text
stringlengths 0
5.91k
|
|---|
170 | 5 Aligning Culture with PSMS Elements
advantage of treating process safety as part of the overall
portfolio of business topics each group addresses. Both options
are acceptable; the key is that management review happens.
5.2 RISK M AN AGEM EN T-RELATED ELEMEN T GROUPIN G
All voluntary and regulatory approaches to m anaging process
safety have some form of risk analysis and risk m anagement as a
central theme. At their core, these management system s seek to
evaluate risk in some way, and to reduce any unacceptable risks
“As Low as Reasonably Practical (ALARP).” The ALARP principle is
explicit in several national regulations and to CCPS Risk Based
Process Safety, and implicit in other regulations.
From the perspective of process safety culture, this grouping
of elem ents drives how companies understand and act on
hazards and risks.
Hazard Identification and Risk Analysis (Element 7) The process of identifying hazards and analyzing risk is
typically performed on every operating unit within a facility m any
times over its lifetim e. The m ethods used m ay be tailored to the
specific situation (Ref 5.4), but generally involve the following
steps:
1. Identify the hazards of the process (e.g. toxicity,
flammability, reactivity, etc.).
2. Estim ate the potential consequences that could occur
under process volumes and conditions.
3. Identify the process deviations that could lead to these
possible consequences.
4. Estim ate the probability that these deviations could
occur.
5. Identify the safeguards that prevent the consequences,
and their probability of failure.
6. Determ ine the process risk. |
320 INVESTIGATING PROCESS SAFETY INCIDENTS
(e.g., recommendation no longer needed due to a change in process
chemistry, alternative recommendation developed, etc.), should be in place.
Chapter 4 addresses the overall management system needs. Specific
suggestions for implementation and follow-up activities are included here in
this chapter. Key considerations for effective recommendation
implementation and follow-up include:
• Assignment of a responsible individual
• Action(s) to implement recommendations
• Challenges to resolving recommendations
• Changes to the management system
• Providing an audit trail
• Tracking action items
• Sharing lessons learned
• Follow-up audit
14.3.1 Assigning a Responsible Individual
An individual, rather than a department or division of the company, should
be named as being responsible for ea ch recommendation. The responsible
individual should determine the most appropriate action(s) to address the
recommendation. This individual shou ld be responsible for the entire
process of implementation, including monitoring the status, resolving any problems, verifying, validating, and documenting that the intended preventive action has been completed and is effective.
Formal hand-off should be planned and documented for shifting
responsibilities to another person in the event of job assignment changes,
retirements, etc.
14.3.2 Due Dates and Priorities to Implement Recommendations
Each recommendation sh ould have a suggested ta rget completion date
reflecting both the urgency and the pr acticality of implementation. Complex
recommendations requiring several steps or an extended time to complete
should be assigned inte rmediate milestones to monitor progress of the
actions. It may also be appropriate to consider additional temporary safety
measures until the main actions have been completed. Alterations to
recommendations and extensions to due dates should be reviewed, in light
of the overall recommendation goal an d subjected to an independent (i.e.
not the Responsible Individual) approval process. |
Appendix A - Human error concepts 375
A.2.2 Non-compliance
The Energy Institute’s ‘Hearts and Mind’ have issued extensive guidance on
“Making Compliance Easier” [120]. The guide pr ovides an up to date view of ‘non-
compliance’ (p7), citing four forms of non-compliance. Their definitions are
reproduced in Table A-1. They note that reckless violations are considered to be
very rare. Their definitions focus on how the organization, the design of
procedures, team norms and knowledg e of risks influence behavior.
Table A-1 ‘Hearts and Minds’ definitions for non-compliance
a) Situational non-compliance
These happen when it is very difficult or impossible to get the job done by
following the procedures strictly. For ex ample, there may not be enough people,
or the right equipment may not be availabl e to follow the procedures as written.
b) Optimizing non-compliance
These happen when people think they can get the job done faster or more
conveniently by not following all the rule s. There are two subtypes of optimizing
non-compliance:
Optimizing for organizational benefit: These happen when people take
shortcuts because they believe that it will help the organization achieve its goals,
e.g., achieve a performance target or meet a deadline. Non-compliance for
organizational benefits may show ways to improve productivity and safety if
brought out into the open, communicated, discussed and approved.
Optimizing for personal benefit: These happen when people take shortcuts
to reach a personal goal (e.g., leaving work on time, or meeting a target),
avoiding using complicated procedures, or because they have found a quicker,
easier or better way of doing the job.
c) Routine non-compliance
A non-compliance of any type can become routine.
These happen when people no longer appreciate the risk of the situation, or
when the rule no longer reflects reality, and not following the rule becomes the
accepted behavior. The rule may be seen as no longer relevant or important.
These non-compliances become routine, ei ther by a whole group or just by one
individual. This indicates that there is an issue around a particular rule, or a
particular individual, or the effort required to follow the rule is perceived to be
greater than the benefits.
d) Reckless violations – a very rare occurrence
In a very small number of cases pe ople commit non-compliance without
thinking, or even caring, about the cons equences to themselves or others,
despite being aware of the potential cons equences. Such ‘violations’ are outside
the scope of this tool. Reproduced from th e Energy Institute [120] |
7 CASE STUDIES/LESSONS LEARNED
This book contains a series of embedded example incidents that
illustrate some of the key issues as sociated with managing abnormal
situations. Using example incidents and case studies in discussions and
formal training sessions can be highly beneficial in helping staff to
understand the underlying causes and learnings arising from these
types of events. Questions to ask staff include:
How would you respond?
How would you ensure that people are out of harm’s way?
How would you decide when to shut down operations?
What do you think we could do differently to avoid a situation like
this from occurring here?
Case studies are available from numerous sources, including
newsletters, incident reports, and various databases as follows:
The Process Safety Beacon , produced by CCPS (CCPS website)
Safety Digest, US Chemical Safety and Hazard Investigation Board
(CSB 2021 news website)
Loss Prevention Bulletin , produced by the IChemE in the UK (IChemE
UK)
Safety Lore , produced by the IChemE Safety Centre in the UK
(IChemE UK)
Learning Sheet , produced by the European Process Safety Centre
(EPSC)
The ICI Safety Newsletters , mainly issued by Trevor Kletz (Kletz T)
Health and Safety Executive UK (HSE Case Studies) (HSE UK)
Chemical Safety Board - reports and videos on major incidents (CSB
website)
European Commission Major Accident Reporting System—a
searchable database of incidents in the EU (eMARS database) |
OVERVIEW OF RISK BASED PROCESS SAFETY 43
RBPS Element 7: Hazard Identif ication and Risk Analysis
Hazard Identification and Risk Analysis (HIRA) are complementary activities
initially identifying process sa fety hazards and their poten tial consequences and later
estimating the scenario risks. HIRA includes recommendations to reduce or
eliminate hazards, reduce potential consequences, or reduce frequency of
occurrence. Analysis may be qualitative or quantitative depending on the level of
risk. HIRA is a core process safety activity.
HIRA analyses vary from simple to comp lex. In addition to basic topics such
as identifying responses to upsets, potentia l leak scenarios, important barriers and
integrity, it must also take into account extreme and remote environments, reservoir
uncertainties, and compounds that affect production (e.g., waxes or radioactive
materials). It also must consider the po tential exposures to people (public can be
nearby onshore, or personnel accommodations may be located next to the facility
offshore or at remote onshore facilities), and to the environment and the asset.
Many different tools are used for HIRA analyses. These range from simple
checklists, through What-If an d HAZOP, to more complex LOPA, QRA, fire hazard
analysis and explosion studies. Inherent safety methods and functional safety
assessments fall within HIRA.
Example Incident: Piper Alpha
The Piper Alpha incident in 1988 resulted from deficiencies in the RBPS risk
management pillar, but also had problems related to safe work practices
(defective work permit system) and emergency management (no safe place for
refuge and backup control if the control room was disabled). The facility was
modified to meet updated environmental regulations. Initially it just handled
liquids and gas was flared, but to avoid this a gas compression and export module
was added. The layout was not ideal and resulted in major process facilities being
too close to the control room. When th e event occurred, the control room was
quickly disabled and the explosion event escalated to multiple pool fires and later
a major jet fire. Personnel congregated in the accommodation module and
perished there due to smoke inhalation. At the time there was no requirement for
detailed risk assessment to track an initial event and how this might escalate to
involve other modules and release more hydrocarbons. The Cullen Inquiry
recommended that a QRA be carried out ad dressing such risks and ensure safety
systems could prevent the escalation.
RBPS Application
The hazards of high-pressure hydrocarbons were reasonably well known at the
time, but not the risk of escalation. Hazard Identification and Risk Analysis sets
out the means to take a hazard identification and extend this with a risk
assessment addressing possible escala tions. Escalation is more important
upstream, especially offshore, where spacing is limited with the small footprint
available. The Piper 25 conference (O il & Gas UK, 2013) and Broadribb (2014)
outline the major learnings and modifications since 1988. |
22 PROCESS SAFETY IN UPSTREAM OIL & GAS
field to recover the oil. Transport of heavy oils for further processing may require
diluent (e.g., kerosene) to make the oil flow more freely in the pipeline and a second
pipe to return recovered diluent back to the well site to repeat the cycle.
Process Safety Issues
Key process safety issues associated w ith onshore exploration include blowouts
(including shallow gas blowouts), hydrocarbon intrusion into freshwater aquifers,
and hydrocarbon loss of containment fro m surface facilities. Further details are
provided in Chapter 4 for drilling and Chapter 5 for onshore production. Blowouts
are prevented by active well management an d early detection of well kick events
that signal a potential influx of hydrocarbons into the wellbore. Primary process
safety controls are (1) the mud column and (2) the blowout preventer (BOP) and the
well pressure containment system. The mu d column uses high-density mud to
hydrostatically prevent formation fluids from entering the wellbore. The BOP
consists of several valves/rams designed to close off the wellbore to control potential
blowouts. The BOP is placed on the surface for onshore dr illing, and at the surface
or on the seabed for offshore. Drilling is us ually carried out by specialist contractors
and there is a need for good communication of process safety between the
owner/operator and the drilling contractor , as is noted in IADC guidance.
2.2.2 Offshore
Drilling offshore is similar to onshore, but the drilling rig is different. In shallow
water up to 300-400 ft (90-120 m) drilling is carried out from a fixed platform or
using a jack-up rig. A fixed platform cannot be moved once installed, whereas a
jack-up is mobile. The jack-up legs have large spud-cans (sometimes with mats) on
the bottom of each leg. Spud-cans use the weight of the rig and ballast water to
penetrate the mud on the sea floor to provide a stable platform for the rig while
conducting well operations. Jack-ups can be self-propelled, but more often they
require tugs to move them between locations.
In deeper water, well operations are acc omplished by floating drill ships which
are ship-shaped, semi-submersibles (i.e., derrick and decks supported by columns
onto pontoons providing most of the buoyancy), or platform-based rigs deployed on
production facilities. Drill ships are generally self-propelled whereas most semi-
submersibles are towed between drilling locations. Position is held during well
operations either by an array of anchors or more commonly by dynamic positioning
(multiple GPS-controlled thrusters). BS EE normally considers deepwater well
operations to be in 1000+ ft (305+ m) or greater of water depth.
Wells drilled in deep water generally have much greater total lengths than
normally seen onshore or in shallow wate r because of the dist ance to the seabed.
The cost of deepwater well operations are higher than onshore or shallow water
operations, mainly due to cost of the rig, logistical support and time. |
466 PROCESS SAFETY FOR ENGINEERS: AN INTRODUCTION
9. The main pieces of equipment on the site will be LNG tanks, unloading arms, heat
exchangers, pumps, and piping. Name three failu res that might occur with this equipment.
10. A consequence analysis is to be performed. List 3 potential scenarios including source,
transport, consequence effects, and potential outcomes.
11. Draw a swiss cheese diagram for one of the scen arios identified in the Preliminary Hazards
Analysis.
12. Suggest three aspects of human factors that should be considered in the project team and
their design of this facility.
13. List 5 things you expect to be on the op erational readiness plan for this project.
14. As the project is 50% through the detailed en gineering, a proposal is made to add an
additional LNG tank. How should this be handled?
15. List three operating practices and three safe work practices that would be appropriate for
this facility when it is operational.
16. List 3 emergencies that should be addresse d in the Emergency Response Plan for this
facility.
17. List 2 means to engage the workforce in the pr oject. List 2 stakeholder groups that should
be involved in the project.
18. List 3 leading and 3 lagging process safety metr ics that might be appropriate for this facility
when it is operational.
19. What action might you take to foster a g ood process safety culture on the project?
Exercise 2: Polymerization Reactor
You have been assigned to a HIRA study team to evaluate hazards associated with a continuous
solution polymerization reactor at your manuf acturing facility which is located near the
Houston, Texas ship channel. This reactor is located within a 2000 m3 (71000 ft3) enclosed
process structure and roughly 250 m (820 ft ) from a 50-person housing complex.
Styrene monomer and ethylbenzene solvent are added to the 2000-gallon reactor by flow
control from their respective storage tanks via pump (not shown). The monomer-solvent
mixture is heated in a shell and tube heat exchanger with 10 barg steam to the normal reactor
operating temperature of 90 C. The exothermic reaction is maintained at 90 °C within the
reactor by temperature control of the vessel ja cket with cooling water. The reactor is well
mixed with a 15-horsepower agitator. The re actor is also maintained under an inert
atmosphere at 0.25 bar gauge using nitrogen by a series of back pressure regulators.
A catalyst solution is added to the reactor from drums by a small metering pump. Drum
weight is monitored by the Catalyst Scale and a low weight alarm. Once a drum is empty, the
operator manually stops the reactor feeds, replac es the empty catalyst drum with a full one,
and restarts the system. |
A.4 Report References | 223
NPO Association for the Study of Failure (ASF) of Japan Incident
Database (Continued)
(For incident reports J1–J163: see www.shippai.org/fkd/en/lisen/cat102.html)
Code Investigation
J81 Explosion of Hydrogen Peroxide Due to the Change of a Feeding
Line to a Vessel at a Surfactant Manufacturing Plant (1989)
J82 Explosion and Fire at an Outdoor Tank to Start Storage Before
Completion of Attached Facilities (1989)
J83 Explosion of a Dryer Due to Unexpected Reaction of Residual Alkali
(1989)
J84 Leakage of Fuel Oil Caused By Damage to a Flexible Hose at a Fuel
Oil Tank Piping (1989)
J85 Explosion Caused By an Overflow of Aqueous Hydrogen Peroxide
at a Peracetic Acid Manufacturing Plant (1988)
J86 Explosion and Fire D Due to a Change from Sodium Salt to
Potassium Salt at a Di-Cumyl Hydroperoxide Manufacturing Plant
(1988)
J87 Rupture of a Chlorosulfonic Acid Tank Due to Pressurizing (1988)
J88 Partial Leakage of Hydrochloric Acid Gas from an Absorber Due to
an Earthquake (1987)
J89 Fire of Ethylene Oxide Adducts at a Manufacturing Plant Not in
Operation (1987)
J90 Rupture of a Solvent Recovery Drum Caused By an Abnormal
Reaction Due to a Temperature Rise at a Sugar Ester Manufacturing
Plant (1987)
J91 Fire Caused By a Thunderbolt That Struck Piping at a Vinyl Chloride
Monomer Manufacturing Plant (1987)
J92 Explosion in Dead Space of a Reactor at a Naphthalene Oxidation
Plant (1987)
J93 Explosion of an O-Nitrochlorobenzene Melting Drum Caused Due
to a Temperature Rise Caused By Reflux Piping Blockage (1986)
J94 White Fumes Generated from a Toluene Diisocyanate (TDI) Solution
Tank Due to Moisture Contamination at an Epichlorohydrin
Manufacturer (1986)
J95 Dust Explosion of Purified Anthracene Powder in a Weighing
Hopper (1986)
J96 Fire Caused By Electrostatic Charge in the Filtration Process of a
Medicine Intermediate (1985) |
Pumps and Compressors
175
be FOB but if the pump gets suction from the top of a
container, the reducer should be FOT. However the fun-damental concept is what was mentioned above. When a pump gets suction from a bottom of a container, there is a chance of getting suspended solids and then we need an FOB reducer and similar logic for using an FOT.
When the suction flange of the centrifugal pump is
smaller than the suction pipe size by more than one size, it is recommended to use multiple reducers in series instead of using one reducer to decrease the size to match with the suction flange of the centrifugal pump. The reason is that a reducer of reduced pipe size by more than one size may generate some disturbance in the liquid and this disturbed liquid, when it gets to the centrifugal pump, cannot be efficiently pumped. In such cases, two reducers in series but not back to back should be used.
Figure 10.4 shows a P&ID representation of a centrifu-
gal pump with the associated reducer and enlarger.
There could be a strainer in the suction of a centrifugal
pump. Installing a strainer in the suction of a centrifugal pump is very common if the installed pump is not sup-posed to receive large chunks of suspended solids in the liquid.
A strainer on the suction side of a centrifugal pump
could be placed for a short term period or a long term period. If the strainer is used for a short term period it can be named as a TSS or “temporary suction strainer. ” It means this strainer should be in place only tempo-rarily during commissioning and then the start‐up. Commissioning, which is the first start‐up of a unit or plant after the construction, is different from other start‐ups during the lifetime of a unit or plant. Because during the construction phase of a plant all vessels are open and pipes are open there could be the chance of a large chunk of solids in the system. These solids could be anything from used welding rods, instrument packages, or even socks. Therefore during the commissioning a pump may see a large solid that could be detrimental for the pump internals, including impellors. So it is a very good idea to place a strainer on the suction side of a centrifugal pump temporarily during commissioning.
However, there are some cases that for whatever
reason there is a still chance of having large solids in the pumping liquid. In such cases the strainer could be placed permanently and during normal operation of the pump.
The size of a strainer opening is decided based on the
smallest clearance in the pump. It is obvious that some centrifugal pumps that are designed to handle large solids like slurry pumps, or some submersible pumps, don’t need a strainer on their suction side.
10.6.1.2 P&ID Dev elopment on the Discharge Side
On the discharge side of a pump (the pump’s down-stream), there could be an enlarger and most likely a pressure gauge, a check valve, and also an isolation valve. After the isolation valve there could be a control loop to control the capacity of the pump. A Tee may exist for minimum flow spillback. The spillback is discussed in Section 10.6.2.
The check valve is a very critical component of a
cen
trifugal pump and it should always be installed on the
discharge side of a centrifugal pump.
The reason for requirement of a check valve is to
pre
vent backward rotation of the impellor in the cen-
trifugal pump when there is a sudden trip in the pump. When there is a sudden shutdown in the pump, the pump won’t rotate and it will stop; however, the pumped fluid on the discharge side of the pump then will no longer be pushed and it may travel back from the dis -
charge side of the pump and into the pump. When the discharge side of a centrifugal pump is a large pipe and/or it is a long pipe the severity of the backward rotation of the impellor is higher. In such cases it may be decided to insert a non‐slam check valve. Backward rotation of the impellor in the pump is bad for at least for two main reasons: it makes the mechanical seal fail and also back rotation is bad for electric motor. A check valve should always be placed on the discharge side of a centrifugal pump and as close as possible to the discharge flange. The criticality of the distance between the discharge flange of a centrifugal pump to the check valve depends on the bore size of the discharge pipe; the larger bore size the more critical it is to keep the check valve closer to the discharge flange.
The last item in the centrifugal pump arrangement is
isolation valves and blinds. Centrifugal pumps, isolation valves, and blinds should be used on both sides.
Up to now, a typical P&ID representation of a centrifu-
gal pump could be like that shown in Figure 10.5.
One (or more) size smaller One (or more) size bigger
Discharg e one size smaller (or same?)
Discharg e one (or more) size smallerPG
TSSPG
Figure 10.4 Cen trifugal pump with associated reducer/enlarger. |
Piping and Instrumentation Diagram Development
40
PID-300-1003Wash water
To wash water pre-h- 8/uni2033 - 3015BC DE FG
Refer to the P&ID for the type 1
sampling system
4×3 4×3 FOFCIS OS
FC
130
FV
130LV
131LC
131
WAT - AA - 4/uni2033 - 3014S1WAT - AA - 4/uni2033 - 3017
WAT - AA - 6/uni2033 - 3018WAT - AA - 6/uni2033 - 30163/4/uni2033 3/4/uni2033
2/uni20336×4 6×4
3/uni20331/uni2033 1/uni2033
1/uni2033 1/uni2033
1/uni2033 1/uni2033
Figure 4.26 A P&ID sheet with a ref erence to a sampling system sheet.
SAMP LING SYS TEMS
8300-25I -001- AMoham mad Toghrae i
0Sampling ty pe OpenSample Fluidity Complete fluidity
Sample health No issueSample Phas e Liquid -Low volatility
Sample Temp. <60°C
Sampling Source
Min.Min.
Sampling Source
Min.Min.
CMS 3/4"
1/2"3/4"
1/2"CMR
Sampling Source
Min.Min.
CMS 3/4"
1/2"CMR Sampling ty pe OpenSample Fluidity Complete fluidity
Sample health No issueSample Temp. >60°CSample Phas e Liquid-Lo w volatility
Sample Phas e Liquid -Low volatility
Sample Temp. >60°C
Sample Fluidity Highly Viscose
Sample health No issue
Sampling ty pe Open
Utility Steam
Sampling SourceMin.
To Safe location3/4"
1/2"Sampling ty pe Closed LoopSample Fluidity Complete fluidity
Sample health No issueSample Phas e Gas
Sample Temp. <60°CSampling System-Type 1 Sampling Syst em-Type 2
Sampling Syst em-Type 4 Sampling Syst em-Type 3
Figure 4.27 A sampling syst em P&ID. |
DETERM INING ROOT CAUSES 251
10.8.2.2 Analyzing a Causal Factor
The following is an analysis of one of these causal factors: contractor
operator (CO) falls asleep. The basi c technique works with any of the
predefined trees commonly used within the process industry. However, for
the purposes of this example, a proprietary tool (Paradies, 2016) has been
selected, and therefore the structure of the tr ee and the terminology used is
specific to that tree.
To analyze the causal fa ctor, the investigator starts at the top of the tree
and works down the tree through a proc ess of selection and elimination. The
investigator asks and answers qu estions to identify the specific root causes
for the causal factor.
In this case, the causal factor (contrac t operator falls asleep) is identified
as a Human Performance Difficulty (one of the four major problem
categories at the top of the tree, see Figure 10.26), and the other three
categories are discarded. (Different predefined trees use different
terminology and structure, but gene rally cover similar choices.)
Figure 10.26 Top of the Predefined Tree
The investigator then follows the Hum an Performance Di fficulty category
through a series of questions (o r subcategories). These questions help the
investigator identify which of several human performance related branches
(sometimes known as basic causes) to investigate further. (Some predefined
trees use statements rather than questions, but the selection process is
similar.
The human performance related branches are:
|
2.6 Understand and Act on Hazards and Risks |53
Figure 2.3 Exam ple risk matrix
Probability Consequence Rare Occasional Regular Frequent Constant
Catastrophic Unacceptable
Severe Reduce risk Reduce risk as risk
High at next opportunity soon as possible
M edium Risk generally
Low acceptable
If the risk related to a given hazard is not within the generally
acceptable category, the com pany must then apply safeguards to
reduce the risk. Again, efficacies of given safeguards are clearly
defined in order of magnitude categories. Safeguards that reduce
probability by 1 order of magnitude shift the risk one cell to the
left in the m atrix. Safeguards that reduce potential consequences
by 1 order of magnitude downwards. It may take several
safeguards to bring the risk to the acceptable level.
Among other benefits, the risk m atrix approach makes it quite
clear how many safeguards are required. Generating risk matrices
can be hard work, however. It helps in defining risk categories to
relate risk levels for the process to risk levels in daily life, such as
the risk of driving, to help everyone can clearly see how the
process risk com pares to something they are familiar with.
The bottom line of understanding and acting on hazards and
risks, as Adm iral Hyman G. Rickover stated on many occasions, to
“face the facts.” As Adm. Rickover built the US Navy’s nuclear
program, he strongly believed that officers managing the program
m ust be prepared to m ake difficult decisions that favor reactor
safety, despite pressures due to cost, manpower, schedule, or
potential bad press involved. Ultim ately, the facts about process |
Piping and Instrumentation Diagram Development
92
P&IDs. For example, some company guidelines ask
for connecting pipes larger than 4 or 6 in. thr
ough
flexible connections when they are connected to tanks to absorb the settlement of the tank. Small bore pipes are waived from this guideline because they can handle a few tank settlement by creating a sagging in the pipe.
Figure 6.56 shows a flexible connection on the inlet of
a centrifugal compressor, although it is not common these days.
Flexible connections used to have a bad reputation
regarding leakage. But now there are better flexible joints in the market.
6.9 Dealing with Unwanted
Tw
o‐Phase Flow in Pipes
The design and implementation of systems in two‐phase flows are more difficult than single‐flow pipes. There are, however, cases in which a two‐phase flow is inevitable.
When the flow is intended to be a single flow, but then
it turns out to be a two‐phase flow, the piping design is based on a single phase, and the two‐phase flow should be eliminated. There are three types of two‐phase flows: liquid–gas, gas–liquid, and solid–liquid.
6.9.1
Liquid–G
as Two‐Phase Flow
In a liquid–gas two‐phase flow, there is a chance of liquid
droplets in the main stream of gas or vapor. The problem arises when transferring a gas or vapor because a liquid can be generated and that is problematic.
Such unwanted two‐phase flows may happen at differ -
ent times. One is when gas comes off of a liquid surface, like in liquid–gas separators. The other case is when transferring hot vapors, like steam.
The first step in dealing with this problem is to prevent
the creation of a two‐phase flow. For example, when transferring a wet gas, heat trace (dashed line beside the main line) may be used. This solution can be seen in Figure 6.57.
The next method is to remove the generated liquid
phase from the gas phase as soon as possible before the creation of a slug of liquids. One example is using a demister in a gas–liquid separator vessel as shown in Figure 6.58. For gas streams that come off of a liquid sur -
face, there is always the chance of carrying liquid drop-lets over into the gas stream.
The other example is using steam trap in steam distri-
bution piping networks. Steam traps remove water con-densation from the steam (Figure 6.59).
In Figure 6.60, a steam trap is shown as a square with
letter T at the middle. Steam traps should be installed at predetermined distances on steam transfer pipes, and the pipes should be sloped toward the stream traps. Failure to install a condensation removal system in steam pipes may lead to steam hammering, which may break the pipes.
There are other symbols can be used on P&IDs for
steam traps if the intention is to use the exact type of steam trap (Figure 6.61).M
Figure 6.56 Fle xible connection on the inlet of centrifugal
compressor.
Figure 6.57 Heat tr acing to prevent the generation of
condensation.
Figure 6.58 Demist er to prevent carrying over of liquid droplet.
Steam +Condensat eS team trap Steam
Figure 6.59 Str eam trap action. |
APPENDIX B – EXAM PLE PROTOCOL 363
Evidence
No items will be removed from the equipment as evidence at this time. If
the valve is to be removed for further ev aluation, a separate protocol will be
prepared.
Safety Provisions
The site safety plan will be followed, including:
• PPE requirements – FRC, steel toes sh oes, hard hat, safety glasses with
side shields, leather gloves, hearing protection
• Gas detector for flammable atmosphe re; for radiograph equipment and
cameras
• The number of people who can be present on the platforms is limited
by size of the platforms.
• Radiograph safety procedures provided by the contractor will be
followed as approved by the radiation safety officer. All non-qualified
personnel will be beyond the minimum safe distance specified by the
subcontractor. The specific safety provisions are provided below.
Approach
The following steps are followed to check the valve position:
1. Place an alignment mark (Mark #1) on the chain wheel and adjacent
housing to document as-found position.
2. Photograph the valve.
3. Measure the height of the va lve stem and photograph with a
measurement device beside the stem.
4. Radiograph the valve according to the following procedure:
• An appropriate radiation source will be selected for all shots. The
camera containing the radiation sour ce is man-portable and requires
no external power source. All film cassettes and su pport stands are
also man-portable. The radiation safety protocol described in this
protocol will be followed.
• To maintain traceability, an alphanumeric identification system will be used to track which valve is being radiographed and the position of
the source relative to the valve and exposure time. Lead lettering will |
185
present themselves. Chapter 7, and al so Chapter 14 contain guidance for
the use of IS in operations and in the programmatic aspects of process
safety programs.
8.7 OPERATIONS & MAINTENANCE
The longest stage in the life cycle of a process is the operations and
maintenance stage. This phase will lik ely last for decades and will span
many changes in personnel, oper ating and maintenance philosophy,
business/financial changes, and pe rhaps ownership. There are two
issues that are important with respect to inherent safety that should be
addressed during this phase:
Preserving the inherent safety features and practices provided
during the process development phase of process life.
Seeking opportunities for continued improvement in inherent
safety.
8.7.1 Preservation of Inherent Safety
A primary objective of any process sa fety program is to maintain or
reduce the level of process safety risk in the process. The design basis of
the process, especially those inherently safer features that are built into
the design and installation of the pr ocess, should be clearly documented
as inherent in nature and how/why they are inherent. This is particularly
important for IS measures because they are not currently required by Recognized and Generally Accept ed Good Engineering Practices
(RAGAGEPs) or other standards that can compared to the design later. Design measures that are inherent sa fety strategies are “inherent” to the
final design decisions made and deep ly embedded in the basic features
of the process. There is generally no standard, report, calculation, of
other reference that shows the IS bene fits that were incorporated into
the final design, unless someone thin ks to create such documentation.
Therefore, without such specific IS documentation it will not be possible
to completely evaluate any anticipate d change, especially those that may
occur years later, to de termine whether it stre ngthens or weakens the
inherent safety features or practices or has a neutral effect. Complete IS documentation is also important because the persons (or even the
organizations) who provided inherent features or practices will likely not
be available to explain the inhere nt safety aspects of the design. |
332 PROCESS SAFETY FOR ENGINEERS: AN INTRODUCTION
resources could be saved. It is an order of magnitude approach which makes it simpler and
quicker to use than a QRA but using a consistent method and values is imperative to having
comparable results.
Quantitative risk assessment is typically rese rved for the highest risks. It can be labor
intensive, and its quality is dependent on th e appropriateness of the data and parameters
used. For those risk-based decisions regarding spending significant funds on project design or
risk reduction measures, QRA can support prudent allocation of resources.
A challenging question for many process safety professionals, and their company
colleagues, is when enough layers of protection have been implemented to yield a residual risk
that is tolerable. Unless levels are defined in regulation, this can be a sensitive question.
Resources are available to assist in creating cr iteria and precedents. Having criteria greatly
supports the making of risk-based decisions and is required for LOPA.
All risk analysis is dependent on understandi ng both consequence and frequency. Several
tools are available to support identification of frequency values including use of historical
records, fault tree analysis, and event tree analysis.
Other Incidents
This chapter began with a description of the Phillips 66 Pasadena explosion. Other incidents
relevant to consequence analysis include all of the incidents listed in Chapters 4, 5, and 6.
Exercises
List 3 RBPS elements evident in the Phillip s 66 Pasadena explosion summarized at the
beginning of this chapter. Describe their shortcomings as related to this accident.
Considering the Phillips 66 Pasadena explosion, what actions could have been taken to
reduce the risk of this incident?
Use a simple fault tree to estimate the overa ll frequency of activation of a relief device
caused by either failure of a pressure regula tor or overheating of tank contents due to a
failed temperature control. Use a failure freq uency of 0.1/yr for the pressure regulator
and 0.2/yr for the temperature control. As sume a high temperature interlock exists
which shuts off heating of the temperature control with a Probability of Failure on
Demand of 0.1. Show your results.
Use a simple event tree to estimate the fr equency for an overfill event where Human
Error of 0.1/year results in connection of a tank truck to an already full storage tank
equipped with a high-level interlock to th e feed pump of PFD =0.01 and flammable gas
detection interlock to the feed valve of PFD=0.1. Show your results.
Estimate the frequency for a “full bore” leak (Full Leaks) of a 150 mm (5.9 in) diameter
pipe with length of 1000 m (3280 ft) from Figure 14.13. Show your results.
For the following scenario statement, estima te the risk reduction factor or number of
protective layers needed to meet a tolerabl e frequency. Use the risk matrix in Figure
14.14 to determine the tolerable frequency. Ta nk T-103, is involved in an overfill event
caused by a level control failure with a subs equence airborne release of 1500 kg (331 lb)
acrylonitrile. This incident may result in toxi c Infiltration to a nearby occupied building
which could result in up to 1 fatality. Show your results. |
EMERGENCY MANAGEMENT 417
WFC itself was destroyed (see Figure 20.4). An FGAN railcar was overturned. Fortunately,
the two anhydrous ammonia tanks on-site were not damaged. A large amount of off-site
property was damaged. The follo wing were severely damaged.
an apartment complex, 122 m (450 ft) from WFC (see Figure 20.5)
an intermediate school, 168 m (552 ft) from WFC
a nursing home, 183 m (600 ft) from WFC
a high school, 385 m (1,263 ft) from WFC
The cause of the initial fire itself is unkno wn. The ATF concluded that the cause was arson
(Ellis 2016). The CSB developed three theories as to why the AN exploded that did not involve
arson (CSB 2013).
The first scenario is that during the early part of fire, soot and other organics contaminated
the FGAN and served to keep heat in. This coul d have caused formation of hot liquid FGAN at
the top of the pile. The liquid layer could have produced oxidizing gases, which would have
created a cloud of oxidizers, NO 2, O2 and HNO 3. All are the decomposition products of AN. This
gas cloud may then have detonated.
The second scenario is that the detonation wa s caused by heat from the exterior walls of
the bin. Photos show that just prior to the detonation, the exterior walls of the bin were
penetrated, which allowed more air in and caused the fire to become even hotter. There could
have been some melting of the FGAN along the exterior wall.
The third scenario focuses on an elevator pit; a bucket elevator was used to unload FGAN
and other materials. There could have been FGAN remnants in the pit. FGAN could have spilled
into the pit if the wall of the AN bin collapsed. The remnant of FGAN could have been
contaminated by burning rubber and the fallin g FGAN, plus the confinement by concrete
elevator walls might have caused the detonation . This is considered the least likely scenario.
Lessons
The RBPS management systems are interlinked, and the West Fertilizer explosion shows how
important this linkage is.
Process Safety Culture. Prior to 2009, WFC had insuranc e through Triangle Insurance
Company. In 2009 Triangle stopped insuring WF C because of losses and a lack of compliance
with Triangle’s recommendations from thei r loss control surveys. Several of the
recommendations involved electrical problems, such as corroded wires and grounds. In one
of its evaluations, a Triangle consultant note d that WFC had no safety program and “had no
positive safety culture”. (CSB 2013).
Compliance with Standards. AN is covered by OSHA’s “B lasting and Explosive Agents”
standard (OSHA 1998); however, this is not widely known throughout the fertilizer industry. AN
is also covered by NFPA 495, “Code for the Ma nufacture, Transportation, Storage, and Use of
Explosives and Blasting Agents” (NFPA 495) and NFPA 400, “Hazardous Material Code” (NFPA
400). Prior to 2002, AN was covered NFPA 490 “C ode for the Storage of Ammonium Nitrate”
(NFPA 490). |
APPLICATION OF PROCESS SAFETY TO WELLS 55
Incident: Deepwater Horizon, April 2010
The Macondo well is located in the Gu lf of Mexico. During actions for a
temporary abandonment of the well, several failures occurred. The final
cementing used a novel formulation, and this failed to seal the well. The
heavy mud barrier was partially circulated out and an underbalanced
situation resulted. Kick signals were misinterpreted, and a loss of well
control followed. Flammable oil and gas were initially diverted into the mud
room, but this soon ended up on the drill floor, where it ignited causing 11
fatalities, total loss of the drill rig, and the largest oil spill in US history.
Process Safety Issues : The Deepwater Horizon had an excellent
occupational safety record. There was a program to address process safety
(see later discussion on tools such as drill well on paper (Section 4.3.2)), but
occupational safety was emphasized more than process safety. There were
multiple technical defects identified relating to the cement job, the kick
detection, and the apparent failure of the BOP. In fact, the BOP includes
multiple safety systems, and some worked properly. The variable bore rams
closed, sealed, and held considerable pressure. The shear ram failed to close
because three drill string pieces were in the ram and the cutting face could
not cut all the pipe in its bore (DNV, 2011). The National Commission made
the following key conclusion.
“The immediate causes of the Macondo well blowout can be
traced to a series of identifiable mistakes made by [the companies
involved] that reveal such systematic failures in risk management
that they place in doubt the safety culture of the entire industry.”
Source: Deepwater Horizon National Commission, 2011
RBPS Application
Process Safety Culture : Requiring a focus on process safety, not only
occupational safety. Insufficient attention was given to a potential loss of
well control with many other conflicting objectives present.
Asset Integrity and Reliability : Ensuring that safety critical equipment such
as a BOP functions reliably is fundamental to process safety. While the BOP
did function, it was presented with a condition which exceeded its design
capability and it failed to seal the well.
Contractor Management : A main characteristic of well construction
operations is the close relationship and dependency of the owner/operator,
the drilling contractor, and the other sp ecialty contractors. Establishing a
clear understanding of who does what in routine, non-routine, and
emergency situations is imperative.
As both are key factors in a safe design, these must be established first in the
well design process. There are many sour ces used to determine these two factors |
Selecting an Appropriate PHA Revalidation Approach 97
Example 6 – Batch Processes:
Since the prior PHA, a batch chemical plant had implemented a few, simple
changes to the batch process equipment. It had also reformulated one product
line in its multi-purpose facility, which required rewriting its batch procedure.
The revalidation approach could be to Redo the PHA for the re-written recipe,
but to Update those recipes and equipment with only minor changes.
Example 7 – New Safeguard Requirements:
Since the prior PHA, a company now requir es the safeguards credited in the PHA
to be included in the facility mechanical in tegrity program. If safeguards (e.g., the
relief valve on a supplier’s railcar) are no t in the ITPM program, the PHA team is
instructed to (1) find or recommend alternative safeguards, (2) make recommen-
dations to include the previously identi fied safeguards in the maintenance
p r o g r a m , o r ( 3 ) v e r i f y t h a t s a f e g u a r d s t h a t c a n n o t b e i n c l u d e d i n t h e s i t e
maintenance program (e.g., due to owners hip of the asset by a third party) are
being properly maintained and that the ITPM records are auditable. Since every
documented safeguard and risk ranking in the PHA HAZOP worksheets will be
re-assessed, the revalidation could perform an Update for changes/incidents and
a Redo of safeguards and risk rankings, without altering causes and
consequences.
Example 8 – Misapplication of Supplemen tal Risk Assessments (e.g., LOPA):
In preparation for the PHA revalidation, it was discovered that the associated
supplemental risk assessment (e.g., LOPA ) was far too liberal in the application
of some criteria (e.g., conditional modi fiers were overused). This misapplication
alone would not be reason to Redo the core PHA (i.e., the HAZOP worksheets);
however, the team should consider whether (1) only a few affected LOPAs can
be Updated , or (2) the issue is systemic and all the LOPAs should be Redone .
5.2 SELECTING THE REVALIDATION OPTIONS
To determine the approach prior to be ginning preparation for the revalidation
meeting, it is necessary to carefully review and integrate the information
gathered and questions answered in previous chapters of this book. This activity
should begin several months before th e revalidation due date, which may be
immovable due to regulatory requirements. The time required for revalidation
preparation, analysis sessions, and document preparation depends upon the
revalidation approach, and the revalid ation approach depends upon an
assessment of the unit’s prior PHA and op erational history. Working back from
the delivery deadline and allowing for contingencies, it is prudent to start the |
135
7.3 Center for Chemical Process Safety (CCPS), Guidelines for
Engineering Design for Process Safety, Second Edition New York: American
Institute of Chemical Engineers, 2012.
7.4 Horn, R.E., Developing Procedures, Policies & Documentation,
Info-Map, Waltham, 1992, page 3-A-2
7.5 Kletz, T.A. Process Plants: A Handbook for Inherently Safer
Design. Philadelphia, PA: Taylor & Francis, 1998).
7.6 Kletz, T.A. and Amyotte, P., Process Plants: A Handbook for
Inherently Safer Design, Second Edition. CRC Press, 2010. |
304
method would be used, with the IS guidewords (Table 12.1) used as
deviations for each node or subsystem. In general, combining the use of
HAZOP, or What-If? Methodology with a checklist provides for creative
brainstorming as well as a detailed me ans to ensure that most issues
have been covered. It should be poin ted out that no checklist is perfect
and there may be opportunities not id entified in the checklist that can
only be discovered through a more subjective analysis.
12.1.3 IS Review Methods
The following review methods can be us ed to ensure that inherent safety
is considered and documented for hazardous processes:
An independent IS analysis done in addition to a PHA, either
in tandem or separately . This analysis should review the
process for ways to eliminate or reduce hazards present in the
covered process and may be achi eved using an IST checklist
(Appendix A) or guideword analysis (Table 12.1).
An IS analysis that is incorporated into the existing PHA review process . In most cases, an initial stand-alone IS analysis
should be conducted for the entire process to ensure it receives
adequate attention. Again, this may be achieved using a checklist
(Attachment A) or guideword (Table 12.1) approach. This type of
analysis would review the processes for ways to eliminate or reduce hazards, as well as to reduce risks using the other risk management strategies (passive , active, and procedural).
12.1.4 Research & Development Application
There are significant benefits to a pplying IS concepts and methodologies
within Research and Development (R&D)/laboratory and pilot plant
operations. IS requirements should be mandated both when a process
hazards review is done, and when a project progresses toward full
development:
A process safety review should be required for each new or
significantly modified R&D “proce ss” (a semi-works, laboratory
experiment, etc.). This review should include hazards identification and evaluation, facility siting, consequence
analysis, human factors, and, importantly, inherently safer design. The emphasis should be on the safety of the pilot plant |
Pressure Relief Devices
219
12.3.1 Active Versus Passive Solutions
T
he above discussion shows that using only one solution
(either passive or active) to deal with high‐pressure sce-
narios may not be a good idea; rather, a combination of these two methods should be used in order to have a safe and economical plant. Therefore, the allocation of issues to these two solutions is the primary task. A summary of the task allocation is shown in Table 12.2.
The concepts stated in Table 12.2 are expanded upon
in the sections below.
12.3.2
Wher
e Could Passive Solutions
Be Used?
There are not very many cases where the passive solu-
tion, or fabricating the equipment based on the highest attainable pressure as the design pressure, is acceptable. This solution can basically be used where it is legal and where a worst‐case maximum attainable pressure exists, and can be calculated. Generally speaking, the regulatory body’s preference is to use active solutions, and they gen-erally prefer to “see” a PRD on every single container. It is not easy to estimate the maximum attainable pressure for all overpressure scenarios.
For example, in a fire scenario, it is difficult to estimate
the maximum attainable pressure because of the unpre-dictable nature of fire. The other example is a blocked outlet of positive displacement (PD) pumps. When a valve on the outlet of a PD pump is accidentally closed, the pressure on the discharge side of the pump will increase. Here there is no “maximum attainable pres -
sure”; pressure will increase until the pump casing rup-tures. In this case, the use of a passive solution is also impossible.
The other example is protecting a centrifugal pump
against high pressure caused by a mistakenly closed valve on its discharge side. In this case, the maximum attainable pressure can easily be estimated and it is what we call the “pump shut‐off pressure” or “pump dead‐head pressure. ” This case could be a good case for using a passive solution to protect the piping and equipment downstream of the centrifugal pump. To do that, the downstream piping and equipment would need to be fabricated based on a design pressure equal to the cen-trifugal pump dead‐head pressure.
Some examples of cases where passive solutions have
been used are internal fire, (external) jet fire, hydraulic hammer, and blocked outlet of centrifugal pumps.
12.3.3
Wher
e Should Active Solutions
Be Used?
The short answer to this question is: in all enclosures.
The long answer would add to that: in high‐pressure
sy
stems as much as possible, unless it is not technically
feasible.
For example, active solutions (i.e. installing PRDs) can
be used to protect against a fire pool (a fire that engulfs the equipment), thermal expansion in pipelines, control valve failed or jammed open, and blocked outlets. Typical scenarios that are unlikely to accommodate active solu-tions include some gas vessels and underwater scenarios, where there is no room for release.
12.4 Safety Relief System
As soon as a pressure relief device is installed on the first pressure safety valve on a process item, a “process relief device system” should be developed (Figure 12.3).
PRDs release fluids from the inside of enclosures to the
outside. Then such released fluid should be collected through a collecting system and directed to a specific type of disposal system that is named an “emergency
di
sposal system. ”
The collecting system is a type of pipe network and will
be discussed in Chapter 16 as part of utility networks.
Emergency disposal systems are briefly discussed at
the end of this chapter.
Table 12.2 Applica tion of passive and active solutions.
Applications
Passive solution:
process design ●If maximum attainable pressure exists and is specifiable
●If it is legally acceptable
Active solution:installing a PRD ●As much as possible
●Unless it is not technically doable (e.g. some gas vessels,
underw
ater
requirements)Main process
A system with
probability of
overpressure
PRD
Collecting
systemDisposal
system
Figure 12.3 Pr essure relief device system. |
75 | 6.1 Focus
6. The airlock doors from the main deck to the office block were rarely
closed.
7. Drillers very much left to their own devices. We had our own isolation
procedures, lock out system, maintenance team.
Rae acknowledged that, although he didn’t recognize it at the time, these
were strong warning signs of breakdowns in management of change,
emergency management, asset integrity, operational discipline, and safe work
practices.
Why wait for an incident investigation to hear the warning signs that
preceded the incident? Ask frontline and leadership personnel at the plant and
site to imagine that they are being interviewed after a major incident. Then
have them complete this sentence: “I knew that this incident would occur
because __________.”
You can also ask personnel, “What is the one thing that concerns you most
about our present operations?” or “If you could improve one process safety
task, what would it be?” These questions can help to identify potential
incidents before they happen. Try this approach during task reviews, toolbox
talks, and safety meetings, or in meetings specifically held for this purpose.
When employees at any level are engaged in these exercises, common
concerns will emerge. They represent the collective gut feel of the plant or
site—they are warning signs of where the PSMS may be weak. Any potential
weaknesses that don’t match up with metrics or audit results should be added
to the list of potential improvement opportunities.
Recent High-Profile Incidents
Consider placing on your list for study any external incidents that involve
a technology, process, or unit operation the company uses. The findings and
recommendations from the investigations of those incidents may be directly
applicable to company standards, policies, and PSMS. In the special case of
licensed technology, the technology licensor may do some of the evaluation
on behalf of all licensees. However, don’t assume the licensor has done the full
evaluation. The company is ultimately responsible for its own operations.
Due Diligence Evaluations and Post-Acquisition Integration
Process safety evaluations conducted during and following acquisition of
a site typically identify gaps that need to be closed relative to the company’s
PSMS, standards, and policies. If the operation of the business, technology, or |
162 GUIDELINES FOR MANAGING ABNORMAL SITUATIONS
Conduct of operations – behaviors, alertness, and diligence of
workforce to recognize and intervene to correct abnormal
situations.
Process safety engineers should consider developing specific
questions on these issues for HIRA team members to use to address
abnormal situations. Other RBPS elements discussed in previous
chapters may also be beneficially audited with a focus on abnormal
situations.
6.5 MANAGEMENT REVIEW AND CONTINUOUS IMPROVEMENT
Management review is the routine evaluation of whether management
systems are performing as intend ed (CCPS 2007a). This ongoing due
diligence review by leadership fills the gap between daily work activities
and periodic auditing. Weaknesses and inefficiencies in a management
system may not be immediately obvious, but the management review
process provides regular checks so th ey can be identified and corrected
before they are revealed by an audit or an incident.
A study by the ASM® Consortium of the root causes of 42 incidents
found that the top ten causes accoun ted for 71% of all the operations
practice failures (Bullemer & Laberg e 2009). Over 10% of these top ten
causes were associated with an in effective continuous improvement
program.
The management review should be led by the facility manager and
involve key subject matter experts, including a senior operator and
maintenance technician. The team shou ld focus on a few RBPS elements
(typically up to three) at each me eting and then evaluate and discuss
records and observations pertaining to management system
weaknesses associated with those RBPS elements. Management should
identify weaknesses, and make re commendations for improvement and
then capture them in an action plan that specifies responsible parties
and completion dates. The next mana gement review should then focus
on action plan progress to driv e continuous improvement, before
reviewing the performance of other RBPS elements.
|
64 | 5 Learning Models
Most process safety leaders today are accustomed to driving change from
the bottom up and should feel familiar with ADKAR®. However, CCPS
recommends that process safety be driven from the top of the company
through, middle managers, to the frontline (CCPS 2019).
5.3.5 IOGP
A working group of the International Association of Oil and Gas Producers
(IOGP) has developed a guide based on how their member companies typically
learn from process safety incidents (IOGP 2016). Although this guide focuses
on learning from internal process safety incidents, the learning processes it
identifies describe the current state of how learning occurs in the international
oil and gas sector learn.
The IOGP working group identified a set of principles that guide how
companies should think about learning from incidents. (They also developed
a map of components, that is, an inventory of how member companies learn.)
The 10 guiding principles include:
• Something must change. If the company plans to learn from incidents, it
must commit to changing what it does based on what’s learned.
• Learning is one way we manage risk. Learning allows companies to control
risks better.
• Sharing is not the same thing as learning. Just because you pass on
information, that does not mean the recipient changes behavior.
• Balance short-term temporary mitigation with long-term sustainable
response. Because long-term solutions take time, short-term interim
actions may be needed while the long-term solutions are developed.
• It’s necessary to focus. Considering the wide range of potential learning, the
company needs to focus on what can really make a difference.
• Learning is a distributed effort. No one person or group can drive all
learning for a company.
• Don’t restrict learning to individual incidents. Look for patterns and themes
across multiple incidents.
• Promote collaboration. Developing solutions is as much a distributed effort
as is learning.
• Leaders make learning work. People do what their leaders value. For
learning to work, leaders must value learning.
• Close the loop. Manage the learning process. Ensure efforts are paying off
as expected and adapt as needed. |
EMERGENCY MANAGEMENT 423
Emergency response exercises can be based on scenarios identified in process hazard
analysis or on past incidents. Exercises can in clude tabletop exercises, tests of communication
systems, and field drills. Drills can be simple involving a small portion of a response team or
can be quite large involving mutual aid supporters and external agencies. Each drill should be
followed by a critique of the response, the communications, and the plan. Findings should be
used to improve the emergency response plan.
Emergency Response Training
Emergency response training should be cond ucted for those with roles defined in the
emergency response plan and others affected by the potential emergency. This may include
employees, contractors, neighbors, and local auth orities. The training should include how they
will be notified and how they should respond in an emergency. Training should be provided
initially and refreshed periodically.
Emergency Response Communications
Communications are critical to an emergency and having effective communications requires
planning. This includes what communications equipment will be used, where this equipment
will be located, who will be communicating, and maintaining a list of current names and contact
information. Although this may be relati vely simple inside a facility, emergency
communications will include contractors, local au thorities, neighbors, and other stakeholders
which can increase the complexity. A strategy should be developed to ensure that facility
responders can quickly and easily communicat e with other responders (community, mutual
aid, etc.). It should also be kept in mind that the emergency itself can challenge
communications. For example, natural disasters can interrupt power supplies and
communication towers leaving cell phones inoperable.
Recovery and Recommissioning
After the emergency has passed, it is time to manage the aftermath and resume operations.
There will likely be damaged equipment to repair , contamination to address, hidden or silent
failures, new hazards associated with old equipm ent, and typical startup challenges. The first
step in this phase is to stabilize and secure equipment to make it safe for investigators and
those working to clean up and repair it. It al so preserves potential evidence, physical and
electronic, that could be helpful in understanding what happened and preventing its
recurrence. The next step is to repair the facility.
The recommissioning plan for a facility following an emergency must be at least as
comprehensive as that for the initial start-up of a new facility. Before starting operations, those
responsible for the startup should be properly trained, it should be verified that the equipment
is ready to receive the chemicals and utilities, and all operational and safety systems should
be functional. Things that may have worked pr operly before the emergency may not work after
it. Do not assume that equipment will perform as expected. Confirm it. Refer to Chapter 17 on
operational readiness.
|
126 Guidelines for Revalidating a Process Hazard Analysis
to resolve that recommendation would be added to the list of safeguards against
the low flow hazard. A complete set of worksheets (some revised, others
unchanged) can then be published in the Updated PHA.
6.3.4 Audit Results
PSM system audit results should have been included in reviewing the operating
experience for the PHA as discussed in Chapter 4. Fundamental deficiencies in
the prior PHA could have led to the decision to Redo the PHA. If a decision has
been made to Update the PHA, audit results could provide insight into items to
focus on, ensuring they were adequately addressed in the prior PHA.
It should be noted that, typically, PSM system audits are conducted using
some sort of sampling scheme to identify documentation that is to be reviewed.
Thus, the fact that a particular PHA report is not mentioned in an audit finding
should not be construed as conclusive proof of the quality of that particular PHA;
the PHA may not have been examined as part of the audit.
6.3.5 Incident Reports
Accident and near-miss reports may id entify loss scenarios that warrant
consideration within the revalidation effort and should be reviewed as part of
the preparation. Where relevant information can be gathered on incidents in
similar facilities or processes, such in formation may also be considered. This
information can be collected from open literature sources or specific databases
such as the CCPS Process Safety Incident Database (PSID).
A major loss, a series of less signif icant losses, or nu merous near-miss
incidents in a process unit or similar unit can be indicators of a weakness in, or
failure of, some PSM system element(s). Incidents involving the subject process
should be scrutinized to see if the particular circumstances prompt a concern
with the quality of the prior PHA. Fo r example, the PHA team may not have
identified the potential for a particular incident. Alternatively, the PHA team may
have identified the potential for the incident, but erroneously judged the
safeguards to be adequate. |
368
|
1 INTRODUCTION
1.1 PURPOSE AND SCOPE OF THE BOOK
This book provides resources for supe rvisors and operators/technicians in
industrial processes, whose correct and timely intervention is often crucial,
either to prevent abnormal situations fr om escalating into a major event, or
to mitigate the consequences if an event occurs. Operations management
and support services personnel (such as those in maintenance, engineering,
and process safety), who read this book will be able to develop relevant
training and support material to prevent or mitigate abnormal situations
from occurring at their facility. This book includes the historical
development of principles and procedures for managing abnormal
situations as well as a summary and review of available resources for
addressing them. It also provides guidance for management and engineers
to develop appropriate training and procedures, and demonstrates how to
institutionalize these into process operations.
Since many of the principles and practices of managing abnormal
situations are transferable across industries, this book provides some
guidance and suggestions on sharin g knowledge and learning from a
variety of industries and disciplines that are leaders in such management.
With that in mind, example incidents (brief examples) and case studies
(detailed studies) are included for front line staff as training aides.
As part of the development of this book, five online training modules
were developed relating to abnormal situations. These training modules
can be used by supervisors, plant en gineers, and trainers to help train
operating teams in diagnosing an ab normal situation. The modules allow
the trainer to step through specific abnormal situations and discuss
diagnosis, actions to be taken, learning, and relevance to their operation
with the team members who are being trained. Details on how to access
this material is provided in Appendix A.
|
Figure 18-7: Challenging skills
|
INVESTIGATION M ANAGEM ENT SYSTEM 77
The initial site visit is the first opportunity to establish the physical
boundaries of the investigation. The team leader should:
• ensure that access to the area is minimized as much as possible, and
• verify that the personnel who en ter the incident area are aware of
evidence preservation considerations.
One of the most critical issues is clearly establishing which groups
have responsibility for which activities and areas. These responsibilities may
change during the investigation. The incident investigation team leader
needs to ensure that thes e responsibilities are clear to all groups to avoid
duplication of effort or omission of critical activities.
Management’s charter to the team should include expectations for
accurately reporting investigation outcomes. However, assigning blame or
recommending disciplinary actions should not be part of a team’s charter. A
high performance team should be as independent and autonomous as
possible, and the leader should encourage this awareness. This helps to
establish an unambiguous signal to all contributors that the investigation process will be implemented impartially. If there is a perception, either
rightly or wrongly, that the team is in any way inhibited or intimidated by
outside influences, participants and reviewers may question the quality,
quantity, and credibility
of the information collected.
It is particularly helpful to have an hourly employee from the same (or
an adjacent) plant on the team to not on ly get their valuable input, but also
to establish credibility with a wider workforce. There has been a tendency in
the past to select staff engineers as incident investigation team members
and ignore operators and technicians. Operators and technicians may know
what really happens better than others, and their involvement on the team can
produce facts that would otherwise not become known. Personnel closest to the incident occurrence, however, may also be those with a personal agenda, so this potential conflict of interest should be considered.
|
8 • Emergency Shutdowns 144
written plans of the steps and PPE to shut the equipment down,
especially during a loss event of hazardous materials and energies.
As noted in Chapter 7, Section 7.2, and, as illustrated in abnormal
and emergency operations flow chart (Figure 1.3), the operations team
has more shut down-related option s due to an unsucce ssful recovery
effort. Shutdowns activated during emergencies tend to focus on the
larger process deviations from the pr ocess aim, as was illustrated with
the different operations team re sponses to deviations during
abnormal operations (Figure 6.1) . When comparing the range of
deviations between the successful re covery efforts (Chapter 6) and the
unscheduled shut-down responses (Chapter 7), the continuum of
responses illustrates that emergency situations typically occur with significant process deviations, such as when the safe operating limits
are approached or exceeded, when there has been a loss event of
hazardous materials or energies, and after an anticipated but
unscheduled natural hazard event which has just occurred (e.g.,
earthquake). In general, the larger deviations will warrant quicker responses, especially when the safe operating limits are exceeded. In
some cases, the quicker engineered emergency shut-down is executed with a Safety Instrumented System (SIS) or an Emergency Shut-down
System (ESS) [79] [80]. This is true whether the emergency shut-down is needed for a continuous or batch process [19]. The immediate
engineering and administrative ac tions are taken and place the
process in a safer state.
Thus, when the released material causes fatalities due to toxic
exposures or to fires and explosions from ignited flammable releases,
the proactive emergency respon ses are too late to prevent harm.
Sometimes the fire or explosion has severely damaged or destroyed the very equipment designed to activate and perform the shut-down or help reduce the consequences of the loss event. For example, ignited releases of flammable materials can result in thermal impact |
22 GUIDELINES FOR MANAGING ABNORMAL SITUATIONS
Example Incident 2.3 – Texaco Refinery, Milford Haven ( cont. )
2) For supervisors / operators / technicians:
Control systems, including each co ntrol valve, should be tested
prior to all startups as part of an operational readiness review. As
a minimum, safety-critical systems should be tested after plant
trips.
Operators must remain alert to recognize abnormal situations,
such as high levels in vessels.
These example incidents illustrate that abnormal situations can result in
major, tragic, and costly process safety incidents. However, most
incidents are preventable if the ab normal situation is recognized,
diagnosed correctly, and corrected in a timely manner. Therefore, it is
essential that operations personne l have the knowledge, skills, and
abilities to manage abnormal situations.
2.4 IMPORTANCE OF TRAINING FOR ABNORMAL SITUATIONS
An abnormal situation is often recogn ized as an abnormal occurrence, but
it may not always be recognized as a po tential process safety issue. In order
to prevent these types of incide nts from occurring, whether the
abnormal situations occur during tran sient or normal operations, it is
imperative that companies understand the hazards, provide workers
with appropriate training, and have in place and enforce robust process
safety policies and procedures for a ll hazardous operations, including
startups and shutdowns. These polic ies and procedures should address
all elements of process safety (CCPS 2007a) and human factors (CCPS
2006, 2004), and specifically provide guidance indicating clearly that
abnormal situations may have a process safety component to them and are
not just operating difficulties.
If properly implemented, the RBPS Operating Procedures element
includes the safe operating limits, consequences of deviation from safe
limits, and the actions required to correct a deviation (CCPS 2007a).
However, with new technology and the increasing complexity of some |
197
others, has worked with the Amer ican Chemistry Council (ACC) to
recommend improvements in equipme nt, routing, and procedures to
enhance safety. The Chlorine Instit ute has for years acted to provide
inherently safer transport of chlo rine. The ACC Responsible Care ,
CHEMTREC , and TRANSCAER programs have resulted in significant
improvement over the years in chem ical transportation safety and
emergency response.
There are regulations in many countries governing the
transportation of chemic als, and any evaluation of transportation risks
and options must include consideration of those regulations. In addition,
some companies have policies th at require going beyond legal
requirements for specific materials. With the addition of security
concerns associated with the shipment of chemicals in the transportation sector, it is important that the implementation of inherently safer strategies take into account, at a minimum, the onsite
transportation infrastructure.
8.10.1 Location Relative to Raw Materials
It may be possible to reduce or elim inate transportation risk by locating
the plant where hazardous raw ma terials or intermediates are
produced, if the risk from tran sporting the raw materials or
intermediates outweighs th e risk of transporting th e final product(s). An
example of this IS practice followed the Bhopal incident in 1984. The few
facilities in the U.S. that produced methyl isocyanate (MIC), the highly
toxic chemical released at Bhopal , began manufacturing MIC and then
using it in situ, thereby stopping th e transportation of MIC to another
location where it was used to manufac ture pesticides. The entire process
of manufacturing and using the material was accomplished in one
location (note that currently in the U.S. MIC use in the production of
carbamate based pesticides has been replaced with a different process
chemistry, which is an application of Substitution ). Locating the starting
and ending points in the value chain of a given chemical product at the
same site will probably provide a dditional opportunities for risk
reduction by inventory reduction, i.e., Minimization . Of course, increasing
the number of processes at a particular site may increase the overall risk
at that site.
|
90 PROCESS SAFETY FOR ENGINEERS: AN INTRODUCTION
Introduction to Chemical Reactivity
The CSB report Improving Reactive Hazard Management analyzed reactive chemistry incidents
showing that they occur in a variety of equipmen t as shown in Figure 5.4 and result in severe
consequences as shown in Figure 5.5. This fo cus on reactive chemical incidents highlighted
that regulations did not address reactive chemical hazards as well as other chemical hazards.
The risks of a chemical reactivity incident result from the potential for an uncontrolled
chemical release leading to the co nsequences shown in Figure 5.5.
Figure 5.4. Equipment involved in reactive chemistry incidents
(edited from CSB 2002)
Figure 5.5. Consequences of reactive chemistry incidents
(CSB 2002)
|
DETERM INING ROOT CAUSES 255
Once all of the root causes are iden tified, the investigator is ready to
develop the corrective actions, as described in Chapter 12.
10.8.2 Quality Assurance
There are a number of quality assuranc e checks that shou ld be considered
when conducting an incident invest igation using predefined trees. Most of
these checks have already been discuss ed, although it is useful to review
them as they relate to th e predefined tree approach.
Predefined trees are designed to captur e root causes, but the predefined
tree may not necessarily be comprehensive enough to identify all root
causes. It is therefore necessary to conduct another completeness test. As
each branch of the predefined tree is considered in turn, the investigator
should ask if there are other root causes asso ciated with that category that
are not listed on the tree.
The ‘root causes’ identifi ed by applying the causal factors to a predefined
tree should be subjected to a management system test to ensure that they
are management system failures. Some predefined trees are quite detailed,
while some proprietary trees do not fully reach the under lying root cause
level. The system test essentially applie s the 5-Whys tool to each cause
identified at the end of the relevant br anches of the predefined tree.
Typically, the team may need to ask “why?” a number of times to reach
underlying root causes.
After the root causes have been identified, a generic cause test should
be applied. By considering the plant operating history, especially other
incidents that may indicate repetitive failures, the investigator may identify
other generic management system prob lems. These generic causes would
not necessarily be apparent from investigating the latest incident alone.
10.8.3 Predefined Tree Summary
Predefined trees are a convenient me ans of identifyin g root causes.
Providing all of the causal factors ha ve been determined correctly, use of a
comprehensive predefined tree should ensu re that most, if not all, root
causes are identified, especially if the management system test is performed.
Several other quality assurance tests should help identify any remaining root
causes. Table 10.3 illustrates the strengths and weaknesses of predefined
trees.
|
396
Figure 15.3 Original batch reaction system
2.metering pump, as compared to the original gravity feed where
the driving force for the Reactant B flow is always present.
3.The maximum flow rate of the metering pump is not capable of generating more heat from reaction than the reactor cooling
capacity can manage. Therefore, it is not possible to overheat the reactor by feeding Reactant B at a rate that exceeds the reactor's capability to remove heat.
4.The maximum capacity of the Reactant B feed tank has been reduced to exactly one batch charge. In this case, the same Reactant B feed tank was re-used, but it was relocated to the lower floor. To reduce its maximum capacity, an overflow was added to the side of the tank at the desired level, with the overflow piped back to the Reactant B storage tank.
|
6 Selecting a type of job aid
6.1 Learning objectives of this Chapter
By the end of this Chapter, the reader should be able to:
• Understand the different types of job aids.
• Select a type of job aid.
• Understand the use of Hazard Identification and Risk Analysis (HIRA),
Task analysis and worker involvement, in the development of job aids.
Selecting a type of job aid can be achieved in two stages.
1. Determine the need for a job aid.
2. Determine the best type of job aid to use.
6.2 Stage 1: Determining the need for a job aid
6.2.1 Overview
Procedures with many tasks are time co nsuming to write and maintain. HIRA may
be used to prioritize the higher risk task s and identify lower risk tasks for which a
job aid may not be required. In order for job aids to be accepted as necessary it is
important to produce them only when they are really needed.
I t i s g o o d p r a c t i c e t o h a v e a S t a n d a r d Operating Procedure (SOP) for safety
critical tasks (i.e., those tasks that, if performed unsuccessfully, will result in a
process safety event, see section 6.2.2.2 fo r more detail). Assigning SOPs to these
types of tasks will produce a consistent an d safe way of performing a task each
and every time and as a basis for training.
However, as noted in 5.4, it is importan t to remember that low risk tasks may
not require any form of job aid to be us ed every time a task is performed. Also,
people should be trained for tasks that must be performed very quickly, such as
emergency response, especially if task co mpletion time frames prohibit reading
through procedures.
For example:
• Frequent, low complexity tasks
Frequent and less complex tasks may not require step-by-step
instructions (or a SOP) to be used ea ch time a task is performed due to
operators having had enough experience performing the task.
Human Factors Handbook For Process Plant Operations: Improving Process Safety and System
Performance CCPS.
© 2022 CCPS. Published 2022 The American Institute of Chemical Engineers. |
APPENDIX A – CONCLUDING EXERCISES 473
Figure A.4. Node 1 – T-1 WWT Equalization Tank
Table A.3. Node 1 – T-1 Intention, Boundary, Design Conditions and Parameters
Node
# Node Node Intention Node
Boundary Design Conditions/
Parameters Operating
Conditions/
Parameters Drawings
Drawing Rev
1 T-1 WWT
Equalization
Tank Supplying variable
low pH
wastewater with
P-1 to T-1, which
serves to equalize
the feed pH being
fed forward to the
process on level
control (LIC-101). P-1 Wastewater
Feed Pump to T-
1 WWT
Equalization
Tank including
the level control
valve (LCV-101) P-1, Centrifugal
Pump, 200 GPM, 80
ft TDH, Ductile Iron.
T-1, Atmospheric
Storage Tank, CS,
33,050 gallons, 25'
Dia. x 15' Height Normal Operating
Range
- 50% level
- Ambient T 65 to
85F
- pH 4.5 to 6 P&ID
Example 0
|
158 GUIDELINES FOR MANAGING ABNORMAL SITUATIONS
Many other metrics could be includ ed to measure the effectiveness
of the facility’s management system s in identifying and preventing
process conditions that could lead to an abnormal situation and a
process safety incident. It is recomme nded that the metrics be indicators
that are specific to a process unit. Th e metrics should specifically reflect
how the unit is managing its operation safely, with an emphasis on those
metrics that can help predict or pr event an abnormal situation from
occurring or escalating to an event. The availability of data for potential
use in metrics has increased significantl y in recent years, as discussed in
5.3.3. However, it is important to ensure that the me trics are accurate
and relevant, to prevent “metric overload”. The use of a “dashboard” to
provide a high-level management summary of metrics data is
encouraged.
6.3 ABNORMAL SITUATIONS AND INCIDENT INVE STIGATIONS
Incident investigation is a way of le arning from incidents to identify
management system issues and weakne sses that can be corrected, in
order to improve the overall effectiveness of the management system. It
is particularly important to investigate high-potential near-misses that
could lead to fatalities, substantial property damage and/or
environmental damage, under differen t circumstances. While it may not
be practical to investigate every abnormal situation in depth, abnormal
situations should be investigated in order to recommend actions to
prevent, or at least minimize, their occurrence in the future.
A near-miss event can result in a serious process safety incident
under slightly different conditions if the underlying cause is not
determined and action taken to preven t it from happening again. Failure
of safety-critical equipme nt/elements such as pressure safety valves and
safety instrumented systems to work on demand, for example, can
rapidly exacerbate an already seriou s abnormal situation if operations
personnel are slow to intervene.
Like any incident investigation, the depth of analysis should be
commensurate with the actual and po tential severity of the abnormal
situation. Several sources of guidan ce are available for determining and
conducting the depth of investigation: Guidelines for Risk Based Process
Safety (CCPS 2007a) ; Guidelines for Investigating Process Safety Incidents
(CCPS 2019) ; Pressure Equipment Integrity Incident Investigation (API 2014) ; |
177
safe state if the BPCS fails to maintain safe operating conditions. A BPCS
should not be used as the sole source of a process safety shutdown. Many of the following guidance items related to the design, operation, and testing of BPCSs and SISs are not inherently safer technology in a strict sense, because they relate to active safeguards. However, much of this guidance can also be considered part of the inherently safer strategy of Simplification .
Inherently safer SISs should be fully independent of the process
control system logic residing in the BPCS, including input/output
(I/O) cards and logic solvers. Common-mode failures can also
result if the BPCS and SIS sh are components, including power
supplies and any other utility system , such as instrument air.
The SIS should normally be fail-safe, i.e., it is designed to achieve or maintain the safe state of the process on loss of power, or a
de-energize to trip design.
The BPCS should also be progra mmed to take its outputs to the
safe state if input or output signals are lost. Due to the potential in most facilities for power or other utility loss, it can be very difficult to achieve adequate ri sk reduction with non-fail-safe
design. Fail-safe designs are si gnificantly less complex, thereby
helping to implement the IS Simplification strategy.
Another example of Simplification is that operators should
always receive notification from the BPCS or the SIS as the
mandatory action points are a pproached. An everyday example
is a high temperature warning ligh t on an automobile. It gives no
indication of an abnormally high temperature until the temperature reaches the high alarm point. A temperature
indicator gives advance warning before the alarm point. The
change in display provides feedback of its functionality by normal fluctuations.
When choosing SIS input variable s, where possible, use direct
readings of the process paramete r being controlled, and not an
indirect reading. If pressure is being controlled, measure pressure directly rather than inferring it indirectly from temperature. This choice eliminates the lag time in processing that occurs when an indirect vari able is chosen. Direct readings
also eliminate potential errors in the inferred relationship |
24 GUIDELINES FOR MANAGING ABNORMAL SITUATIONS
recall their training and adapt quickl y and responsibly when things go
wrong.
In addition, during times of transien t conditions, such as unit startup,
the risk of operators having conflicti ng understandings of the state of
the process unit is greate r. Leadership should en sure that experienced,
technically trained personnel superv ise and support operators during
unit startups and shutdowns, and that effective communication and
feedback is essential to establish and maintain a mutual understanding
of the process unit and its expected future state.
Some cultural cons iderations related to mana ging abnormal situations
can have an influence on human factor s. Corporate and fa cility leadership
can drive or limit the organization’s safety culture, through words or actions.
For example, if leadership seeks to assign blame for process upsets and
incidents, the workforce will be unlik ely to report events that could
otherwise be learning opportunities and serve to reduce or mitigate
future abnormal situations. Leadersh ip must also put safety before
production and profit. For instance, there should be no pressure on
operators to continue operations, if shutdown is the correct response to
an abnormal situation (“Stop Work Authority”).
Abnormal situations introduce stress, and operators under stress
can make poor decisions, which th en exacerbate the situation. How
companies prepare and equip their operators to deal with these
problematic and stressful situations is critical to ensuring the return of
the unit to a safe state. Often, process safety incidents are a result of
organizations failing in this area.
Similarly, Conduct of Operations (COO) and Operational Discipline
(OD) are closely tied to an organiza tion’s Process Safety Culture (CCPS
2007a, 2011b). If leadership enforces high standards, then a robust COO
will ensure that operational tasks, such as establis hed management
systems and procedures, are executed in a deliberate and structured
manner. Furthermore, OD is associ ated with the organizational and
individual behaviors, and will dictate how well the management systems
and procedures are applied.
A strong and positive process safe ty culture should ensure that
operational tasks, such as operatin g and maintenance procedures, safe
work practices, and shift hand over communication, are followed
routinely and diligently. Consequently , if operations personnel perform |
HUMAN FACTORS 369
amount by which the nominal HEP can be multiplie d. HEART classifies a task into one of the 9
Generic Task Types (GTT) and assigns the nomi nal human error potential (HEP) to the task.
Error Producing Conditions (EPC) that may affect task reliability are identified. The task HEP is
calculated. HEART can help identify areas for im provement and includes strategies to reduce
errors. HEART is used in nuclear, process, me dical, and transportation industries and is
described in several papers by J. C. Williams. (Williams 1992 and HSE 2009)
What a New Engineer Might Do
A new engineer should look beyond equipment de sign and consider the role of the human in
the system of people, facilities and equipmen t, and management systems that defines the
workplace. Even in small projects and simple systems, engineer the human machine interface
to support human success.
New engineers will likely be involved in HAZO P studies and various projects that involve
teamwork. Helping to facilitate meetings and supporting good team communications can lead
to more efficient and effective teams. Simple cr itical task analysis methods can be led by new
engineers which would not only improve tasks but also build a relationship with operators and
maintenance technicians.
Tools
Human factors resources include the following.
CCPS Human Factors for Process Plant Operations: A Handbook. This book describes
human factors concepts and principles in an ea sy to understand manner. It describes how to
support human capabilities including identificati on and design of job aids to do so. (CCPS
expected 2022)
Flight-crew human factors handbook, CAP 737. The aviation industry has built significant
knowledge in human factors. (CAA) This handb ook addresses both individual and work team
human factors. “The knowledge in the handb ook was intentionally simplified to make the
document more easily accessible, readable and more usable in the practical domain.” (CAA)
Critical Task Analysis. Several approaches for critical task analysis are mentioned in Section
16.6. (NOPSEMA 2020, HSE 2000, Miller 2019)
Human reliability assessment tools. Refer to the discussion on THERP and HEART in Section
16.6.
HSE. HSE provides guidance and many references on human factors topics and human
reliability on the Health and Safety Executive webpage at hse.gov.uk.
Summary
It is helpful to think of the workplace as a thr ee-part system of people, facilities and equipment,
and management systems. Humans are an important part of this engineered system. Focusing |
Figure B-5 Interaction of the key valves and vessels
(adapted from UK HSE [87]).
|
E.14 No Incidents? Not Always Good News |301
E.14 N o Incidents? N ot Always Good N ews
The monthly KPIs for process safety incidents and
near m isses at a refinery had been very low for
several years. The new Refinery Manager was
pleased with this KPI, especially since in his first year it was zero.
In his previous refinery where he had been the Operations
Manager, the sam e KPI had been favorable but not that good.
He asked the PSM S Coordinator how the KPI was derived. He
learned that during acquisition negotiations five years earlier, the
previous owner had been challenged by several potential buyers
about the high rate of near m isses. The near m isses were not
serious and no actual incidents had occurred, but the com pany
attempted to lower their bid because of it.
After the acquisition, the refinery began investigating and
addressing near m isses less form ally. Consequently, when the KPI
program was put implemented, the near m iss result was very
positive.
Further review revealed that during the previous two years
several SISs had been activated during plant upsets or transients.
These had not been classified as near m isses because, according
to an e-mail, “the safeguards had worked as designed and that’s
not a near m iss because that was what they are supposed to do.”
Following this discovery, the facility revised the definition of
the near m iss KPI to align with the API and OGP standard for near
m iss reporting. This standard recognizes that a SIS trip usually
represents a close approach to the capability of the equipment to
contain the process, and therefore truly a near m iss. B y tracking
these types of near m isses, the facility has an opportunity to learn
about the process, culture, and PSM S without suffering any
adverse consequences. As a result, the data reported monthly
returned to values that were more typical for a large refinery. B ased on
Real
Situations |
44 PROCESS SAFETY FOR ENGINEERS: AN INTRODUCTION
Table 3.2 continued
Source Examples
NACE International standards on corrosion management
National Fire Protection
Association (NFPA) NFPA 30, “Flammable and Combustible Liquids Code”
NFPA 70®, “National Electrical Code®”
NFPA 652, “Standard on the F undamentals of Combustible
Dust”
Organization for Economic
Cooperation and Development – “Guiding Principles on Chemical Accident Prevention,
Preparedness, and Re sponse”, 2003 (OECD)
The Chlorine Institute (CI) Chlorine Customers Generic Safety and Security Checklist
The Instrumentation, Systems,
and Automation Society (ISA)
The Fertilizer Institute Recommended Practices for Lo ading/Unloading Anhydrous
Ammonia (TFI)
Standards are also written by companies and, again, can be focused on a specific technical
topic or can present a management system. Some companies have their own engineering
design standards that may take an industry code and amend it with details relevant to their
business. They may create a standard for a topic in their business that is not addressed by an
industry code or standard.
The ExxonMobil Operations Integrity Manage ment System (OIMS) is a management
system. The framework as shown in Figure 3.3 in cludes 11 elements and is the cornerstone of
their Safety, Security, Health and Environmental performance. (EM 2009)
Figure 3.3. ExxonMobil Operations Integrity Management System
(EM 2009)
|
3.2 Characteristics of Leadership and Management in Process Safety Culture |91
separately from occupational safety, treating each with equal
importance and considering their unique differences.
Likewise, external recognitions of good safety perform ance
should be considered carefully before assuming they address
process safety. If a facility has earned the prestigious OHSAS
18001 certification, its safety management system m ay address
process safety, but often it does not. Likewise, a facility that earns
Voluntary Protection Program (VPP) Star status from US OSHA
should be justly proud. However, VPP has historically focused
m uch more on occupational safety than on process safety, and in
recent years several VPP sites have experienced serious process
safety incidents.
Use Metrics Prudently The absence of process safety incidents and near misses does
not necessarily m ean that all is well, for two reasons. First, process
safety incidents are rare by nature, and facilities can go many
years without incident even as the conditions for an incident grow
m ore and more likely. Second, the apparent absence of incidents
m ay only be an indicator that incidents and near misses are not
being reported. Even favorable results on leading indicators could
be m isleading if they are the result of “check-the-box” behavior,
which can occur when m anagement values the metrics over
actual process safety perform ance.
Leaders should look behind the metrics. If lagging and leading
indicators of the health of the PSMS are always positive and no
problems are being identified, this could be an indicator of check-
the-box mentality and should at least initially be a cause for
concern. First verify that the metrics represent actual good
perform ance in the field. If good perform ance is achieved,
celebrate the teamwork and technical performance that achieved
it, not the m etrics themselves. Celebrating the metrics could |
114 PROCESS SAFETY FOR ENGINEERS: AN INTRODUCTION
handling specific hazardous chemicals such as hy drogen sulfide monitors or phosgene badges.
Fixed detectors can be located at a potential leak source such as a pump moving a hazardous
chemical. Fixed detectors can also be in the form of line-of-sight detectors designed to detect
a gas between two, distant points, for example, along a side of a process unit.
Alarm or other notification systems are typica lly activated when a fixed detection system
detects a predetermined chemical concentration le vel. Personnel should be trained to take the
appropriate action when notified by the alar m. Communication systems may extend into the
community to advise people to shelter in place.
A safe haven is needed for employees wh ere toxic concentration levels may be
dangerously high. A safe haven design includes an air-tight or pressurized facility and provides
sufficient atmosphere for the number of occupants for the duration needed.
What a New Engineer Might Do
As with flammable, explosive, and reactive mate rials, an engineer should understand the toxic
properties of materials being hand led or processed. Tools are provided in the next section that
can assist in identifying toxic properties and what exposure levels are of concern.
It is also important that an engineer follow the safety guidelines for toxics in terms of their
own protection. This includes handling chemicals in ways defined by safe operating practices
and using required personal protective equipment.
A new engineer may be asked to identify toxic hazards or to use a risk analysis, such as a
What-If analysis (see section 12.3.3) to iden tify toxic release scenarios. They may apply
inherently safer design strategies to minimize th e risk due to toxics. They could be involved in
the design of mitigation devices such as scru bbers, incinerators, and thermal oxidizers. They
be asked to ship a material or manage an in coming shipment. In these cases, communication
is important. The GHS provides classification for toxics such that there should not be any
miscommunication during the shipment or in th e labeling of chemicals. A new engineer could
be asked to use toxic release modeling to und erstand a potential incident and create an
emergency response plan.
A new engineer can benefit from reviewing the CSB investigations and videos relevant to
this chapter as listed in Appendix G.
Tools
Resources necessary to understand toxic hazards include toxicological data
resources and quantitative methods to determine the risk.
Toxicological data . Toxicological data can be found in many of the same tools and
documents listed in Sections 5.8 for reactive chemicals data. These data sources provide the
process safety information valuable in understan ding the hazards of these chemicals. Chapter
13 provides guidance on determining the cons equence effects and potential outcome of a
toxic chemical release. |
Pressure Relief Devices
237
One PRD on one enclosure
One PRD on tw o interconnected enclosure
Possibly one PRD in not enough and each
enclosure needs its own PRD
Definitely one PRD in not enough and each
enclosure needs its own PRD(d)(c)(b)(a)
Figure 12.35 Ev olution of one enclosure to two enclosures and the concept of a PRD. |
72 | 2 Core Principles of Process Safety
While the ultimate goal is to develop a single culture that
applies broadly across the com pany, subcultures can exist
within the organization. Process safety cultures can differ
between work groups and shifts in a facility, between unions
and management, among others. A survey of nine hourly and
salaried work groups in a refinery (Ref 2.5) clearly showed
culture differences between the groups and a wide divergence
in responses between workers and m anagement. Advancing
culture under such a situation may require initially addressing
each of the subcultures differently before they can be m oved
to the common desired culture. Also, the diversity provided by
subcultures can also be a source of opportunity in culture
improvement efforts, both in term s of helping identify
problems as well as providing a range of positive exam ples.
Exercise patience in culture changes Changing and im proving PS culture is like turning a large ship;
it starts with a decision for a new course, takes a long time to
reach the new heading and requires continued effort to
m aintain that direction. Leaders need to realize that culture
changes take months if not years to become fully
implemented. Our tendency is to expect prompt progress
toward a new goal. If the culture message is not consistent
across the organization and across time, it will be m arked as a
passing fad and the opportunity for lasting cultural change will
be lost.
2.11 SUMM ARY
The core principles described in this chapter describe process
safety culture in a high-level roadm ap to culture and how to
improve it. The first three principles (e.g. Establish an Imperative
for Process Safety, Provide Strong Leadership, and Foster Mutual
Trust) provide a necessary foundation for implementing the other
seven principles. |
APPENDIX G –CLASSIFYING LOSS OF CONTAINM ENT 411
Flowchart
The criteria for reporting incidents as a PSI described above are illustrated in
the attached flowchart (Figure G.1).
Figure G.1 Determining if an Incident M eets Definition of a Reportable
Process Safety Incident (PSI) under the Definitions of the CCPS Industry
Lagging M etric
(N
Process Safety Incident Severity
A severity level will be assigned for each consequence ca tegory for each
process safety incident utilizing the criteria shown in Table G.2.
|
8. Life Cycle Stages
8.1 GENERAL PRINCIPLES AC ROSS ALL LIFE CYCLE STAGES
As discussed previously, a process goes through various stages of
evolution, including:
•Concept
•Research
•Design development
•Detailed engineering design
•Procurement, construction, and commissioning
•Operations and maintenance
•Change management
•Decommissioning
This progression is typically referred to as the process life cycle. In this
chapter, each stage will be described and how IS concepts and strategies
can be employed in each stage will be detailed. Throughout the life cycle
of a process, opportunities will arise to apply the concepts and practices
of inherently safer strategies. Thes e opportunities should be evaluated
to determine if the strategies can be applied, and whether the risks and
costs are commensurate with the possibl e reductions in risk. In addition
to the eight stages listed above and in Figure 8.1, a discussion of IS
aspects of Transportation has been added to this chapter.
Transportation is associated with the lif e cycle of a process, as it is a part
of the operation’s value chain activi ties. IS strategies can be applied
onsite when planning and executin g the transportation of hazardous
materials within the facility as we ll as to and from the facility.
This chapter demonstrates that appl ying inherently safer strategies
can enhance process safety, while al so improving economic and other
objectives, such as quality, productivity, security, energy conservation,
and pollution prevention. This app lication, utilizing formal review
methods by trained individuals as id entified in Chapter 10, will link the
general principles of inherently safe r concepts to all life cycle stages.
136 (VJEFMJOFTGPS*OIFSFOUMZ4BGFS$IFNJDBM1SPDFTTFT"-JGF$ZDMF"QQSPBDI
#Z$$14
¥5IF"NFSJDBO*OTUJUVUFPG$IFNJDBM&OHJOFFST |
42 PROCESS SAFETY IN UPSTREAM OIL & GAS
of access to this knowledge was highlighted is the Longford incident (see box below)
where the important knowledge on brittle fracture susceptibility was not available
locally at the time. Important knowledge includes incident lessons and datasets,
updated engineering standards, equipment drawings and specifications, operational
experience and upsets, and new or updated process safety tools.
During engineering projects (see Chapter 7) teams can change at each stage,
and it is important that relevant process sa fety knowledge is transmitted along with
the design.
Example Incident: Longford
A well-known example of a depleting well leading to a process safety incident is
the Longford gas plant fire in Australia (Hopkins, 2000). The field, located in the
Bass Strait, was gradually producing more heavy ends. A separation column in
the plant no longer could function effectively and ultimately during an upset
allowed natural gas liquids (NGLs) to enter part of the plant not designed for it.
This reduced the temperature within a heat exchanger to -45°C causing the
exchanger to be completely frozen. Operators tried to diagnose the issue, but
process safety personnel who might have known about embrittlement had been
transferred to Melbourne and were not readily available to assist with the
diagnosis and warn of the dangers of cold temperature embrittlement. The
operators introduced hot oil to unfreeze the exchanger, but the resulting thermal
stress led to a brittle fracture rupturing the vessel causing injuries and fatalities
for those nearby. A long-lasting fire ensued and ultimately gas supply for the
entire state had to be terminated for several days resulting in significant economic
losses.
RBPS Application
Management of Change – MOC would have identified the change in processing
conditions that might allow NGLs to reach parts of the process not intended for
this service. MOC also includes organizational changes and this would have
addressed the significance of relocating pr ocess safety experts remote from the
facility.
Hazard Identification – As part of the MOC process, a HAZID would have
identified that the change in incoming fluids could allow NGLs to reach parts of
the facility not intended for this service. This could lead to flash vaporization and
cold temperatures in places with mild steel material.
Process Safety Knowledge – Since there was the potential for NGLs to vaporize
and drop temperatures to -40°C, this could cause embrittlement in mild steel.
Personnel should have been made aware of this threat to ensure their operational
responses would not impose major thermal stresses if this occurred.
Note: some additional issues for Longford are presented in Chapter 5. |
22. Human Factors in emergencies 291
Inaccurate assessment of a situation can be due to several factors such as:
• Cues from the environment may be misinterpreted, misdiagnosed, or
ignored. This creates an incorrect mental picture of the problem.
• Risk levels may be miscalculated.
• The time available to deal with the situation may be misjudged.
Additional factors contributing to inaccurate
situation assessment include: loss of situation
awareness, confirmation bias, escalation of
commitment, and/or tunnel vision.
Mnemonic and decisions aid are available for
individuals to help them make decision s during emergency situations. Examples
of decision-making aids and mnemonics are shown in Table 22-4.
Table 22-4: Emergency decision-making aids
Aid or
mnemonic Definition Use
DODAR DODAR is a cyclical model of decision-
making, consisting of the following steps:
• Diagnosis – What is the problem?
• Options – What are the options?
• Decisions – What are we going to
do?
• Assign the tasks – Who does what?
• Review – What happened? and
What are we doing about it? Useful in
emergency
situations to
provide the
steps of dealing
with abnormal
situations.
See Chapters 20 and 21 for
more information on how to
avoid perception bias,
including confirmation bias
and tunnel vision.
|
MANAGEMENT OF CHANGE 391
pre-start up or post-start up. It should be verifi ed that the pre-start up action items have been
implemented before the equipment is started up.
Other Incidents
This chapter began with a description of the F lixborough Explosion. Other incidents relevant
to management of change include the following.
Union Carbide MIC Release, Bhopal, India, 1984
Hickson Welsh Jet Fire, Yorkshire, U.K., 1992
Texaco Oil Refinery Explosion and Fire, U.K., 1994
Esso Longford Gas Plant Explosion, Australia, 1998
Georgia Pacific Hydrogen Sulfide, Pennington, Alabama, U.S., 2002
Hayes Lammerz Dust Explosion, Indiana, U.S., 2003
Formosa Plastics VCM Explosion, Illinois, U.S., 2004
BP Isomerization Unit Explosion, Texas City, Texas, U.S., 2005
Buncefield Storage Tank Overfl ow and Explosion, U.K., 2005
T-2 Laboratories Reactive Chemicals Explosion, Florida, U.S., 2007
Valero-McKee LPG Refinery Fire, Texas, U.S., 2007
Imperial Sugar Dust Explosion, Georgia, U.S., 2008
Deepwater Horizon Well Blowout, Gulf of Mexico, U.S., 2010
Williams Olefins Heat Exchanger Rupture, Louisiana, U.S., 2013
DuPont MMA Release, LaPorte, Texas, U.S., 2014
Exercises
List 3 RBPS elements evident in the Flixbo rough explosion and fire summarized at the
beginning of this chapter. Describe their shortcomings as related to this accident.
Considering the Flixborough explosion and fire , what actions could have been taken to
reduce the risk of this incident?
What is a “replacement-in-kind”?
What changes were made in the Esso Longford gas plant explosion incident?
What change was made in the Imperial Sugar dust explosion incident?
References
API RP 752, “Management of Hazards Associated with Location of Process Plant Buildings”,
American Petroleum Institute, Washington, D.C., 2009.
CCPS Glossary, “CCPS Process Safety Glossary ”, Center for Chemical Process Safety,
https://www.aiche.org/ccps/resources/glossary .
CCPS 2005, “Building Process Safety Culture: T ools to Enhance Process Safety Performance,
Flixborough”, American Institute of Chemical En gineers, Center for Chemical Process Safety,
New York, NY. |
189
Figure 8.5: An example of poor assignment of equipment identification
numbers
Figure 8.6: An illogical arrangement of burner controls for a kitchen
stove. From Ref 8.69 Norman.
|
45
Improving the reliability of critical pieces of equipment may eliminate or
significantly reduce the need fo r in-process storage of hazardous
chemical intermediates.
When designing a proce ss facility or unit, the dimensions of every
item of process equipment should be specified as large enough to
accomplish its intended purpose, and no larger. Required surge
capacities, either for normal operat ions or for emergency situations,
sometimes demand larger equipment. They are part of the intended purpose of a process design and must be maintained. Utilization of this
extra space should be kept to a mi nimum, although the process may be
modified in the future to take adva ntage of additional process capacity.
Raw material and in-process interm ediate storage tanks should be
minimized, if feasible. The need for all in-process inventories should be periodically reviewed and evaluated, particularly those of hazardous
materials.
Minimizing the size of equipment not only enhances inherent
process safety, but it can often sa ve money. If equipment can be
eliminated from a manufacturing proc ess, it will eliminate the need for
associated design, engineering, purc hasing, operating, and maintenance
costs. Equipment which is eliminat ed also cannot release hazardous
material or energy into the su rrounding environment. The true
engineering art is to determine how to accomplish a given task with a
minimum of equipment, and with the required equipment of the smallest size. Siirola (Ref 3.19 Siirola) discusses process synthesis strategies that are helpful in design ing and optimizing a process route to
minimize the required equipment and operations.
The term “process intensification” is used synonymously with
“minimization,” though the former is often used more specifically to
describe new technologies that re duce the size of unit operations
equipment, particularly reactors. It has been defined as “any chemical
engineering development that leads to a substantially smaller, cleaner, safer, and more energy efficient technology (Ref 3.17 Reay). The
European Federation of Chemical Engineering has held biennial
conferences on process intensification since 2007, and other international conference s have taken place as well. These conferences
presented several interesting possibilit ies for a range of unit operations,
including reaction, gas-liquid contac ting, liquid-liquid separation, heat |
Appendix 225
The HAZOP Team will investigate the equipment design, as well,
based on the scenarios being reviewed . The equipment should be fail-
safe, with features that automatically counteract the effect of an anticipated deviation such as a power loss. A system is fail-safe if the failure of a component, signal, or utility, initiates action that returns
the system to a safe condition [34] . Depending on the scenario and
process application, a fail-safe valve may need to close (or remain closed), open (or remain open), or remain unchanged in its current operating position, whether fully cl osed, fully open, or anywhere in
between. Thus, for unexpected shut -downs of the equipment, it
should fail-safe.
In addition, hazards analysis te ams should know the history of
issues that have occurred as th ey develop potential scenarios.
Potential fire, explosion, or toxic release issues during transient
operating modes with start-ups and shut-downs include the following
items (Adapted from [8, p. 7]):
1. Fires burning or resulting in an explosion when fuel mixes
with oxygen in th e presence of an ignition source.
2. Explosions that damage nearby equipment causing additional
releases of other flammable mater ials that may then ignite
and burn (the domino effect).
|
42
2.12 Center for Chemical Process Safety, Process Safety
Glossary, American Institut e of Chemical Engineers,
www.aiche.org/ccps/re sources/glossary.
2.13 Center for Chemical Process Safety, Report: Definition for
Inherently Safer Technology in Pr oduction, Transportation, Storage,
and Use (for U.S. Department of Homeland Security), 2010.
2.14 Council of the European Un ion, Council Directive, Control
of Major-Accident Hazards Involving Dangerous Substances (Seveso III),
2012/18/EU, June 19, 2012.
2.15 Hendershot, D.C. , Some thoughts on the difference
between inherent safety and safety . Process Safety Progress 14 (4),
227-228, 1995.
2.16 Hendershot, D.C., Implementing Inherently Safer Design in
an Existing Plant, Process Safety Progress 25(1), American Institute of Chemical Engineers, 2006.
2.17 The Institution of Chemical Engineers & The International
Process Safety Group, Inherently Safer Process Design, 1995.
2.18 Khan, F., Evaluation of Available Indices for Inherently Safer
Design Options, Process Safety Progress (22) 2, American Institute of Chemical Engineers, 2003.
2.19 Kletz, T.A., Plant Design for Safety, Rugby, Warwickshire,
England: The Institution of Chemical Engineers, 1991.
2.20 Kletz, T.A., Cheaper, Safer Plants, or Wealth and Safety at
Work. Rugby, Warwickshire, England: The Institution of Chemical Engineers, 1984.
2.21Kletz, T., Amyotte, P., Process Plants – A Handbook for
Inherently Safer Design, 2
nd Ed., CRC Press, 2010.
2.22 Lutz, W., Take Chemistry and Physics into Consideration in All
Phases of Chemical Plant Design, Process Safety Progress (14) 3, American Institute of Chemical Engineers, 1995. |
RISK BASED PROCESS SAFETY 25
Many different tools are used for HIRA. Hazard identification tools include simple
checklists, What-If analysis and HAZOP analysis. Risk assessment tools include fire hazard
analysis and explosion studies, LOPA, and QRA. Inherent safety methods and functional safety
assessments fall within HIRA.
That which has not been identified cannot be prevented or mitigated. HIRA results should
be tracked using a risk register or other tracking system. This is to ensure that no identified
issue is inadvertently neglected.
Pillar: Manage Risk
This pillar addresses many important topics for operational safety and management of risks.
These include operating procedures, safe work practices, contractor management, training,
operational readiness and conduct of operations . This pillar also addresses asset integrity,
management of change, and emergency management.
RBPS Element 8: Operating Procedures
Standard operating procedures (SOP) requires written instructions for all phases of
operation including routine, non-routine, startup, shutdown, and emergency. Good
procedures also describe the process, hazards, tools, protective equipment, and controls in
sufficient detail that operators understand the hazards, can verify that controls are in place,
and can confirm that the process responds in an expected manner.
These procedures describe how the operation is to be carried out safely, define safe
operating limits, explain the consequences of devi ation from safe operating limits, identify key
safeguards, and address special situations and emergencies.
Operating procedures have improved substant ially from the past approach of simply
taking start-up procedures from the design co ntractor. Presently, procedures are designed
with operating personnel engagement, are peri odically updated based on feedback and any
modifications, and use modern layouts with gr aphics and photographs to convey key safety
messages. Risks from deviations are highlighted – e.g. if equipment purging is required before
start-up, the procedure highlights safety risks with shorter duration purging. Barrier
management is an important aspect of process safety and the procedures highlight relevant
barriers potentially affected by the procedure.
RBPS Element 9: Safe Work Practices
Safe work practices are requirements estab lished to control hazards and are used to
safely operate, maintain, and repair equipmen t and conduct specific types of work. They
include control of work (job safe ty analysis (JSA), permits and oversight), opening pipework or
vessels, energy isolation, and other activities . These practices are used when developing
detailed work plans, ensuring that requiremen ts are met, and the appropriate safeguards
have been or will be implemented for the work . They cover non-routine work and are often
supplemented with permits. These fill the gap between operating and maintenance
procedures and the hazards and risks specific to the work being conducted at the time.
Typically, several parties are involved in safe work practices including the owner and its
contractors. Interface documents dictate what safe work practices are used and specify who
approves the work. |
147
its structural integrity. Reaction stability is a complex function of
temperature, concentration, impuriti es, and degree of confinement.
Knowledge of the reaction onset temperature, the rate of reaction as a function of temperature, and heat of reaction is necessary for analysis
of a runaway reaction. Process cond itions that result in the rapid
decomposition of reactants can also physically overpressurize a vessel
(Ref 8.13 CCPS 1995).
Toxic Hazards . The dispersion and consequences resulting from the
release of toxic materials require complex analyses that attempt to
simulate many physical and chemic al phenomena in nature and model
the movement and change in concentr ation of released materials in the
atmosphere versus time (Ref 8.25 CCPS 1996) (Ref 8.12 CCPS 2000).
Toxicological effects for humans are often expressed as a concentration (i.e., parts per million), and a number of resources are available to understand acute and chronic toxic effects. As with information
regarding flammability properties, th e SDS is the primary reference for
toxicological data. Howe ver, caution must be exercised in using
toxicological data. Some data ar e intended to describe chronic
exposures to workers, while others are intended to measure the short-
term (acute) exposures associated wi th accident situations. Most of
these data have been extrapolated from laboratory animal experiments and need to be corrected for the size and physiology of humans (Ref 8.81 Patty’s); (Ref 8.72 Rand) (Ref 8.79 USDOE); (Ref 8.1 ACGIH) (Ref 8.64 NIOSH); (Ref 8.12 CCPS 2000).
Physical Hazards . Other hazards can present a process safety risk. For
example, the simple overpressurizati on of a vessel or tank above its
maximum allowable working pressure (MAWP) (or design pressure) can
result in a rupture and release of its contents. The codes for pressure
vessel design, e.g., the ASME Boiler & Pressure Vessel Code, NR-13 in Brazil, and the European Pressure Eq uipment Directive all contain safety
margins such that vessel failure shou ld occur at some pressure above
the MAWP, however, the operating history, corrosion and damage mechanism environment, inspection practices, and maintenance will
determine how effective these safety margins are over the life of the
vessel. Another example of severe physical hazards is the hazard represented by high speed rotating equipment, e.g., the failure of a turbine blade at operating speed. |
and Inherently Safer Processes , October 8-11, 1996, Orlando, FL (pp. 416-
428). New York: American Institute of Chemical Engineers.
Berger, S.A., and Lantzy, R.J. (1996). Reducing Inherent Risk
Through Consequence Modeling. In H. Cullingford (Ed.). 1996 Process
Plant Safety Symposium , Volume 1, April 1-2, 1996, Houston, TX (pp. 15-23).
Houston, TX: South Texas Section of the American Institute of Chemical
Engineers.
Berglund, R.L. and Snyder, G.E. (December,1990). Waste
minimization: The sooner, the better. Chemtech, 740-746.
Black, H. (1996). Supercritical carbon dioxide: the “greener”
solvent. Environmental Science and Technology 30 (3), 124A-7A.
Blumenberg, B. (1992). Chemical reaction engineering in today's
industrial environment. Chemical Engineering Science, 47 (9-11,) 2149-
2162.
Bodor, N. (October, 1995). Design of biologically safer
chemicals. Chemtech, 22-32.
Borman, S. (November 30, 1992). Aromatic amine route is
environmentally safer. Chemical and Engineering News , 26-27.
Bradley, D. (August 6, 1994). Solvents get the big squeeze. New
Scientist , 32-35.
Bradley, D. (April 29, 1995). Incredible shrinking visions. New
Scientist , 46-47.
Brennan, D.J. (May, 1993). Some challenges of cleaner
production for process design. Environmental Protection Bulletin 024, 3-
7.
Burch, W.M. (1986). Process mo difications and new chemicals.
Chemical Engineering Progress, 82 (4) 5-8.
Callari, J. (November, 1992). En vironmental pressures force
widespread change. Plastics World , 40-43.
Calvert, C. (November, 1992). En vironmentally-friendly catalysis
using non-toxic supported reagents. Environmental Protection Bulletin
021, 3-9. 472 |
RISK MITIGATION 347
Figure 15.9 b. The right side (consequence legs) of a bow tie for loss of containment
(CCPS 2018)
Risk Reduction Measures
Processes that pose risk are provided with risk reduction measures. These measures may
prevent an incident from occurring or mitiga te the consequences as illustrated in Bow Tie
Analysis. They may be safeguards , barriers, or IPLs as discu ssed in section 15.3. Additionally,
they may be passive or active.
Passive Hardware - A barrier system that is continuously present and
provides its function without any required action. (CCPS 2018)
Active Hardware - A barrier system that requires some action to occur
to achieve its function. All aspects of the barrier detect-decide-act
functions are achieved by hard ware or software. (CCPS 2018)
Active barriers may be pieces of equipment, human action, or a combination of the two,
for example, an operator closing a valve in response to an alarm. Table 15.1 provides a list of
potential risk reduction measures and classifi es them in these categories. Many of these
measures are fundamental to process safety. Se lect the appropriate risk reduction measure
for the application. It is also important to recall the concepts of inherently safer design (Section
10.7.2) and the hierarchy of controls (Secti on 10.7.3) when making decisions on the
implementation of risk reduction measures.
Not all measures meet the requirements to be considered an IPL; however, that does not
make them less important in the overall management of process sa fety. Items such as training,
procedures, and signage are key aspects of preventing process safety incidents.
|
404 PROCESS SAFETY FOR ENGINEERS: AN INTRODUCTION
Car Seals. Car seals, are devices for physically lock ing valves in position where the position
is critical to safe operation. (see Figure 19.4) They play an important role in the effectiveness
of process safety systems. P&IDs should indicate whether the valve is car sealed open or car
sealed closed (CSO or CSC). By itself, a car seal does not necessarily prevent a position critical
device from being moved to an unsafe position. It is what the seal represents and how it is
managed that keeps the device in the appropriate position. The seal can be broken in an
emergency to change the position of a valve. Chains and locks are sometimes used instead of
car seals.
Figure 19.4. Car seal on a valve handle
(Wermac)
Good practices for position critical devices in clude having a written procedure, frequently
verifying correct position, using MOCs for changing the position, and verifying position as part
of a PSSR.
Equipment Labeling . The purpose of proper equipment and piping labeling is to support
plant operations and maintenance activities. G ood practices include labeling of all equipment
(including spares), safety instrumented system components, piping, utilities, and safety critical
double block and bleed.
Process Oversight
Process Readings and Evaluation . Operators collect information on the process status and
evaluate that information to determine if processes are running efficiently and meeting
|
134 PROCESS SAFETY IN UPSTREAM OIL & GAS
FEL-2 develops selected options further and usually one option is progressed
to initial mass and energy balances, outline layout and PFDs, and equipment lists to
allow an outline costing to be developed.
Once the final option is selected, the initial Hazard Identification and Risk
Analysis studies, risk register, Inherently Safer Design review, and Concept Risk
Analysis are all updated. Final option selection involves a multi-variable balancing
of project finance, potential process safety and environmental impacts, and project
risks (e.g., construction risks, weather).
Initial engineering design activities are undertaken including: establishing the
design philosophy, identifying relevant regul ations and standards, facility siting or
module layout, conducting preliminary studies for fire and explosion analysis, fire
and gas detection, fire hazards, firewater and foam needs, blowdown and
depressurization, transportation risk, and security vulnerability.
The ISD is updated in FEL-2 as part of the final option selection process. Sutton
(2011) describes the application of inherent safety into an offshore FEL stage design.
Options are screened out based on good design principles and the ISD hierarchy –
using Figure 7-2 as a guide for ranking. The ranking is discussed in the box
explaining the concept of Inherently Safer Design. ISD options include those related
to workforce exposure and the use of reduced personnel levels or even unmanned
facilities, thus reducing risks. Other ISD options may relate to facility layout and
sizing of hydrocarbon storage. There may be a trade-off where safety is enhanced,
but environmental impacts increased (e.g., offshore subsea processing mostly
eliminates safety issues, but smaller undetect ed leaks persist for longer than if on a
manned facility). Both aspects are important, and the design selection should
consider each in the option selection.
Where a CRA (either QRA or consequence analysis) is performed, it is refined
at the FEL-2 stage when outline design activity occurs, and it is part of the ranking
of the different options. As before, the risk estimates are at more of a screening level
of detail but may be sufficient to differentiate options.
There may be adequate informa tion at this stage to address Asset Integrity and
Reliability. It might establish the need for more costly alloy materials to deal with
corrosive fluids. It might also address whether a single equipment item is sufficient
or there should be redundancy to provide increased reliability.
The relationship between various risks and risk reduction options should be
considered at this stage. For example, where a normally unmanned installation is
proposed which reduces personnel risks, then the equipment should be designed
with sufficient integrity to minimize the need for maintenance (which could
inadvertently increase manning levels). |
6.3 Improving the Process Safety Culture of an Organization |227
Create clarity of purpose.
In this step, the clarified roles and responsibilities are rolled
out, along with the process safety culture mission, vision, and
goals. Leaders should conduct awareness orientations about the
desired process safety culture for all personnel. The orientation
should cover the core principles of process safety culture, the new
expectations, and the plan to assess and improve culture.
The orientation should be conducted by the organizational
leaders personally. Culture comes from strong, committed
leadership, and a leader who is absent from the roll out, or one
who introduces the orientation and then leaves may be perceived
as less than committed. It is also important to involve lower-level
leaders in conducting the orientation. When workers see their
supervisors aligning with the new culture, they will be more
m otivated to follow.
Ideally, everyone should be connected in some way to
addressing the cultural gaps and working towards the vision.
Following the orientation, normal work teams should be engaged
to address the cultural gaps identified in the assessments.
Training to im prove overall process safety competency should
also be done. There may be som e resistance to this. Some m ay
already believe they are fully knowledgeable, and some managers
m ay believe they do not need to know process safety because
their subordinates handle it. Leaders do not have to conduct this
training (as for the orientation). However, they should show a
visible commitm ent to the training. For exam ple, they might
personally kick-off training sessions by explaining their
importance and their expectations of trainees.
Some sort of visual representation of progress should be
created, and then m aintained. This could be based on metrics (e.g.
% of workforce involved) or on milestones (e.g. depicted on a
flowchart color-coded to show items completed, in progress, not
yet started, etc.) Once started, this should be continued. If the |
Containers
145
9.4 Transferring Fluids Between
C
ontainers
The duty of containers is storing or holding fluids and
also allowing fluids to flow out.
The latter duty is not always an easy one. A similar
issue in our daily lives is shown in Figure 9.2.
When transferring fluid from point “ A” to point “B, ”
point “ A” should be able to manage a lack of fluid and somehow break the vacuum created, and point “B” should be able to manage the accumulation of fluid and pressurization at point “B. ” If either of the points, or both of them, fail to handle the pressure changes, the flow will be stopped. This is one of the most commonly overlooked requirements in developing container sys -
tems in P&IDs (Figure 9.3).
From a theoretical viewpoint there shouldn’t be such
an issue at all. A plant theoretically operates in a steady state and wherever there is flow‐in into a container, there should be a flow‐out equal to the flow‐in. However in practice there are units/plants that are working in batch‐wise or semi‐continuous modes of operation. Even in fully continuous operation plants there few times that the plant operates in a fully steady state condition. Therefore there is always the chance of liquid accumulation in con-tainers and the creation of low pressure in the source tank and creation of high pressure in the destination tank.
This issue is mainly for liquids, in tanks and not vessels,
and specifically for larger tanks.
There should be provision to take care of the atmos -
phere above the liquid level. Without such provision flow is stopped, or when the liquid level in the container decreases a vacuum will be created in the space in the top of the container. If this vacuum is not broken the container will collapse. In the case that the liquid level in a container increases, the space in the top of the container will be over pressurized. If this overpressure is not released the container will explode.
There are at least four different ways to deal with this
issue. They are explained below and are shown in Figure 9.4.Solution 1: do nothing. This solution can be used when the
amount of vacuum or overpressure is very slight and at the same time the liquid is near its boiling point. This means the liquid can easily be converted to vapor and vapor can be easily converted to liquid. In this solution the slight vacuum created will be compensated for by additional liquid evaporation and the slight overpres -
sure will be mitigated by a small conversion of vapor to liquid. This solution is not very reliable and is rare.
Second hole is created
Liquid exiting smoothly Exiting liquid is “glugging”
Figure 9.2 Pr oblem of pouring liquid out of a can.
Vacuum
Here!Over pressure
here!
Container AW ith or
without pumpContainer B
Figure 9.3 Fluid tr ansfer between containers.
Equilibrium line(a)
(b)
(c)
(d)
Figure 9.4 Differ ent ways of facilitating flow. |
LESSONS LEARNED 349
been inhibited. The associated re commendation was to reinforce the
management of change system . This i s an issue that occurs across many
industries and these de tails could be extracted from the case study,
summarized on a single sh eet as shown below in Figure 16.1 (Safety Alert)
and used to communicate the co mmon learning that applies.
Figure 16.1 Example Safety Alert
|
192 | 5 Aligning Culture with PSMS Elements
communication , particularly in traditionally hierarchical cultures
where asking questions for the purpose of learning is an
appropriate way to initiate communication.
Key employees in PSMS roles should not restrict com petency
building to internal developm ent efforts. Since process safety
incidents tend to be rare events, it is im portant for process safety
personnel to participate in local, national, or global industry or
technical organizations, meetings and conferences. This gives
them direct access to lessons learned from other com panies
across the industry. A person’s access to lessons learned is greatly
facilitated by sharing their own lessons learned, Doing so fosters
mutual trust . Some companies may be uncom fortable with this
level of sharing. However, the value of sharing is so high that
com panies should find ways to appropriately manage the details
of what is shared while enabling the exchange of lessons-learned.
There is ongoing debate whether senior facility and corporate
leaders should have experience in the processes and technologies
that they m anage. The debate addresses other technical and
m anagerial disciplines beyond process safety. The school of
thought embraced by followers of the HRO approach (Appendix
D) believe that technical competence in the discipline is essential,
particularly when it com es to preventing catastrophic incidents.
For example, in the nuclear industries of many companies, facility
m anagers must spend a m inimum amount of time in a nuclear
safety role. The other school of thought believes that leaders do
not need to have had experience in the discipline; they need only
surround them selves with staff having the necessary com petency.
Certainly, having the technical expertise helps, particularly in
preventing the normalization of deviance . However, either approach
can work, from a process safety culture perspective. The bottom
line, with or without technical experience, leaders should: Understand the PSMS and its underlying principles,
Know the hazards their organization is m anaging, •
• |
EQUIPMENT FAILURE 191
Table 11.1. Failure modes and design considerations for fluid transfer equipment
Failure
mode Causes Consequences Design considerations
Stopping Power failure
Mechanical failure
Control system action
(failure or intended) Consequence to upstream
or downstream equipment
(HIRA needed)
See Reverse Flow Power indication on pump
Low flow alarms/interlocks
Level alarms and interlocks in
other equipment
Deadhead
ing or
Isolation Pump/compressor
outlet blocked in by:
Closed valves
(manual, control,
block) on discharge
side,
Plugged lines
Blinds left in Loss of containment due to,
high temperature and
pressure causing seal,
gasket, expansion joint,
pump or piping failure.
Possible phase changes,
reactions. Overpressure protection.
Minimum flow recirculation lines.
Alarms/interlocks to shut down
the pump or compressor on low
flow or power
Limit closing time for valves
Cavitation
/ Surging Blocked suction by:
Closed inlet valves
Plugged
filters/strainers Loss of containment due to
damage to seals or
impellers Low flow alarms/interlock to shut
down the pump or compressor
Vibration alarms/interlocks
Differential pressure alarm on
strainers
Reverse
Flow Pump or compressor
stops Loss of containment
upstream
Overpressure upstream
Contamination upstream Non-Return (Check) valves on
discharge side (Check valves are
difficult to count on; their
dangerous failure modes are
difficult to diagnose or test for
until they are actually needed.)
Automatic isolation valves
Overpressure protection
upstream
Positive displacement pump
Seal
Leaks Particulates in feed
Loss of seal fluids or
flushes
Small bore
connections
Age (wearing out) Loss of containment due to
damage to seals Alarms or interlocks on seal fluid
system to shutdown
pump/compressor
Double mechanic al seals with
alarm on loss of one seal
Sealless pumps
Contamin
ation /
change of
fluid Liquid in compressor
feed Compressor damage
See Seal leaks Knock out pots before
compressor
Figure 11.11 shows a Pump Application Data Sheet . The first block of information, Liquid
Properties, specifically asks fo r safety information such as flammability, toxicity, regulatory
coverage. Other properties that could be of inte rest could be thermal stability or reactivity of |
2.8 Defer to Expertise |57
Employees should feel that they can stop any activity when
they notice a potential hazard, even when stopping may have an
impact on production or costs. They should feel that these actions
can be taken without retribution from either fellow workers or
m anagement, and that second-guessing from any party regarding
the consequences of such actions will not occur.
Empowerm ent promotes feelings of self-worth, belonging and
value. Employees should be involved in training, should be
consulted about the content of the PSMS, and should participate
to the extent possible in all process safety activities.
2.8 DEFER TO EXPERTISE Atlantic Ocean, Offshore of Florida, USA, January 28, 1986
Shortly after launch, the external fuel tank of the space shuttle
Challenger exploded, dooming the shuttle and all 7 crew
m embers. The fuel tank was breached by a sizeable leak of hot
gases through two O-rings that sealed a joint in one of the
orbiter’s two booster rockets. On the day of launch, the
tem perature was significantly colder than the O-rings were
designed to seal. However, NASA management failed to Defer to
the Expertise of the booster rocket program engineer and
launched anyway.
The investigation revealed a significant Normalization of
Deviance (Ref 2.32), as NASA launched shuttles at colder and
colder tem peratures, accom panied with greater and greater
burn-through of the O-rings.
The Challenger space shuttle disaster of January 28, 1986,
was a major turning point in the consideration of culture in
highly technical operations. However, NASA failed to Learn to
Assess and Advance the Culture , leading to the loss of the shuttle
Columbia from another Normalization of Deviance situation. |
DETERM INING ROOT CAUSES 227
A larger version of the tree is shown as Figure 10.14, although it is not
completely developed. (The figure is turned for better viewing.)
Figure 10.14 Logic Tree, Slip/ Trip/ Fall Incident
|
Table 26-3 continued
Investigatory
tools Description of tool
Tripod Beta
[115] The Tripod Beta technique consists of three steps:
1. “What happened?” – develop a diagram that shows the sequence of events in the accident.
2. “How did the incident happen?” – identify failed, inadequate, missing , and effective barriers. This is to
identify risk management measures that should have been in place.
3. “Why did the accident happen?” – create a causation path that iden tifies immediate causes and related
human failures of failed barriers, pre-conditions in fluencing the immediate causes, and underlying (root)
causes that created the preconditions.
(adapted from [115] )
Precondition
Underlying
cause
Immediate
cause
Agent
Barrier
Event
Object |
Chapter No.: 1 Title Name: Toghraei c14.indd
Comp. by: ISAKIAMMAL Date: 25 Feb 2019 Time: 12:29:27 PM Stage: Proof WorkFlow: CSW Page Number: 269
269
Piping and Instrumentation Diagram Development, First Edition. Moe Toghraei.
© 2019 John Wiley & Sons, Inc. Published 2019 by John Wiley & Sons, Inc.Companion website: www.wiley.com/go/Toghraei_PID
14.1 Introduction
This chapter is not meant to be a definitive course in
control system design. Instead, I want to take you through the process principles involved in designing a system. There are two main methods that can be used:
●Design by analysis. This method is used when the unit operation is complex enough that it requires mathe-matical equations and chemical process data to render a solution. For instance, this method would be used by control engineers in the design of the control system for a fractionation tower or distillation column. In these operations, where composition is a vital process parameter, you may have many side streams that draw off different end products at various stages in the ves -
sel. So you can imagine that this requires quite a lot of analysis to design the control system correctly.
●Design by intuition. This method involves a mixture of gut feeling, practical experience, and observation to provide a control solution. This is the way I learned, and it is also the method used by most designers. This method can be used because most process operations are not complex enough to warrant a full mathematical analysis for control purposes. Apart from that, many items of equipment like pumps, heat exchangers or boilers have established a control methodology that works and has been tried and tested over many years, so it can be learned easily.
In this chapter, we focus on the second method: design-
ing a control system by intuition.
14.2 Control System Design
There are four steps involved in designing a control system:
1)
Se
lecting the parameter you want to control and the
location of the sensor.
2) Iden
tifying the manipulated stream, or the stream on
which you want to place a control valve.3) De
termining the set point.
4) Building the c
ontrol loop.
14.3 Selecting the Parameter to Control
In this step, you are basically selecting the type of sensor
and its location by identifying the process variable you need to control. In the majority of cases, the sensor should be placed on the stream whose process parameter you want to control.
Later you will see the two main different types of con-
trol loop architectures, which are “feedback” and “feed-forward. ” The above statement is valid only for feedback (FB) loops. It means in feedforward (FF) loops, the sen-sor does not necessarily need to be located on the stream that is to be controlled.
How do we go about selecting the right parameter to
control?
The approach that I take is to use Table 14.1 as a rule of
thumb to help with parameter selection.
This table is useful because, for each parameter, it gives
you examples of where the sensor should be placed and also the point in the process where the parameter must be adjusted. There are a few points to note about this table:
1)
Alwa
ys control the inventory in your process.
2) Pre
ssure for gas vessels works similar to level for liq-
uids. Both of them work for inventory control.
3) Do t
emperature control wherever there is a piece of
equipment that causes a change in temperature.
4) Do c
omposition control wherever there is a piece of
equipment that causes a change in composition.
5) Comp
osition control is not common for at least two
reasons: one, because theirs sensors (process analyz-
ers) are slow and not very reliable, and two, because composition is generally a function of other parame-ters and by controlling temperature, pressure, etc. the control of composition can be achieved.
6)
Flow rat
e is only controlled in piping systems (obvi-
ously), and control loops are located near fluid movers.14
Application of Control Architectures |
Pipes
77
The limitations could be the available pipe sizes or una-
vailability of some valves in the spec.
There are some cases that we are looking for some-
thing which is not available in the selected pipe spec.
There are at least four ways to deal with this issue:
1) Change t
he whole pipe spec by changing the interpre-
tation of the piping spec commodity in a new, less radical way and move the fluid name to another less restrictive pipe spec.
2)
Change t
he pipe spec to a less restrictive but still
acceptable spec (a more rigid pipe spec).
3) Change t
he process design to obey the restrictions of
the current pipe spec.
4) As
k the material group to update the current piping
spec to cover the item of request.
It means there could be a specific piping spec for the
utility water that has pipes only from 2 to 10 inche
s when
you are looking for a 16‐inch pipe in the plant.
The solutions are:
1) Is ther
e another less restrictive pipe spec, for exam-
ple, for raw water? Does it have 16‐inch pipe in it? If the answer is yes, use it and change all the pipe specs to that one.
2)
Is t
here another less restrictive, but more expensive
pipe spec, for example, for potable water? Does it have 16‐inch pipe in it? If the answer is yes, use it and put spec breaks for the 16‐inch pipe to use this less restrictive but more expensive pipe spec.
3)
Can
you replace the 16‐inch pipe with two 12‐inch
pipes in parallel to get over the limitation in the exist -
ing pipe spec?4) As a last resort, ask the material group to update the
piping spec table and extend their piping spec to cover the item of request (i.e.16‐inch pipe). This solution is not the best solution because companies are usually not willing to change their piping spec and changing the piping spec may take a few weeks to a few months.
The other limitation could be unavailability of some
valves. In some piping specs, some specific valve types are not available. For example, in one piping spec, there could be no ball valve. So, if this is the specific pipe spec, it needs to be ensured that no ball valve is installed on the pipe. If there is a need to put a ball valve, the preceding solution can be used. An additional solution for valves is changing the valve to a similar valve with an actuator. Using this trick, the valves become beyond the piping specs and are transferred to the Instrumentation and Control group who may accept and approve the requested valve.
6.3.4
Pipe Siz
e
The pipe size, or pipe diameter, to carry a fluid from
point A to point B is already specified during the design phase of project. However, it is a good idea to have some practical understanding about these parameters.
Pipe size is generally mentioned as part of a pipe tag.
However, the way that the pipe size is mentioned is dif -
ferent. Without going through different pipe standards, generally, in North America the pipe size is reported as nominal pipe size (NPS), which is an approximate size and not necessarily beyond the pipe diameter. NPS is generally stated in inches and an 18‐inch pipe can be written as 18
in. or 18″ pip
e. Another way of stating
Piping material spec
Out steam:
Commodity: Water
T=T0 °C
P=P0 kPagCommodity: Water
T=T 1/uni2192 T2 °C
P=P 1/uni2192 P2 kPagCommodity: xxxxx
Commodity: xxxxx
Commodity: xxxxx
Commodity: xxxxx
Commodity: xxxxx
Figure 6.13 Specifying pipe spec. |
344 Human Factors Handbook
• Conducting two to three discussion se ssions over a period of time with
individuals to:
o Fully capture their perceptions of the incident.
o Allow for individual later memory recall.
o Allow time for post incident-stress recovery. For example,
people may remember more information if asked about the
incident a few days or a week after it happened. This is because
people sometimes need time to process any negative emotions
and to allow their stress levels to fall.
26.5.3 Avoiding bias in investigations
Incident investigations should avoid investigation bias to ensure the gathered
information is an objective and true reflec tion of events. Some investigation biases
and strategies to lessen the impact of these biases are provided in Table 26-2. |
Ancillary Systems and Additional Considerations
403
he may need to choose the inherent solution. In the
inherent solution all the design pressures of connected items are equalized to the highest value. This is a very conservative approach that not all companies welcome because it may increase the cost of project to a high, unacceptable value. For example one company may say: “all the tanks connected to a VRU system through a vapor collection network should have the same design pressure (@ design temperature). ”
Although this logic cannot be completely overruled, it
is less likely to happen in a large scope and it is also very expensive to implement.
A more common example of where this situation hap-
pens a lot during P&ID development is what is shown in Figure 18.33.
What should we do when tying‐in two pipes together
with two different design pressures?
It is obvious that the operating pressure of the two sep-
arate pipes will both be changed to new values after the tying in. In an inherent solution the design pressure of the lower rating pipe should be increased to the higher rating (Figure 18.34).If this solution is not acceptable from an economical
viewpoint the limiting pressure can be implemented. In this solution a pressure regulator together with a PSV can be placed on the lower rating pipe, right after the tying in point. A single or double check valve could be placed to prevent high pressure to “migrate” to the upstream of a lower rating pipe.
Placing a check valve for this purpose is very tricky and
is not always acceptable. The reason is that a check valve may prevent reverse flow but not the reverse pressure! Then putting in a check valve doesn’t necessarily prevent high pressure from reversing to upstream of the tying point.
This lack of reverse pressure prevention is because
no check valve can 100% prevent backflow. Generally speaking a conventional swing check valves passes flow in the reverse direction in about 10% of main flow.
18.8.5.2 Design P ressure of Connected
Equipment–Sensor
For connected equipment–sensor, there are at least two available solutions: equalizing the design pressure to the highest value and do nothing!
The solution of limiting and allocating pressure is
generally not available for this case.
Some companies prefer to put the design pressure of
all instruments connected to – for example – a piece of pipes equal to the rating of the pipe. This could be an expensive approach and not all companies like it. It is not strange to see that the design pressure of connected instruments to a piece of pipe is lower than the rating of the pipe with no means of pressure limitation and with no concerns!
This “do nothing” approach could be taken by some
companies based on the logic that: “losing the pipe and rupturing it is not affordable by us but we can afford to lose a small instrument if the pressure goes beyond the design pressure of the instrument. ”
Therefore the answer to the question of: “should the
design pressure of a flow sensor be the same as the design pressure of the pipe it is installed on it?” The short answer is: “not necessarily. ”Design P=500
Design P=300
Figure 18.33 Tying t ogether two pipes with different ratings.
Design P=500
Design P=300Design P=500
Figure 18.34 Equaliza tion of pipe rating after tying them in
together. |
6 PROCESS SAFETY IN UPSTREAM OIL & GAS
This selected list of incidents with lo ss of containment shows how serious major
process safety events can be to people, the environment, and to business. Marsh
(2020) lists many serious upstream losses in its 100 largest losses review. This
source lists major losses due to process leaks (e.g., Piper Alpha), blowouts (e.g.,
Deepwater Horizon), harsh weather (e.g., Ocean Ranger), failure of marine systems
(e.g., Kolskaya towing incident), struct ural failures (e.g., Alexander Kielland leg
collapse and sinking) and transportation (e.g., helicopter incidents). Details on all of
these are available from the relevant regula tor or by a literature search. While this
book focuses on process safety, readers ca n learn from all incidents in efforts to
improve overall upstream safety.
Although this book does not delve into specific regulations, certain regions
require process safety and other major hazards to be addressed as part of the
permitting process. Examples include the US SEMS and OHSA 1910 rules, Europe
(EU) Offshore Oil & Gas Safety Directiv e, Norway PSA requirements, Australia
NOPSEMA rules, and Abu Dhabi ADNOC requirements. The implementation
requirements are different, but all require major hazards to be identified, assessed
for risk, and managed with an effective safety and environmental management
system. A short summary of international regulations is provided in Section 2.8.
This book is needed for several reasons.
1.Major incidents in the upstream industry such as Piper Alpha (IChemE,
2018) and Deepwater Horizon (National Commission, 2011) show that
robust process safety management is beneficial not only to reduce events
but also to demonstrate to the public that the industry is managing its risks
effectively. This latter aspect is important for the community and regulators
to have confidence in the industry and thus allow continued or new
operations. This book is intended to help improve process safety
performance, thus supporting the indu stry as a whole. A similar argument
applies to the downstream industry as well.
2.Newcomers to the industry can benefit from a text specifically explaining
process safety in the context of the upstream industry. SPE has a substantial
library of books in its textbook series. These focus on technical aspects of
well design and upstream operations, rather than process safety. This book
fills a gap in SPE literature.
3.The upstream industry can leverage learnings from both its own incidents
and those from downstream related to hard-won lessons of major incidents
such as Flixborough and Bhopal (both described in Lees, 2012). These
lessons have been codified by CCPS in a series of over 100 Guideline texts
addressing process safety, including Risk Based Process Safety (CCPS,
2007a). CCPS was created as the US downstream industry response to the
Bhopal disaster in 1984. It has had its objective to put into the public
domain the best practices for process safety. Most of these are equally
applicable to upstream as well, alth ough the technical terms and examples
may differ. This book is an access point to many of the other CCPS texts. |
60 Human Factors Handbook
Figure 6-4: Task safety criticality rating
(adapted from [33] )
6.2.3 Other factors
Guidance is provided Table 6-1 and Tabl e 6-2 for rating the remaining factors.
Some low complexity tasks may be perfor med frequently, such as depressurizing
oil storage tanks every day. However, so metimes the circumstances may change.
For example, a change in wind direction and speed may require special
precautions, such as turning off ignition sources downwind of the tanks. This could
be a low frequency task and higher complexity.
|
174 INVESTIGATING PROCESS SAFETY INCIDENTS
examine the actual failure sites to identi fy the nature of the failure, such as
fatigue, stress corrosion cracking, in tergranular stress corrosion, or
embrittlement.
The timeline tool pulls all of this in formation together into a manageable
record of events and sequence provid ing a perspective conducive to proper
causal analysis.
8.4.2 Constructing a Sequence Diagram
Organizing Data with Sequence Diagrams
Sequence diagrams are a more elaborat e graphical depiction of a timeline
that allow the investigator to present related events and conditions in
parallel branches. As with a timeline, begin construction of the sequence
diagram at the earliest opportunity, as soon as the initial facts become
known about the incident. By starting earl y, the investigation can spot
missing information or inconsistencies in the “facts” and focus upon resolving those gaps.
A diagram depicting the sequence of events leading to an incident has
a number of advantages over a simple timeline that can be summarized in
three main areas: investigation, identi fying actions, and reporting as shown
below (Ferry, 1988).
Investigation
• Summarizing the events in the form of a diagram provides an aid to
developing evidence, identifying ca usal factors, and identifying gaps
in knowledge.
• The multiple causes leading to an incident are clearly illustrated.
• Diagrams enable all involved in the investigation to visualize the
sequence of events in time, and th e relationships of conditions and
events.
• A good diagram serves to communicate the incident more clearly
than pages of text and ensures a more accurate interpretation.
Identifying Actions
• The diagram provides a cause-or ientated explanation of the incident.
• Areas of responsibility are clearly defined.
|
Chapter No.: 1 Title Name: <TITLENAME> c19.indd
Comp. by: <USER> Date: 25 Feb 2019 Time: 12:34:26 PM Stage: <STAGE> WorkFlow: <WORKFLOW> Page Number: 405
405
Piping and Instrumentation Diagram Development, First Edition. Moe Toghraei.
© 2019 John Wiley & Sons, Inc. Published 2019 by John Wiley & Sons, Inc.Companion website: www.wiley.com/go/Toghraei_PID
19.1 Introduction
The purpose of this chapter is to show a general meth-
odology for the development of P&IDs and a general methodology for checking P&IDs.
19.2 General Procedure for P&ID
De
velopment
The question is how to develop the P&ID of an item that is new to you.
Let’s look at a process plant from a bird’s eye view
(Figure 19.1).
The main units in process engineering are: conversion
units and separation units. The conversion units could be physical conversion units or chemical conversion units.
The other items, that we name general items, can be
considered as “peripheral” items, and their duties satisfy the main conversion units.
We have already learned how to develop P&IDs for
general items like pipes, pumps, compressors, heat exchangers, etc. However, it is not always the case that a design engineer (in the role of P&ID developer) should develop the P&ID for general/popular items (e.g. con-tainers, fluid movers, heat exchangers, etc.). In those cases where he is faced with new items (less popular items like a liquid extraction tower, filter press, etc.) he should have the capability of developing the P&ID. It is not very easy to develop the P&ID when you are totally unfamiliar with the item, but it is not impossible.
The first step is to learn the function of the new piece
of equipment and its principles of operation. Talking to the vendor and several users of the piece of equipment helps a lot in developing a good P&ID for a piece of equipment. Interviewing vendors is easy because they want to sell their equipment to you but finding good users for the equipment is not easy.
First of all users are generally hesitant to talk because
the transferred information may be considered as proprietary information and inhibited. The second issue is that getting unbiased information is very difficult. The third issue is every user’s experience is gained in a specific service, specific weather, etc. and may not be considered as a “general” idea. In the end, it is the skill-fulness of the P&ID development engineer to “extract” the pure facts from the interviews.
P&ID development is nothing but developing provi-
sions to cover all four stages of the life cycle for every single piece of item on the plant.
These four stages are, again, normal operation, non‐
routine operation, maintenance/inspection, and the absence of the item from operation.
Here we develop this strategy in two sections, a piping
and equipment section and an instrument, control, alarm, and SIS section.
19.2.1
P&ID Dev
elopment: Piping
and Equipment
Out of the four stages of each piece of equipment, the
“normal operation stage” generally doesn’t need much from a piping viewpoint. The majority of items are needed for the three other phases of operation.
Each of the items needed to be added to cover these
fours stage may need an additional control system.
For the piping and equipment section these sample
questions could be asked:
1)
The e
quipment may need partial recycling if the
function of the equipment improves because of the
recycling. Examples of such equipment are reactors with equilibrium reactions and the units where “probability” is a factor (like the floatation process in mining).
2)
How low c
an the flow rate be for the equipment to
work comfortably, and is there any expectation that the flow will go below that “minimum acceptable flow”? What happens if the flow goes below the “minimum acceptable flow” in the short and long term?19
General Procedures |
345
processing of chemicals having explos ive properties; 3) improved hazard
consideration for hydrogen; 4) additi onal special process hazards; and,
5) inclusion of toxi city in assessment.
The Mond Index divides the plant into individual units and takes into
consideration plant layout and the creation of separating barriers
between units. The hazard potential is initially ex pressed in terms of a
set of indices for fire, explosion, and toxicity. The hazard indices are then
reviewed to determine if design changes reduce the hazard, and the
revised values. Factors for preventative and protective features are applied, and then final values of the indices are calculated.
13.7.4 Proposed Inherent Safety indices
The Integrated Inherent Safety Index (I2SI), developed by Khan and
Amyotte (Ref 13.19 Khan) , addresses the economic evaluation and
hazard potential identification for each option within the process life
cycle. I2SI is comprised of sub-indi ces accounting for hazard potential,
inherent safety potential, add- on control requirements, and the
economic aspects of the options. Th e two main sub-indices are a hazard
index and an inherent safety potentia l index. The hazard index measures
the damage potential of the process, considering the process and hazard
control measures. The inherent safe ty potential index addresses the
applicability of inherent safety prin ciples to the process. The two sub-
indices are combined to produce the I2SI value.
The Prototype Index of Inherent Safety (PIIS) for process design was
developed by Edwards and Lawrence (Ref 13.9 Edwards 1993). The PIIS is based on a chemical score and a process score. The chemical score
takes into consideration inventory, flammability, explosiveness and
toxicity. The process score addresses parameters, such as temperature and pressure.
When using an inherent safety in dex, the user should take the
necessary steps to ensure that he/she understands the basis of the index. The developers of the in dices use their own judgment and
experience in deciding what factors are analyzed and in determining the
weighting—sometimes transparentl y, sometimes hidden—of those
factors and how they are combined. Th e user must be sure that these
subjective decisions are in line with their organization’s philosophy and
goals. |
450 PROCESS SAFETY FOR ENGINEERS: AN INTRODUCTION
became an accepted part of every flight and wi th each successful landing the original concerns
seem to have faded away. They loss their sense of vulnerability to a major incident related to
foam failure.
Normalization of deviance - A gradual erosion of standards of
performance as a result of increase d tolerance of nonconformance. (CCPS
Glossary)
In the words of the CAIB report, “Cultural tr aits and organizational practices detrimental
to safety were allowed to develop, including: reliance on past success as a substitute for sound
engineering practices (such as testing to und erstand why systems were not performing in
accordance with requirements); organizational barriers that prevented effective
communication of critical safety information an d stifled professional differences of opinion;
lack of integrated management across program elements; and the evolution of an informal
chain of command and decision-making processes that operated outside the organization’s
rules.” (CAIB 2003)
Lessons
Process Safety Culture. An important aspect of a good safety culture is maintaining a sense
of vulnerability. An example of the poor safety cu lture at NASA is the denial of requests by the
Debris Assessment Team for imaging of the wing while the shuttle was in orbit. The team
concluded, based on modeling that “some loca lized heating damage would most likely occur
during re-entry, but they could not definitively state that structural damage would result.” The
Mission Management Team eventually conclude d the debris strike was a “turnaround” [time
between launches] issue. As stated in the CAIB report “Organizations that deal with high-risk
operations must always have a healthy fear of failure – operations must be proved safe, rather
than the other way around. NASA inverted this burden of proof.”
The CAIB found “NASA ʼs safety culture has become reactive, complacent, and dominated
by unjustified optimism. Over time, slowly and unintentionally, independent checks and
balances intended to increase safety had been eroded in favor of detailed processes that
produce massive amounts of data and unwa rranted consensus, but little effective
communication. Organizations that successfully de al with high-risk technologies create and
sustain a disciplined safety system capable of identifying, analyzing, and controlling hazards
throughout a techno logy’s life cycle.”
Overview
Chapter 21 addresses topics that focus on th e personnel management aspects of process
safety management. This chapter addresses topi cs that focus on the business management
activities used to sustain process safe ty management including the following.
Incident investigation
Measurement and metrics
Auditing
Management review and continuous improvement |
Appendices 173
CRITERIA Yes/No
operations, startup, cleanout , maintenance, and shutdown
configurations)?
Prior PHA Documentation (Section 3.2 and Chapter 8)
Is the PHA documentation sufficient, or can sufficient
documentation be reconstructed to:
• Indicate PHA team meeting dates?
• Verify the five-year history of process safety incidents, and
any others with the potential for catastrophic
consequences, were revi ewed by the PHA team?
• Verify the PSI used by the team was current and adequate
to ensure a thorough study?
• Identify the hazards, engineering and administrative
controls (safeguards), and co nsequences if those risk
controls fail?
• Verify that facility siting was addressed?
• Verify that human factors were addressed?
• Verify that a range of the possible safety and health effects
w a s e v a l u a t e d ( e . g . , b y r i s k r a n k i n g o r s o m e o t h e r
documented technique)?
• Verify compliance with any additional regulatory or
internal requirements?
Note: if only a few of these bullets are an issue, they are probably fixable via an
Update; however, if several of them are an issue, then there are sufficient
deficiencies to warrant a “No” answer. |
164 | 12 REAL Model Scenario: Overfilling
of my buddies, asking if their companies made any upgrades on their tank
gauging equipment. Once I hear back from them, I’ll let you know what their
first-hand experience has been with the upgrades.”
Frederik said, “I should check with my peers at other companies to see
what their experience has been as well.” Alexandre followed up, “We were
really attracted by the lack of mechanical parts in the radar gauges, but since
the initial costs are so high, I still need to crunch some numbers to see if we
can justify the expense.”
Pamela and Frederik smiled broadly. Their employees had done a stellar
job analyzing the situation and making a sound proposal. “Looks like we have
the start of a plan that we can lay out to Jan for feedback and, hopefully,
approval. We just need to flesh out a few more details and we should be good
to go. Let’s plan on getting back together in another week to finalize the plan,”
said Pamela. Before closing the meeting, Alexandre read back the action items,
saying, “Here’s what I have, and please let me know if I missed anything”:
• Review current PHA to see if we need any updates to our emergency
response to flooding. Check on the frequency in which we review PHAs.
(Pamela and Alexandre).
• Follow-up with colleagues about first-hand experience with tank gauge
upgrades. (Frederik and Reed).
• Lifecycle cost analysis for tank gauge upgrade. (Alexandre).
Pamela said, “I think you’ve captured everything. Good meeting.” They left
the meeting excited about the opportunities to improve the safety and
operations of the facility, but also concerned that the costs might be so
prohibitive that they would be stuck with the status quo. Upgrading the gauges
and planning for a flooding event that might never happen could be quite
costly.
12.6 Prepare
Frederik, Pamela, Alexandre, and Reed filed into the meeting room. Reed
started first. “Following up on my action item about talking to my friends, my
buddy Pieter said his company recently upgraded to radar gauges.” Reed went
on, “Like Alexandre mentioned in our previous meeting, we were leaning
toward radar because it has no mechanical parts, but I was concerned about
the complexity of the set-up and how my fellow operators would accept this
new type of equipment. |
9 • Other Transition Time Considerations 178
C9.4.2-3 – Small caustic leak issue upon new refinery start-up [94]
Incident Year : 2012
Cause of the incident occu rring during the initial start-up : During a
longer -than- expected repair time for a leak discovered during a new
refinery start -up, undiluted caustic cont inued to be added to the
crude already charged to and circulating in the partially shut -down
system (a “warm circulation”). Th e undiluted caustic in the system
upon full restart vaporized as the temperatures increased, corroded
thousands of feet of stainless st eel pipe, fouled almost 50 heat
exchangers, and damaged instrumen tation, the distillation tower,
and components in the furnace.
Inciden t impact : Soon after running the refinery at its normal
elevated temperatures and pres sures, a series of quickly-
extinguished fires on the new pipe line occurred, and then a heater
ruptured once crude flow was resumed. Accelerated corrosion
caused significant pipeline and equipment damage and a
subsequent significant delay in and cost of the refinery start -up.
Risk management system weaknesses:
LL1) Note: No formal incident investiga tion report has been made
publically available.
The following issues may have cont ributed to this incident: 1) the
unanticipated delay in fixing the a le ak (this was not recognized as a
change in the planned start -up); and 2) the unanticipated effect of
continuing to add caustic to the small amount of crude still in the
unit during the warm circulation (either from a failed valve on the
caustic system or by not shutting th e caustic addition system off).
Relevant RBPS Elements:
Process Knowledge Management
Hazard Identification and Risk Analysis
Management of Change
Operational Readiness
|
276
• Are all tripping hazards minimized and all walking
surfaces tractional during all weather conditions?
• Are low noise equipment and machinery taken into consideration when ma king new purchases?
• Is shift rotation optimized to avoid fatigue?
• Are awkward positions and repetitive motions
minimized?
• Are attempts made to completely eliminate raw materials, process intermediates or by-products?
• Are elbows, bends, and joints in piping minimized?
Substitute • Can a less toxic, flam mable or reactive material be
substituted for use?
• Is there an alternate way of moving product or equipment as to eliminate human strain?
• Can a water-based product be used in place of a
solvent or oil-based product?
• Are all allergenic materials, products and equipment replaced with non- allergen ic materials, products and
equipment when possible?
Moderate • Can potential rele ases be reduced via lower
temperatures or pressures, or elimination of equipment?
• Are all hazardous gases, liquids and solids stored as far away as possible to eliminate disruption to people, property, production and environment in the event of an incident? |
Selecting an Appropriate PHA Revalidation Approach 91
(1) this worksheet alone or (2) the work sheet along with PHA documentation that
has been revised per the changes and incidents serves as the Update to the PHA.
Using only the change and incident review worksheet is sometimes referred to
as a “Focused” or “Simple” Update .
Regardless of the Update approach, the amount of time needed to perform
the Update will depend upon the extent and number of changes made to the
process since the prior PHA, and the number and significance of incidents that
have occurred.
Revalidation teams may encounter situ ations where the prior PHA was of
very high quality (i.e., no apparent gaps or deficiencies) and where there have
been no significant incidents or changes within the subject process. Such
circumstances may permit a much simpler revalidation effort and report, limited
to affirming the continued validity of the prior PHA. When is an Update by “Change an d Incident Review ” Appropriate?
Facilities should thoroughly evaluate the prior PHA (as discussed in
Chapter 3) before initiating an Update by Change and Incident Review
(Focused Update ). Because many scenarios documented by the previous
team are deemed unchanged and are no t being reviewed, (1) the prior PHA
must be of high quality and complete and (2) every change and relevant
incident must be thoroughly documented and available for the revalidation
team to review.
For example, if a process subject to PHA revalidation is an ammonia tank
only, with one pertinent MOC, one minor incident, and a high-quality prior
PHA, Updating the PHA by Change and Incident review can result in a
complete, up-to-date, and adequate PHA.
On the other hand, if a process is complex with multiple changes and several
incidents, using the Change and Incident Review alone may not produce a
PHA that accurately describes the cu rrent process risks or may produce a
document that will be difficult to revalidate in the future.
Performing a Focused Update by Change and Incident Review on an
incomplete or otherwise unacceptable PHA can only result in an incomplete
or unacceptable PHA revalidation. |
6.2 Assess the Organization’s Pr ocess Safety Culture |215
the interview. Helping the interviewee to clarify and/or deepen
his/her responses communicates respect and interest.
Probe constructively. This m ay be needed when
interviewees provide inconsistent, conflicting, or incomplete
responses. Interviewers should phrase inquiries to focus on the
data rather than confronting or criticizing the respondent. If
possible, the conflict should not even be mentioned. Instead draw
the interviewee into the process to clarify the inform ation.
When probing suspected negative behaviors, avoid negative
and potentially accusatory questions such as, “Do you make
unauthorized changes in the plant without using MOC?” Instead,
pose a scenario and observe the response. To probe
unauthorized changes, the interviewer could ask, “It is 2:00 AM
Saturday morning. A part needs to be replaced but the
replacement-in-kind part is not available. What would you do?”
The verbal and non-verbal responses should reveal the true
situation.
Confirm input. The interviewer should sum marize or
paraphrase the inform ation learned frequently during the
interview. Called active listening, it involves paraphrasing answers
in the form of closed questions. Active listening clarifies the
interviewee’s response, while showing interest in understanding
the response accurately.
Watch non-verbal signals. As the saying goes, only 10
percent conversations are verbal; the other 90 percent is tone and
body language. Answers that appear inconsistent with body
language or tone, and sudden changes in either may signal that
the interviewer is getting close to sensitive topics (Ref 6.3)
Provide feedback, as appropriate. The interviewee may
request feedback at various stages in the interview process.
B ecause policies may vary from company to com pany regarding
m aking recomm endations and suggestions directly to facility
personnel, interviewer should understand those policies prior to |
68 PROCESS SAFETY IN UPSTREAM OIL & GAS
Incident: Snorre A Blowout, No rway North Sea, November 2004
Snorre A was a large integrated tension leg platform with processing, drilling
and accommodation modules. Activity levels were high with SIMOPS covering
production, drilling and well intervention underway. During a workover
operation prior to further drilling in a well, a gas blowout occurred on the seabed
with the subsequent gas flowing to th e surface and under the facility. Ignition
did not occur. There were 216 persons on board at the time of the incident, 181
of whom were evacuated to other installations while the other 35 persons
remained on Snorre A to carry out emergency response and well control tasks.
The gas blowout was stopped and the well brought under control the day after.
No one was injured in connection with the incident.
Process Safety Issues : Complex defects in the well due to corrosion and other
factors were not sufficiently managed. Shore-based HAZOPs to address the
problems being encountered were carried out but not communicated to offshore
personnel. The Petroleum Safety Authority (PSA) identified multiple safety
barriers that failed, and these allowed the incident to occur. The PSA concluded
that total loss of the facility was possibl e and that this serious near miss was one
of the worst events in Norway.
Source: PSA, 2005
RBPS Application
Process Safety Culture : Multiple organizational issues were identified including
too slow integration of Snorre into the Statoil organization following the
acquisition of Saga Petroleum, critical questioning of operations was not
welcomed, and management was not sufficiently engaged. RBPS suggests how
to enhance process safety culture.
Asset Integrity and Reliability : Offshore personnel allowed the BOP to be partly
disabled as only the annular preventer was available. RBPS offers guidance on
means to ensure full availability of critical barriers.
Key Process Safety Measure(s)
Conduct of Operations: Significant planning and engineering is required to work in
an HPHT environment, including the specification and use of equipment and drilling
and completion fluid and cement specification. Personnel should be trained in HPHT
operations. |
310 INVESTIGATING PROCESS SAFETY INCIDENTS
Start writing portions the report as soon as the investigation begins.
Focusing on the result can help keep the team focused on the investigation
process and the product.
13.6 INVESTIGATION DOCUM ENT AND EVIDENCE RETENTION
The investigation team’s work often ends with the approval and distribution
of the report and recommendations. Once the investigation team disbands,
investigation records may be lost over time or destroyed in accordance with
company retention policy. Some jurisdictions require that incident reports and other documents be retained including drafts , all documents reviewed
during the investigation and emails pertaining to the incident report. Litigation may impose other record retention requirements. Consult with the company’s legal representative to determine the record retention
requirements.
Investigation record retention may differ from normal company record
retention policies. The report an d its associated linked and referenced
documents can be an issue. If the do cuments are not cat egorized and stored
properly, corporate record retention systems can delete them. If links are used, and files are moved, the links ca n be broken. Investigation documents
may have to be compiled and stored in a location that is protected from
automated deletion.
Physical and electronic evidence may also have to be retained,
sometimes for years due to litigation. Longer term evidence preservation
and storage should be ar ranged. Items that are weather or temperature
sensitive should be stored in an environmentally controlled room or building. Chemical samples and fracture surfaces pose challenges due to aging in
storage, even in environmentally contro lled conditions. Performing analyses
while evidence is fresh and producing good documentation is often the best
approach when long term degradation is unavoidable. The documentation
should be retained for the dura tion of the legal proceedings.
Corporate counsel and management will ultimately decide when certain
investigation materials and evidence ma y be discarded. Some materials may
be retained permanently, such as the incident investigation report and the documentation of resolution of the action points.
|
168 Human Factors Handbook
Figure 15-3: Working nights
15.3 Managing fatigue risk
15.3.1 Fatigue risk policy
A formal fatigue risk management policy and set of arrangements should be in
operation. Typical parts of a fatigue risk policy are shown in Figure 15-4. This
should include a commitment to manage fatigue and satisfy national and local
laws and regulations as well as guidance such as the “IOGP Report 626 – Managing
fatigue in the workplace” [61]. This should include training for all key roles on how
to prevent fatigue, how to recognize fatigue and deal with it and an overview of
the company fatigue risk management program.
Further guidance on fatigue risk management is also available from the Energy
Institute [62].
The policy on maximum working hours and rest breaks should take account of
the physical and mental demands of tasks. More demanding tasks require more
rest. In addition, it should control people volunteering for over time. The policy on
the maximum hours worked should also limit permitted voluntary over time and
avoid a small number of workers taking on excessive hours.
|
30 GUIDELINES FOR MANAGING ABNORMAL SITUATIONS
3.1.2 Additional Focus Areas
Other key areas that require consideration for the management of
abnormal situations include:
3.1.2.1 Instrument Failures
Potential failure modes should be id entified and appropriate diagnostic/
troubleshooting skills, tool s, and techniques, along with associated training
should be developed for operators to handle situations including:
Valve Failures: A faulty or failed automatic valve can contribute to
abnormal situations. Additionally, the possibility of an automatic
valve not failing to a safe position during an electrical or pneumatic
supply loss should be discussed and understood by the plant
personnel.
Sensor Failures: A faulty or fa iled sensor can contribute to
abnormal situations. In addition, operators should be aware of
potential situations when sensors and instruments are off-line for
calibration or repair.
I/O Card failures: Input and Output card failures in the process
control network can occur and are of ten difficult to quickly diagnose
and address.
Bypassing alarms and trips: There are occasions when it may be
necessary to override such system s temporarily, for example, when
a sensor fails. If not properly managed, however, this can either be
the direct cause of an abnormal situation or remove a layer of
defense if an abnormal situation arises from another cause. Good
PSM systems include a rigorous management procedure to control
alarm suppression and interlock bypassing, which is further
discussed in Chapter 5.
Further details are provided in Chapter 4, Education for Managing
Abnormal Situations.
3.1.2.2 Services Failure Including Power Blackout
Service failure includes situations in which one or more of the services
(including electricity, steam, air, water, inert gas) is lost due to an outage
or other unforeseen reason. During a power blackout, although an
uninterruptible power supply (UPS) us ually allows the DCS to continue |
Edward’s pipeline company John’s plant company
Figure 4.23 A Batt ery Limit P&ID.
Figure 4.24 A Utilit y Distribution P&ID. |
Chemical Hazards Data Sources
Learning Objectives
The learning objective of this chapter is:
Identify sources of chemical hazards data and understand the data provided.
Incident: Concept Sciences Explosion, Allentown, Pennsylvania, 1999
Incident Summary
“At 8:14 pm on February 19, 1999, a process vessel containing several hundred pounds of
hydroxylamine (HA) exploded at the Concept Sc iences, Inc. (CSI), production facility near
Allentown, Pennsylvania. Employees were distilling an aqueous solution of HA and potassium
sulfate, the first commercial batch to be proce ssed at CSI’s new facility. After the distillation
process was shut down, the HA in the proce ss tank and associated piping explosively
decomposed, most likely due to high concentration and temperature.
Four fatalities resulted, including CSI employ ees and a manager of an adjacent business.
Two CSI employees survived the blast with modera te-to-serious injuries. Four people in nearby
buildings were injured. Six firefighters and two security guards suffered minor injuries during
emergency response efforts.
The production facility was extensively damage d (Figure 7.1). The explosion also caused
significant damage to other buildings in the Lehigh Valley Industrial Park and shattered
windows in several nearby homes.” (CSB 2002)
Key Point:
Hazard Identification and Risk Analysis - Hazard review methodologies
need to be appropriate to the haza rds being managed. A high hazard
warrants a detailed review.
Detailed Description
Pure HA is a compound with the formula NH 2OH. Solid HA consists of co lorless or white crystals
that are unstable and susceptible to explosive decomposition and explodes when heated in
air above 70 °C (158 °F). HA is usually sold as a 50 wt. % or less solution in water. The Chemical
Safety Board Investigation report quoted CSI’s safety data sheet (SDS) as stating “Danger of fire
and explosion exists as water is removed or evaporated and HA concentration approaches
levels in excess of about 70%”. HA can be ig nited by contact with metals and oxidants. |
APPENDIX D – REACTIVE CHEMICALS CHECKLIST 483
or lower (for systems being cooled) than th e bulk mixture temperature. For exothermic
reactions, the temperature may also be higher near the point of introduction of reactants
because of poor mixing and localized reaction at the point of reactant contact. The location
of the reactor temperature sensor relative to the agitator, and to heating and cooling
surfaces may impact its ability to provide good information about the actual average
reactor temperature. These problems will be more severe for very viscous systems, or if
the reaction mixture includes solids which can foul temperature measurement devices or
heat transfer surfaces. Either a local high temperature or a local low temperature could
cause a problem. A high temperature, for exampl e, near a heating surface, could result in
a different chemical reaction or decomposition at the higher temperature. A low
temperature near a cooling coil could result in slower reaction and a buildup of unreacted
material, increasing the potential chemical ener gy of reaction available in the reactor. If
this material is subsequently reacted becaus e of an increase in temperature or other
change in reactor conditions, an uncontrolled reaction is possible due to the unexpectedly
high quantity of unreacted material available.
11. Understand the rate of all chemical reactions.
It is not necessary to develop complete ki netic models with rate constants and other
details, but you should understand how fast reactants are consumed and generally how
the rate of reaction increases with temperat ure. Thermal hazard calorimetry testing can
provide useful kinetic data.
12. Consider possible vapor phase reactions.
These might include combustion reactions, other vapor phase reactions such as the
reaction of organic vapors with a chlorine atmosphere, and vapor phase decomposition
of materials such as ethylene oxide or organic peroxide.
13. Understand the hazards of the products of both intended and unintended reactions.
For example, does the intended reaction, or a possible unintended reaction, form viscous
materials, solids, gases, corrosive products, highly toxic products, or materials which will
swell or degrade gaskets, pipe linings, or ot her polymer components of a system? If you
find an unexpected material in reaction equipm ent, determine what it is and what impact
it might have on system hazards. For example, in an oxidation reactor, solids were known
to be present, but nobody knew what they were. It turned out that the solids were
pyrophoric, and they caused a fire in the reactor.
14. Consider doing a Chemical Interaction Matrix and/or a Chemistry Hazard Analysis.
These techniques can be applied at any stage in the process life cycle, from early research
to an operating plant. They are intended to provide a systematic method to identify
chemical interaction hazards and hazards re sulting from deviations from intended
operating conditions.
D.2 Reaction Process De sign Considerations
1. Rapid reactions are desirable. |
Appendix 212
Timened*
downsExtended*
Shutdowns
or
TurnaroundsStart-up
afterwardsNormal or
Abnormal OperationsShut-down
(depends on situation)Start-up
afterwards
Start-up after an
unscheduled or emergency
shutdown period
Transient
Operating ModeTransient
Operating ModeStart-up after a
planned or extended
shutdown periodResulting in:
Recovery
Unscheduled Shutdown
Emergency ShutdownNormal Operations
Abnormal Operations
Transient
Operating Mode
Incident Incident Incident
Figure A.2-1Timeline of when incidents occurre d during the transient oper ating mode (continued). |
20 PROCESS SAFETY IN UPSTREAM OIL & GAS
Figure 2-8. Typical types of production installations in use on the deepwater
outer continental shelf (OCS)
Courtesy of BOEM
Figure 2-9. Typical types of production installations in use on the outer
continental shelf (OCS)
Courtesy of BOEM
|
E.36 Operating Blind |325
the aluminum flame arrestor had corroded to the point it no
longer functioned and the plastic tank could not withstand the
pressure and stresses of the internal and external fire.
The investigators further discovered that the facility did not
have a perm it-to-work system, that it was seriously overdue on
equipment inspection, and that its frequency of safety training
had been steadily decreasing over the prior eight years. B ased on
interviews, the last training involving methanol hazards had
occurred twelve years earlier.
In its investigation report, the CSB m ade recomm endations to
regulatory agencies, standards organizations, and the
engineering com pany that installed the m ethanol system.
However, it made no recommendations to the facility, which is
ultimately responsible for worker and process safety. Which
culture factors could CSB have explored in this investigation?
Did facility managem ent and workers understand the hazards
and risks of its processes? What caused the decrease in training
frequency? Was the imperative for safety weakening?
High consequence scenarios related to the intended project
are easy to imagine. Hot cuttings could partially or com pletely
m elt through the plastic roof of the tank or piping. Methanol
venting from the tank as the sun heats it could ignite from cutting
sparks. A cut-off roof section could be dropped edge-on and slice
through the tank or piping. What caused workers and
m anagem ent to not think about any of this? Or if they thought
about it, what caused them to not act to protect against these
seem ingly likely deviations?
Maintain a Sense of Vulnerability, Understand and Act Upon
Hazards/Risks.
E.36 Operating Blind
A worker was lining up valves to transfer kerosene
and gasoline from one terminal to a neighboring Actual
Case
History |
KEY RELEVANCE TO PROCESS PLANT OPERATIONS 53
3.3.5.6 Measurement and Metrics
Leading and lagging indicators of pr ocess safety performance, including
incident and near-miss rates as well as metrics that show how well key
process safety elements are being pe rformed. This information is used
to drive improvement in Process Safe ty. Safe Operating Limit Excursions
and Demand on Safety Systems ar e two example metrics that are
relevant to abnormal situations. Me trics will be further discussed in
Chapters 5 and 6.
3.3.5.7 Management Review and Continuous Improvement
The practice of managers at all levels of setting process safety
expectations and goals with their staff and reviewing performance and
progress towards those goals. This may take place in a staff or
“leadership team” meeting or individua lly. The practice may be facilitated
by the process safety lead bu t is owned by the line manager.
3.4 PROCEDURES AND OPERATING MODES FOR MANAGING
ABNORMAL SITUATIONS
This section addresses how to writ e and structure procedures that
incorporate principles describing how to manage abnormal situations that
can then be used by operating pers onnel to make appropriate decisions
during periods of abnormal situations. However, many abnormal situations
will not be anticipated, and for those, a more holistic approach to abnormal
situation manageme nt is required, which in cludes not only written
procedures, but also training personnel to recognize an abnormal situation,
protocols for dealing with it, and providing the resources to respond to
events that may not be foreseen.
3.4.1 General Principles for Procedure Development
It is outside the scope of this book to provide detailed guidance, or
templates, for development of normal operating procedures, however,
these can be found in ot her references including Guidelines for Writing
Effective Operating and Maintenance Procedures (CCPS 1996). However, for
managing and controlling abnormal si tuations, some high-level human
behavior principles apply, as summariz ed in the sources discussed in this
section. |