Energy Optimizer Override Value Index (EOOVI)

GMI-RAIL-030Rail TransportationPilot-ReadyPattern O-5: Override Value / Trust

Definition

Ratio of outcomes (fuel consumption, arrival time, and safety events) when engineers override the energy management system (EMS) versus outcomes when the EMS runs uninterrupted, measuring whether overrides improve or degrade performance.

The Measurement Gap

The Measurement Gap

Railroads know how often engineers override the energy management system (EMS). What they don't know is whether those overrides make things better or worse. Override frequency is tracked as a compliance metric (high override rates trigger training interventions) but nobody systematically compares outcomes on override segments versus autonomous segments to determine whether the engineer's judgment actually improved the trip.

This gap matters because override policy is currently set by assumption, not evidence. If overrides are valuable on specific territory types or under specific conditions, suppressing them through compliance pressure destroys information. If overrides are consistently costly, the railroad is paying for interference it could prevent. Without outcome comparison, neither question can be answered.

Decision Supported:

Are engineer overrides of the EMS valuable corrections or costly interference? Should override authority be expanded, constrained, or differentiated by override type? EOOVI provides the outcome data to make these decisions by comparing fuel consumption per gross ton-mile and schedule adherence on override versus autonomous segments.

Formula

Plain Language: Compare fuel consumption per gross ton-mile and schedule adherence on trip segments where the engineer overrode the EMS versus segments where the EMS ran autonomously. Compute the outcome ratio.

Notation:

EOOVI_fuel = Fuel_per_GTM_override / Fuel_per_GTM_autonomous
EOOVI_schedule = Arrival_deviation_override / Arrival_deviation_autonomous

Variables:

Variable Definition Unit Source Classification
Fuel_per_GTM_override Fuel consumption per gross ton-mile on segments where engineer overrode the EMS gal/GTM Locomotive telemetry NATIVE
Fuel_per_GTM_autonomous Fuel consumption per gross ton-mile on segments where the EMS ran autonomously gal/GTM Locomotive telemetry NATIVE
Arrival_deviation_override Mean arrival time deviation (minutes late) for trips with overrides Minutes Operations system NATIVE
Arrival_deviation_autonomous Mean arrival time deviation for trips without overrides Minutes Operations system NATIVE

Sign Convention: EOOVI_fuel < 1.0 means overrides save fuel (good override). EOOVI_fuel > 1.0 means overrides waste fuel. Same logic for EOOVI_schedule: < 1.0 means overrides improve schedule performance.

Edge Cases:

  • Override for safety: if the engineer overrides for an obstruction, track hazard, or signal indication, exclude from EOOVI calculation. Track safety overrides separately.
  • Override correlation with territory difficulty: engineers may override more on difficult territories (steep grades, heavy curvature). If those territories inherently have worse fuel economy, EOOVI will be biased. Control for territory type.
  • Minimum sample: 50 override events and 50 autonomous events per measurement period, matched by territory type.

Proposed formula, pending validation pilots.

At a Glance

Category
Autonomous Train Operations and Human Capability
← Automated Dwell Equity Score (ADES)Energy-Schedule Net Value Index (ESNVI) →

Assess Your Organization

The AI-GPS Assessment evaluates your AI governance maturity and recommends KPIs like this one for your implementation roadmap.

Take the Free AI-GPS Assessment