wangdx25/oil038_sample

OIL-038 — Synthetic Equipment Failure Dataset (Sample) (TsFile)

Apache TsFile version of xpertsystems/oil038-sample.

Overview

A schema-identical preview of OIL-038, the XpertSystems.ai synthetic equipment-failure and predictive-maintenance dataset for oil & gas rotating and stationary assets (pumps, valves, compressors). The full product covers ~6,500 facilities across a 3-year horizon with ~6 million telemetry rows. This sample is the generator's demo mode (20 facilities × ~18 assets × 90 days) covering all 16 product tables.

• Tables: 16 TsFile table(s), 115,031 rows in total.
  • alarm_events: 15,460 rows, 2023-01-01 00:00:00 → 2023-03-31 00:00:00.259000 (UTC)
  • compressor_failures: 92 rows, 2023-01-01 05:00:00 → 2023-03-31 09:00:00 (UTC)
  • downtime_events: 396 rows, 2023-01-01 05:00:00 → 2023-03-31 19:00:00.001000 (UTC)
  • environmental_conditions: 30,510 rows, 2023-01-01 00:00:00 → 2023-03-31 00:00:00.338000 (UTC)
  • equipment_failure_labels: 339 rows, 1970-01-01 00:00:00 → 2023-03-31 19:00:00.001000 (UTC)
  • equipment_master: 339 rows, 2002-09-08 00:00:00 → 2024-12-15 00:00:00 (UTC)
  • facility_master: 20 rows, 1970-01-01 00:00:00 → 1970-01-01 00:00:00.019000 (UTC)
  • lubrication_analysis: 4,407 rows, 2023-01-01 00:00:00 → 2023-03-26 00:00:00.338000 (UTC)
  • maintenance_work_orders: 396 rows, 2023-01-01 10:02:32.650000 → 2023-04-01 01:16:32.170000 (UTC)
  • pump_failures: 129 rows, 2023-01-01 05:00:00 → 2023-03-31 19:00:00.001000 (UTC)
  • reliability_metrics: 339 rows, 1970-01-01 00:00:00 → 1970-01-01 00:00:00.338000 (UTC)
  • root_cause_analysis: 396 rows, 1970-01-01 00:00:00 → 1970-01-01 00:00:00.395000 (UTC)
  • sensor_drift: 1,013 rows, 2024-01-03 00:00:00 → 2025-12-27 00:00:00.002000 (UTC)
  • thermal_profiles: 30,510 rows, 2023-01-01 00:00:00 → 2023-03-31 00:00:00.338000 (UTC)
  • valve_failures: 175 rows, 2023-01-01 12:00:00 → 2023-03-31 00:00:00 (UTC)
  • vibration_signals: 30,510 rows, 2023-01-01 00:00:00 → 2023-03-31 00:00:00.338000 (UTC)

Schema (TsFile structure)

alarm_events.tsfile

• Time (INT64, milliseconds) — the source timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (9): alarm_id (string), equipment_id (string), facility_id (string), alarm_code (string), severity (string), alarm_value (double), response_delay_minutes (double), acknowledged_flag (int64), alarm_flood_flag (int64)

compressor_failures.tsfile

• Time (INT64, milliseconds) — the source failure_timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (15): failure_id (string), equipment_id (string), facility_id (string), failure_mode (string), severity_score (double), root_cause (string), detected_by (string), prior_mtbf_hours (double), estimated_mttr_hours (double), unplanned_shutdown_flag (int64), surge_index (double), vibration_mm_s (double), bearing_temp_f (double), discharge_temp_f (double), compression_efficiency_pct (double)

downtime_events.tsfile

• Time (INT64, milliseconds) — the source downtime_start column.
• TAG: none — the table is stored as a single device.
• FIELD (10): downtime_event_id (string), failure_id (string), equipment_id (string), facility_id (string), shutdown_type (string), downtime_hours (double), production_loss_bbl (double), repair_cost_usd (double), safety_impact_flag (int64), environmental_impact_flag (int64)

environmental_conditions.tsfile

• Time (INT64, milliseconds) — the source timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (7): equipment_id (string), facility_id (string), ambient_temp_f (double), humidity_pct (double), corrosive_index (double), salt_fog_exposure (int64), h2s_exposure_flag (int64)

equipment_failure_labels.tsfile

• Time (INT64, milliseconds) — the source last_failure_timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (9): equipment_id (string), facility_id (string), equipment_type (string), failure_probability_30d (double), failure_probability_90d (double), criticality_score (double), risk_rank_score (double), predictive_maintenance_priority (string), recommended_action (string)

equipment_master.tsfile

• Time (INT64, milliseconds) — the source installed_date column.
• TAG: none — the table is stored as a single device.
• FIELD (13): equipment_id (string), facility_id (string), equipment_type (string), manufacturer_family (string), equipment_age_years (int64), criticality (string), rated_capacity (double), design_margin (double), normal_operating_pressure_psi (double), normal_operating_temp_f (double), maintenance_strategy (string), sensor_coverage_score (double), baseline_mtbf_hours (double)

facility_master.tsfile

• Time (INT64, milliseconds) — synthetic row index over the stable source order; this table has no clock column.
• TAG: none — the table is stored as a single device.
• FIELD (9): facility_id (string), facility_type (string), region (string), operating_environment (string), facility_age_years (int64), process_complexity_score (double), maintenance_maturity_score (double), corrosive_index (double), safety_criticality (string)

lubrication_analysis.tsfile

• Time (INT64, milliseconds) — the source sample_timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (8): sample_id (string), equipment_id (string), facility_id (string), oil_quality_score (double), contamination_index (double), water_ppm (double), particle_count_iso_code (string), viscosity_change_pct (double)

maintenance_work_orders.tsfile

• Time (INT64, milliseconds) — the source created_timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (11): work_order_id (string), failure_id (string), equipment_id (string), facility_id (string), repair_type (string), priority (string), completion_status (string), planned_labor_hours (double), actual_labor_hours (double), spare_parts_delay_hours (double), maintenance_quality_score (double)

pump_failures.tsfile

• Time (INT64, milliseconds) — the source failure_timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (16): failure_id (string), equipment_id (string), facility_id (string), failure_mode (string), severity_score (double), root_cause (string), detected_by (string), prior_mtbf_hours (double), estimated_mttr_hours (double), unplanned_shutdown_flag (int64), cavitation_score (double), seal_leak_rate_lpm (double), bearing_temp_f (double), pump_efficiency_pct (double), suction_pressure_psi (double), discharge_pressure_psi (double)

reliability_metrics.tsfile

• Time (INT64, milliseconds) — synthetic row index over the stable source order; this table has no clock column.
• TAG: none — the table is stored as a single device.
• FIELD (15): equipment_id (string), facility_id (string), equipment_type (string), criticality (string), baseline_mtbf_hours (double), equipment_age_years (int64), maintenance_strategy (string), failure_count (int64), downtime_hours (double), observed_mtbf_hours (double), mttr_hours (double), availability (double), reliability_score (double), criticality_weight (double), risk_rank_score (double)

root_cause_analysis.tsfile

• Time (INT64, milliseconds) — synthetic row index over the stable source order; this table has no clock column.
• TAG: none — the table is stored as a single device.
• FIELD (12): rca_id (string), failure_id (string), equipment_id (string), root_cause (string), contributing_factor_1 (string), contributing_factor_2 (string), rca_method (string), human_factor_flag (int64), design_factor_flag (int64), process_factor_flag (int64), corrective_action_required (string), rca_confidence_score (double)

sensor_drift.tsfile

• Time (INT64, milliseconds) — the source last_calibration_date column.
• TAG: none — the table is stored as a single device.
• FIELD (7): sensor_id (string), equipment_id (string), facility_id (string), sensor_type (string), drift_percentage (double), calibration_status (string), signal_quality_score (double)

thermal_profiles.tsfile

• Time (INT64, milliseconds) — the source timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (6): equipment_id (string), facility_id (string), temperature_f (double), thermal_stress_index (double), cooling_efficiency_pct (double), thermal_cycle_count_30d (int64)

valve_failures.tsfile

• Time (INT64, milliseconds) — the source failure_timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (15): failure_id (string), equipment_id (string), facility_id (string), failure_mode (string), severity_score (double), root_cause (string), detected_by (string), prior_mtbf_hours (double), estimated_mttr_hours (double), unplanned_shutdown_flag (int64), leakage_rate_scfh (double), actuator_response_ms (double), stiction_score (double), position_error_pct (double), packing_leak_flag (int64)

vibration_signals.tsfile

• Time (INT64, milliseconds) — the source timestamp column.
• TAG: none — the table is stored as a single device.
• FIELD (8): equipment_id (string), facility_id (string), rms_velocity_mm_s (double), fft_peak_hz (double), kurtosis (double), crest_factor (double), health_index (double), load_factor (double)

Conversion notes

• One TsFile per source CSV (16 tables); each table is stored as a single device, so the rig / well / tank / equipment / sensor identifiers stay ordinary FIELDs.
• Time = the table's own clock column as INT64 milliseconds (event timestamps, sample or inspection dates, and daily record dates). The clock columns used here are created_timestamp, downtime_start, failure_timestamp, installed_date, last_calibration_date, last_failure_timestamp, sample_timestamp, timestamp.
• Tables with no clock column (facility_master, reliability_metrics, root_cause_analysis) use a synthetic row index over their stable source order as Time.
• No column is dropped. Rows that share a timestamp are ordered by a 1 ms step in stable source order, so daily tables keep their exact calendar day.

Usage

Read this dataset with the Java SDK. It is written with the zstd compressor. The Apache TsFile Java SDK reads every compressor, but the Python SDK (pip install tsfile) implements only UNCOMPRESSED, SNAPPY, GZIP and LZ4; opening this file with it yields no rows.

Read a converted file with the Apache TsFile Java SDK (org.apache.tsfile:tsfile):

import org.apache.tsfile.file.metadata.IDeviceID;
import org.apache.tsfile.read.TsFileReader;
import org.apache.tsfile.read.TsFileSequenceReader;
import org.apache.tsfile.read.common.Path;
import org.apache.tsfile.read.expression.QueryExpression;
import org.apache.tsfile.read.query.dataset.QueryDataSet;
import org.apache.tsfile.write.schema.MeasurementSchema;

import java.util.ArrayList;
import java.util.List;

String path = "alarm_events.tsfile";
try (TsFileSequenceReader sequenceReader = new TsFileSequenceReader(path);
     TsFileReader reader = new TsFileReader(sequenceReader)) {
    // Devices are enumerated as IDeviceID objects. Tables with TAG columns
    // have multi-segment device ids that cannot be rebuilt from their string
    // form, so always use getAllDevices() rather than getAllDeviceIds().
    for (IDeviceID device : sequenceReader.getAllDevices()) {
        List<Path> paths = new ArrayList<>();
        for (MeasurementSchema schema : reader.getMeasurement(device)) {
            String name = schema.getMeasurementName();
            if (name != null && !name.isEmpty() && !name.equalsIgnoreCase("Time")) {
                paths.add(new Path(device, name, true));
            }
        }
        QueryDataSet dataSet = reader.query(QueryExpression.create(paths, null));
        while (dataSet.hasNext()) {
            System.out.println(dataSet.next());
        }
    }
}

Source & license

• Original dataset: xpertsystems/oil038-sample
• Source card title: OIL-038 — Synthetic Equipment Failure Dataset (Sample)
• License: cc-by-nc-4.0 (as declared by the original dataset).