From 3.7 million trips
to one operating decision
An explainable review system for taxi operations. Every alert ends with evidence, an investigation prompt, and a metric to test.
From data burden to operating decision
Three operating decisions
Select a case to follow the evidence chain.
Early-January alert geography
Top alert zones plus the Times Square trip-volume deviation.
Hover or focus a highlighted zone.
Geometry: official NYC TLC taxi zones, February 2026 snapshot.
Trip-volume deviations concentrated in several zones. An operator would inspect source records and local context before deciding whether any staging test is justified.
Measured and unmeasured impact
System impact
- 94.8%fewer alerts than the first configuration
- 436 / 436alerts linked to evidence receipts
- 8 / 8deterministic tests passed
Business impact
- Tripscompleted per driver-hour
- Milesdriven without a passenger
- Timepassenger pickup and airport waiting
Move from map signal to proof
- 1Select an alert patternConcentration, airport, or duration shift
- 2Inspect highlighted zonesFind the place and hour
- 3Review the evidenceSignal before explanation
- 4Choose whether to investigateHuman decision required
- AChoose test zonesFive alert-informed zones
- BChoose control zonesFive comparable zones
- CChange one decisionStaging, dispatch, or ETA
- DMeasure the differenceMoney, time, risk, service
It turns anomaly detection into an operating test.
Signal → evidence → human decision → controlled action → money and service metric.