instruction stringlengths 17 119 | input stringclasses 2
values | output stringlengths 7 321 | bucket stringclasses 6
values | source stringclasses 5
values | reviewed bool 1
class |
|---|---|---|---|---|---|
What is Step 1 of MDMP? | Step 1 is Receipt of Mission. The staff is alerted, gathers planning tools, and receives the commander's initial planning guidance to begin the process. | mdmp_steps | mdmp_steps.md | true | |
What happens during Mission Analysis? | Mission Analysis is Step 2. The staff analyze the higher headquarters order, develop a running estimate, define the problem, draft the commander's intent, develop CCIRs, and develop evaluation criteria for later COA comparison. | mdmp_steps | mdmp_steps.md | true | |
What is COA Development? | COA Development is Step 3. The staff generate courses of action and screen them to ensure they are feasible, acceptable, suitable, distinguishable, and complete before war gaming. | mdmp_steps | mdmp_steps.md | true | |
Which MDMP step is war gaming? | War gaming is Step 4, COA Analysis. Staff visualize each COA through critical events and record strengths and weaknesses without head-to-head comparison of COAs. | mdmp_steps | mdmp_steps.md | true | |
When do staff use a decision matrix? | Staff use a decision matrix during Step 5, COA Comparison, to compare COAs against the evaluation criteria developed during mission analysis. | mdmp_steps | mdmp_steps.md | true | |
What is Step 6 of MDMP? | Step 6 is COA Approval. The commander selects or modifies the recommended course of action. | mdmp_steps | mdmp_steps.md | true | |
What is Orders Production? | Orders Production is Step 7. The staff produce the OPORD and supporting products to execute the approved COA. | mdmp_steps | mdmp_steps.md | true | |
List the MDMP steps in order. | 1 Receipt of Mission, 2 Mission Analysis, 3 COA Development, 4 COA Analysis (War Gaming), 5 COA Comparison, 6 COA Approval, 7 Orders Production. | mdmp_steps | mdmp_steps.md | true | |
What is the commander's role in MDMP? | The commander drives the process, provides planning guidance, approves evaluation criteria, selects the COA at Step 6, and issues intent that guides all steps. | mdmp_steps | mdmp_steps.md | true | |
When are evaluation criteria first developed? | Evaluation criteria are developed during Mission Analysis (Step 2) and may be refined after war gaming. They should not be invented for the first time at COA comparison. | mdmp_steps | mdmp_steps.md | true | |
What is Step 3 of MDMP called? | Step 3 is COA Development — generating and screening courses of action. | mdmp_steps | mdmp_steps.md | true | |
What step follows war gaming? | COA Comparison (Step 5) follows war gaming (Step 4). | mdmp_steps | mdmp_steps.md | true | |
What step precedes Orders Production? | COA Approval (Step 6) precedes Orders Production (Step 7). | mdmp_steps | mdmp_steps.md | true | |
Is Receipt of Mission the same as Mission Analysis? | No. Receipt of Mission is Step 1 — alerting the staff and beginning planning. Mission Analysis is Step 2 — analyzing the problem and developing criteria and intent. | mdmp_steps | mdmp_steps.md | true | |
What is COA Analysis also called? | COA Analysis is also called war gaming — MDMP Step 4. | mdmp_steps | mdmp_steps.md | true | |
At what step does the commander approve a COA? | The commander approves the COA at Step 6, COA Approval. | mdmp_steps | mdmp_steps.md | true | |
What is produced in Step 7? | Step 7 produces the OPORD and supporting orders products. | mdmp_steps | mdmp_steps.md | true | |
How many steps are in MDMP? | MDMP has seven steps, from Receipt of Mission through Orders Production. | mdmp_steps | mdmp_steps.md | true | |
What is the purpose of Step 2? | Step 2, Mission Analysis, defines the problem, develops running estimates, drafts intent, develops CCIRs, and establishes evaluation criteria. | mdmp_steps | mdmp_steps.md | true | |
What is the purpose of Step 3? | Step 3, COA Development, creates viable courses of action and screens them before war gaming. | mdmp_steps | mdmp_steps.md | true | |
Can we compare COAs during the war game? | No. During Step 4 war gaming, staff analyze each COA separately and record strengths and weaknesses. Head-to-head COA comparison belongs in Step 5. | step_boundaries | mdmp_steps.md | true | |
What's the difference between Step 4 and Step 5? | Step 4 war games each COA individually. Step 5 compares all COAs against evaluation criteria using a decision matrix or similar tool. | step_boundaries | mdmp_steps.md | true | |
When should staff rank COAs? | Staff rank or compare COAs systematically in Step 5, not during Step 4 war gaming. | step_boundaries | war_gaming.md | true | |
Is COA Comparison the same as war gaming? | No. War gaming is Step 4 (COA Analysis). COA Comparison is Step 5, where criteria are applied across COAs. | step_boundaries | mdmp_steps.md | true | |
Can we pick a winning COA at the end of Step 4? | The commander selects the COA at Step 6. Step 4 produces analysis; Step 5 produces a comparison recommendation. | step_boundaries | mdmp_steps.md | true | |
What mistake do staffs make between Steps 4 and 5? | A common mistake is comparing or ranking COAs during the war game instead of fighting each COA separately and saving comparison for Step 5. | step_boundaries | war_gaming.md | true | |
Does war gaming replace COA comparison? | No. War gaming (Step 4) informs but does not replace COA comparison (Step 5). | step_boundaries | war_gaming.md | true | |
When do numeric criterion scores get applied across COAs? | Numeric scores are applied in the Step 5 decision matrix, using inputs derived from Step 4 war-game findings. | step_boundaries | war_gaming.md | true | |
Is Step 5 before or after the commander approves a COA? | Step 5 (COA Comparison) is before Step 6 (COA Approval). The commander approves after comparison. | step_boundaries | mdmp_steps.md | true | |
Define course of action. | A course of action (COA) is a broad potential solution that accomplishes the mission, including scheme of maneuver, main effort, supporting efforts, and key tasks. | glossary | glossary.md | true | |
Describe commander's intent in MDMP planning. | Commander's intent is a concise expression of the operation's purpose and desired end state, guiding initiative when the plan no longer fits the situation. | glossary | glossary.md | true | |
What is a decision matrix? | A decision matrix is a tabular comparison of COAs against evaluation criteria, used in Step 5 to support the commander's decision. | glossary | glossary.md | true | |
What is an OPORD? | An OPORD is an operations order issued after COA approval during Orders Production (Step 7). | glossary | glossary.md | true | |
What does CCIR stand for? | CCIR stands for commander's critical information requirement — information the commander needs for timely decisions. | glossary | glossary.md | true | |
What is a running estimate? | A running estimate is a continuous assessment of facts and assumptions affecting the mission, updated throughout MDMP. | glossary | glossary.md | true | |
What is the main effort? | The main effort is the unit or axis assigned the decisive task and given priority of support. | glossary | glossary.md | true | |
What tool aligns actions and reactions during war gaming? | A synchronization matrix aligns friendly actions, reactions, and counteractions across time and space during war gaming. | glossary | glossary.md | true | |
Define COA comparison. | COA comparison is MDMP Step 5 — systematic comparison of courses of action using evaluation criteria after war gaming. | glossary | glossary.md | true | |
Summarize mission analysis for a staff section. | Mission analysis is MDMP Step 2 — understanding the problem, defining tasks, developing criteria, and drafting intent. | glossary | glossary.md | true | |
For OBJ FALCON, how should staff compare COA-A Direct Assault and COA-B Double Envelopment on simplicity vs disruption? | Fictional 3rd BCT river crossing scenario. | COA-A Direct Assault typically scores higher on simplicity with a single main axis. COA-B Double Envelopment may achieve greater enemy disruption through flanking but at lower simplicity due to coordination between fixing and enveloping forces. Compare both in Step 5 against the Simplicity and Enemy Disruption criteria... | scenario_coaching | obj_falcon_fictional.md | true |
Which fictional COA prioritizes speed to objective in the OBJ FALCON scenario? | COA-C Air Assault Link-up typically offers the highest speed to objective if the air assault secures the far bank and ground link-up occurs within 24 hours, but it carries the highest synchronization risk. | scenario_coaching | obj_falcon_fictional.md | true | |
Why might COA-E Turning Movement score lower on mission accomplishment for OBJ FALCON? | COA-E bypasses OBJ FALCON to seize the enemy supply hub. It may disrupt the enemy strongly but can score lower on direct mission accomplishment against seizing the river crossing itself. | scenario_coaching | obj_falcon_fictional.md | true | |
What war-gaming method fits a focused fight at a river crossing? | The box method is often appropriate when focusing on a defined objective area such as a river crossing and town center. | war_gaming | war_gaming.md | true | |
What are the three war-gaming methods? | Belt, avenue-in-depth, and box methods. | war_gaming | war_gaming.md | true | |
What does feasible mean in COA screening? | Feasible means the COA can be accomplished with available resources and time. | coa_screening | coa_screening.md | true | |
What does distinguishable mean for a COA? | Distinguishable means the COA is meaningfully different from other options, not a cosmetic variant. | coa_screening | coa_screening.md | true | |
When should COAs be screened with FASDC? | During COA Development (Step 3), before war gaming in Step 4. | coa_screening | coa_screening.md | true | |
How does COA-D Feint and Main trade off in the fictional scenario? | COA-D uses deception for high enemy disruption and good mission outlook, but feint timing adds synchronization risk and moderate simplicity. | scenario_coaching | obj_falcon_fictional.md | true | |
Which MDMP step is Receipt of Mission? | Receipt of Mission is Step 1. The staff is alerted, gathers planning tools, and receives the commander's initial planning guidance. | mdmp_steps | mdmp_steps.md | true | |
What occurs during Step 1 of MDMP? | Step 1, Receipt of Mission, alerts the staff, gathers planning tools, and issues initial planning guidance to begin MDMP. | mdmp_steps | mdmp_steps.md | true | |
What is Step 4 of MDMP called? | Step 4 is COA Analysis, also called war gaming. Staff visualize each COA through critical events. | mdmp_steps | mdmp_steps.md | true | |
Identify Step 5 of the military decision-making process. | Step 5 is COA Comparison — comparing all COAs against evaluation criteria using a decision matrix or similar tool. | mdmp_steps | mdmp_steps.md | true | |
During which MDMP step do staff generate courses of action? | Staff generate courses of action during Step 3, COA Development, and screen them before war gaming. | mdmp_steps | mdmp_steps.md | true | |
What MDMP step follows COA Comparison? | COA Approval (Step 6) follows COA Comparison (Step 5). | mdmp_steps | mdmp_steps.md | true | |
What step comes immediately before COA Development? | Mission Analysis (Step 2) comes before COA Development (Step 3). | mdmp_steps | mdmp_steps.md | true | |
In which step does the staff issue the OPORD? | The staff produce the OPORD during Step 7, Orders Production, after the commander approves a COA at Step 6. | mdmp_steps | mdmp_steps.md | true | |
What is the first step of the military decision-making process? | The first step is Receipt of Mission (Step 1). | mdmp_steps | mdmp_steps.md | true | |
What is the final step of MDMP? | Orders Production (Step 7) is the final step — producing the OPORD and supporting products. | mdmp_steps | mdmp_steps.md | true | |
When do staff develop initial CCIRs? | Staff develop initial CCIRs during Mission Analysis (Step 2). | mdmp_steps | mdmp_steps.md | true | |
What is the Army's iterative staff planning methodology? | The Military Decision-Making Process (MDMP) per FM 5-0 / ADP 5-0 — seven steps from Receipt of Mission through Orders Production. | mdmp_steps | mdmp_steps.md | true | |
When does COA screening occur in MDMP? | COA screening occurs during COA Development (Step 3), before war gaming in Step 4. | mdmp_steps | mdmp_steps.md | true | |
What is Step 2 of MDMP? | Step 2 is Mission Analysis — analyzing the higher HQ order, developing running estimates, drafting intent, CCIRs, and evaluation criteria. | mdmp_steps | mdmp_steps.md | true | |
May staff use a decision matrix during Step 4? | No. War gaming in Step 4 records findings per COA. The decision matrix is used in Step 5 for systematic comparison against criteria. | step_boundaries | mdmp_steps.md | true | |
Is head-to-head COA comparison allowed during war gaming? | No. Step 4 fights each COA separately. Head-to-head comparison against criteria belongs in Step 5. | step_boundaries | war_gaming.md | true | |
What happens in Step 5 that does not happen in Step 4? | Step 5 compares all COAs against evaluation criteria systematically. Step 4 analyzes each COA individually without ranking or head-to-head comparison. | step_boundaries | mdmp_steps.md | true | |
Can the commander approve a COA during COA Comparison? | No. COA Comparison (Step 5) produces a recommendation. The commander selects or modifies the COA at Step 6, COA Approval. | step_boundaries | mdmp_steps.md | true | |
Should evaluation criteria first appear at Step 5? | No. Evaluation criteria are developed during Mission Analysis (Step 2) and refined after war gaming. They should not be invented for the first time at Step 5. | step_boundaries | coa_screening.md | true | |
What is wrong with ranking COAs while fighting them in the war game? | Ranking during Step 4 blurs the process and biases narratives before criteria are applied systematically in Step 5. | step_boundaries | war_gaming.md | true | |
Does Step 4 replace the need for a decision matrix? | No. War gaming (Step 4) informs the decision matrix but does not replace COA Comparison (Step 5). | step_boundaries | war_gaming.md | true | |
When is the final COA selected? | The commander selects or modifies the final COA at Step 6, COA Approval, after staff comparison in Step 5. | step_boundaries | mdmp_steps.md | true | |
What feeds the Step 5 decision matrix? | War-game findings from Step 4 — strengths, weaknesses, synchronization issues, and qualitative or numeric assessments per COA. | step_boundaries | war_gaming.md | true | |
Can staff compare COAs against criteria during Step 4? | No. Staff record per-COA analysis in Step 4. Systematic comparison against evaluation criteria is Step 5. | step_boundaries | mdmp_steps.md | true | |
What step uses evaluation criteria systematically across all COAs? | COA Comparison (Step 5) applies evaluation criteria across all COAs using a decision matrix or brief. | step_boundaries | mdmp_steps.md | true | |
Is war gaming the same as COA approval? | No. War gaming is Step 4 (COA Analysis). The commander approves a COA at Step 6. | step_boundaries | mdmp_steps.md | true | |
What MDMP step immediately follows COA Analysis? | COA Comparison (Step 5) follows COA Analysis / war gaming (Step 4). | step_boundaries | mdmp_steps.md | true | |
What step precedes COA Comparison? | COA Analysis / war gaming (Step 4) precedes COA Comparison (Step 5). | step_boundaries | mdmp_steps.md | true | |
Where does systematic COA comparison belong in MDMP? | Systematic COA comparison belongs in Step 5, after war gaming is complete. | step_boundaries | mdmp_steps.md | true | |
What is action-reaction-counteraction? | Action-reaction-counteraction is the sequence recorded in the synchronization matrix during war gaming to visualize how friendly actions, enemy reactions, and friendly counteractions unfold. | war_gaming | war_gaming.md | true | |
How is a synchronization matrix used in war gaming? | The synchronization matrix aligns friendly actions, enemy reactions, and counteractions across time and space during Step 4 war gaming. | war_gaming | war_gaming.md | true | |
When should staff use the belt war-gaming method? | Use the belt method to fight the battlefield in sequential belts, often east-west or along operational phases. | war_gaming | war_gaming.md | true | |
When is the avenue-in-depth method appropriate? | Avenue-in-depth follows a single avenue of approach through critical events — useful when the fight is dominated by one axis. | war_gaming | war_gaming.md | true | |
When should staff choose the box method for war gaming? | The box method focuses on a defined area — an objective, engagement area, or choke point such as a river crossing. | war_gaming | war_gaming.md | true | |
How should staff fight COAs during the war game? | Fight one COA at a time (or one friendly COA against one enemy COA in sequence). Do not compare or rank COAs during the war game. | war_gaming | war_gaming.md | true | |
What are typical war-game outputs? | A synchronization matrix, refined COAs, updated evaluation criteria, and a war-game workbook or narrative that feeds the COA comparison brief. | war_gaming | war_gaming.md | true | |
What do staff record per COA during Step 4? | Strengths, weaknesses, advantages, disadvantages, synchronization risk, likely enemy reactions, and refinements for each COA. | war_gaming | war_gaming.md | true | |
Can war gaming refine evaluation criteria? | Yes. Staff refine COAs and evaluation criteria based on war-game findings before Step 5 comparison. | war_gaming | war_gaming.md | true | |
What is COA Analysis in MDMP? | COA Analysis is Step 4 — war gaming. Staff visualize each COA through critical events without head-to-head COA comparison. | war_gaming | war_gaming.md | true | |
Is war gaming Step 4 or Step 5? | War gaming is Step 4 (COA Analysis). Step 5 is COA Comparison. | war_gaming | war_gaming.md | true | |
What is the purpose of war gaming? | To visualize each COA through critical events, identify strengths and weaknesses, capture synchronization issues, and feed Step 5 comparison. | war_gaming | war_gaming.md | true | |
Should staff compare two friendly COAs side by side during the war game? | No. Staff analyze each COA separately in Step 4. Side-by-side comparison belongs in Step 5. | war_gaming | war_gaming.md | true | |
What happens to invalid assumptions during a war game? | Staff update assumptions when the war game reveals invalid premises. | war_gaming | war_gaming.md | true | |
What are decision points in war gaming? | Points where the commander may need to decide or refine the plan based on how the fight unfolds during visualization. | war_gaming | war_gaming.md | true | |
Describe the belt war-gaming method. | The belt method fights the battlefield in sequential belts, progressing through phases or geographic slices. | war_gaming | war_gaming.md | true | |
Describe the avenue-in-depth war-gaming method. | Avenue-in-depth follows one avenue of approach through critical events from start to objective. | war_gaming | war_gaming.md | true | |
What should staff list before starting a war game? | Friendly forces, assumptions, critical events, decision points, and the war-gaming method to use. | war_gaming | war_gaming.md | true | |
How do war-game findings support Step 5? | Findings become inputs to the decision matrix — narrative assessments or numeric scores per criterion and COA. | war_gaming | war_gaming.md | true | |
Does war gaming include OPORD production? | No. War gaming is Step 4. OPORD production is Step 7 after COA approval. | war_gaming | war_gaming.md | true | |
What visualization happens in MDMP Step 4? | Staff visualize each COA through critical events using action-reaction-counteraction and record results per COA. | war_gaming | war_gaming.md | true | |
Can numeric scores be derived from war gaming? | Yes. Staff may translate qualitative war-game observations into numeric inputs for the Step 5 decision matrix. | war_gaming | war_gaming.md | true | |
What is the difference between war gaming and COA comparison? | War gaming (Step 4) analyzes each COA separately. COA comparison (Step 5) compares all COAs against evaluation criteria. | war_gaming | war_gaming.md | true |
End of preview. Expand in Data Studio
mdmp-staff-planning-pairs
Leak-reviewed instruction-tuning pairs for MDMP staff-planning coaching. Public doctrine summaries and fictional scenarios only — no proprietary algorithms, customer data, or classified content.
Disclaimer: Unofficial educational dataset. Not affiliated with the U.S. Army.
Dataset description
324 human-reviewed {instruction, input, output} pairs for fine-tuning a Mistral-7B instruct model on Military Decision-Making Process vocabulary, step boundaries, and coaching responses.
- GitHub source: dlens/mdmp-assistant
- Model trained on this data: mistral7b-mdmp-lora
Fields
| Field | Description |
|---|---|
instruction |
User question or task prompt |
input |
Extra context (empty string if none) |
output |
Ideal assistant response |
bucket |
Category for balancing (see below) |
source |
Corpus filename the answer is grounded in |
reviewed |
true after human leak + doctrine review |
Buckets
mdmp_steps— step identification and orderingstep_boundaries— Step 4 vs Step 5 and related trapsglossary— term definitionswar_gaming— methods, synchronization matrixcoa_screening— FASDC, criteria developmentscenario_coaching— fictional scenario trade-offs
Example
{"instruction": "What MDMP step is war gaming?", "input": "", "output": "War gaming is Step 4, COA Analysis. Staff visualize each COA through critical events and record strengths and weaknesses without head-to-head COA comparison; comparison belongs in Step 5.", "bucket": "mdmp_steps", "source": "mdmp_steps.md", "reviewed": true}
Leak-review policy
Before inclusion, each pair is checked for:
- Customer names, OPNAV, or real unit designations
- Proprietary algorithm or product terms
- Verbatim golden-eval questions (held out in
eval/golden_questions.jsonon GitHub)
Usage
import json
pairs = []
with open("pairs.jsonl", encoding="utf-8") as f:
for line in f:
if line.strip():
pairs.append(json.loads(line))
Split into train/eval with scripts/split_data.py (85% / 15%).
License
Apache 2.0
- Downloads last month
- -