Definition
Ratio of actual utilization variance across the fleet under AI-optimized dispatch compared to expected variance under uniform assignment, revealing whether the dispatch algorithm systematically concentrates wear on compliant vehicles while underutilizing assets with open maintenance items.
The Measurement Gap
Legacy fleet metrics create a dangerous blind spot around dispatch-driven wear concentration. The standard metrics that mask this problem include:
Fleet utilization rate: reports aggregate percentage of fleet in active service but never reveals whether the same subset of vehicles absorbs disproportionate mileage while others sit underutilized.
Vehicle availability: tracks whether assets are mechanically ready for dispatch but says nothing about whether available vehicles are actually being dispatched equitably.
Miles per vehicle: reports fleet-wide averages that smooth over the variance entirely. A fleet averaging 8,000 miles per month per vehicle could have compliant trucks running 12,000 while trucks with open items run 4,000, and the average would never surface the split.
None of these metrics expose the distribution pattern underneath. When AI dispatch optimizes for compliance and cost simultaneously, it naturally gravitates toward vehicles with zero open work orders, creating a two-tier fleet where compliant assets accumulate accelerated wear and non-compliant assets fall further behind on utilization, a self-reinforcing cycle no legacy metric detects.
Decision Supported:
Is AI dispatch creating a bimodal fleet where some trucks run hard and others sit idle. Driven by compliance status rather than operational need?
Formula
Plain Language
DWCI compares how spread out vehicle mileage actually is versus how spread out it would be under even distribution. A value of 1.0 means dispatch is distributing miles as evenly as a baseline period or theoretical uniform model. Values above 1.0 mean dispatch is concentrating miles on fewer vehicles. The supplemental Compliance Concentration Ratio (CCR) isolates whether that concentration follows compliance status.
Notation
$$DWCI = \frac{CV_{actual}}{CV_{expected}}$$
Where:
$$CV = \frac{\sigma_{miles}}{\mu_{miles}}$$
Supplemental:
$$CCR = \frac{\overline{Miles}{zero_open_items}}{\overline{Miles}{1+_open_items}}$$
Variables
| Variable | Description | Source |
|---|---|---|
| CV_actual | Coefficient of variation of miles per vehicle during measurement period | Computed from telematics/ELD mileage data |
| CV_expected | Baseline coefficient of variation from pre-AI dispatch period or theoretical uniform distribution | Historical data or modeled baseline |
| σ_miles | Standard deviation of miles driven across all active vehicles in the fleet | Computed from telematics/ELD |
| μ_miles | Mean miles driven across all active vehicles in the fleet | Computed from telematics/ELD |
| Miles_zero_open_items | Average miles for vehicles with zero open work orders | CMMS work order status + telematics |
| Miles_1+_open_items | Average miles for vehicles with one or more open work orders | CMMS work order status + telematics |
| CCR | Compliance Concentration Ratio: values above 1.5 indicate compliant vehicles running 50%+ more miles | Derived from CMMS + telematics join |
Proposed formula, pending validation pilots.