Overview#
A transformer that scores 38 out of 100 on composite health, with a 47% estimated failure probability over the next 30 days, should have a work order raised before it fails during a heat event, not after. The Predictive Maintenance domain automates that judgement: it scores every asset across five weighted health factors, schedules preventive maintenance automatically when scores fall below thresholds, and forecasts how many outages and crews a given weather event is likely to produce.
Key Features#
- Five-factor weighted health scoring: age (20%), condition (25%), telemetry (25%), maintenance (15%), environmental (15%)
- Failure probability estimation for 30-day and 90-day horizons
- Risk classification: critical, high, medium, and low, with recommended actions and inspection schedules
- Automated preventive maintenance schedule generation for assets below health thresholds
- Work order auto-creation for overdue maintenance with priority-based scheduling
- Pre-configured maintenance definitions for 10 asset types across electric, water, and gas utilities
- Weather impact profiles for 7 event types with outage and restoration multipliers
- 23 weather alert type mappings for automated forecast triggering
- Historical outage correlation for confidence scoring
- Critical facility and life-support customer risk assessment
- Compliance framework tagging for regulatory maintenance requirements
Use Cases#
Relevant sectors include critical infrastructure, utility operations, and public safety.
- Calculating health scores across utility asset fleets to prioritise maintenance spending
- Automatically scheduling preventive maintenance work orders for at-risk assets
- Forecasting weather impact on utility infrastructure with crew pre-staging recommendations
- Tracking compliance with regulatory maintenance requirements across utility types
Integration#
The Predictive Maintenance domain reads from utility asset inventory, work order systems, telemetry alarm history, historical outage data, and organisation configuration. It creates preventive maintenance work orders and persists health scores, schedules, and weather forecasts.
Open Standards#
- ISO 55000 (Asset Management): The health scoring and preventive maintenance scheduling framework aligns with ISO 55000 and its UK predecessor BSI PAS 55, using condition, age, and maintenance history factors consistent with the standard's asset lifecycle approach.
- NERC CIP (Critical Infrastructure Protection): Transformer oil testing and substation thermographic survey schedules are tagged against the NERC CIP compliance framework, ensuring electric utility assets meet reliability regulatory requirements.
- EPA Safe Drinking Water Act (SDWA): Valve exercising and hydrant flow test PM definitions carry the
epa_sdwacompliance tag, linking scheduled work orders to US federal drinking water infrastructure obligations. - PHMSA 49 CFR Part 192: Gas main leak surveys, gas valve exercising, and cathodic protection surveys reference PHMSA pipeline safety regulations directly in their maintenance definitions.
- IEEE C57.104: Transformer dissolved gas analysis (DGA) procedures are described per IEEE C57.104, the standard guide for interpreting combustible gases in oil-immersed transformers.
- NFPA 291 / BS EN 14384: Hydrant flow testing explicitly references both NFPA 291 (US recommended practice) and BS EN 14384 (European fire hydrant standard) for recording static pressure, residual pressure, and flow rate.
- EN 13306 / EN 15341: EU maintenance terminology (EN 13306) and maintenance KPI definitions (EN 15341) are cited in the scheduler to ensure PM type classifications and performance metrics are interoperable with European utility operators.
- GraphQL: All predictive maintenance queries and mutations (health scores, failure forecasts, weather impact, PM calendar) are exposed through a GraphQL API, enabling structured, typed access from any compliant client.
Last Reviewed: 2026-02-24 Last Updated: 2026-04-14