Definition
Difference in delivery quality metrics between AI-recommended relay loads and comparable through-run loads on the same lanes, revealing whether relay decisions save money net of any service quality degradation they introduce.
The Measurement Gap
Relay decisions are evaluated on their cost efficiency without accounting for the service quality trade-offs they create. The blind metrics include:
Cost per load: compares relay and through-run economics but never connects the cost savings to the service quality outcomes on those same loads. A relay that saves $200 in driver costs but generates a $500 damage claim and a lost customer never shows up as a net loss.
Relay savings reports: track estimated savings from HOS-optimized relay recommendations as a standalone figure, divorced from service outcome data. The savings are reported; the service failures are reported separately in a different dashboard by a different team.
On-time delivery rate (aggregate): blends relay and through-run loads into a single OTD figure that masks whether relays systematically underperform. A fleet running 92% OTD overall could have 95% OTD on through-runs and 82% on relays: the aggregate hides the pattern.
The fundamental gap: no metric compares relay and through-run outcomes on a lane-matched, apples-to-apples basis. Without this comparison, fleets cannot determine whether relay recommendations are genuinely net-positive or simply shifting costs from one line item (driver expense) to another (service degradation, claims, customer churn).
Decision Supported:
Are relay recommendations saving money net of service quality degradation, or are they moving costs between budget lines while eroding delivery performance?
Formula
Plain Language
ROD compares delivery outcomes for relay loads against through-run loads on matched lanes. Each quality dimension (on-time delivery, damage rate, customer complaints) is measured separately as the difference between relay and through-run performance. A composite score averages the three normalized differentials. A separate economic calculation determines whether the HOS savings from relays exceed the costs of any service degradation.
Notation
$$ROD_{metric} = Outcome_{relay} - Outcome_{through}$$
Component metrics computed separately:
- $ROD_{OTD} = OTD_{relay} - OTD_{through}$
- $ROD_{damage} = DamageRate_{relay} - DamageRate_{through}$
- $ROD_{complaint} = ComplaintRate_{relay} - ComplaintRate_{through}$
Composite:
$$ROD = \frac{1}{3}\sum(ROD_{metric,normalized})$$
Each metric normalized to 0–1 scale before averaging.
Net economic:
$$ROD_{cost} = (HOS_savings \times relay_count) - (excess_damage_claims + service_penalties)$$
Variables
| Variable | Description | Source |
|---|---|---|
| ROD_OTD | On-time delivery differential between relay and through-run loads on matched lanes | TMS delivery records |
| ROD_damage | Damage rate differential between relay and through-run loads | TMS + claims system |
| ROD_complaint | Customer complaint rate differential | TMS + CRM |
| HOS_savings | Estimated per-relay cost savings from hours-of-service optimization | ELD + TMS derived |
| relay_count | Number of relay loads in measurement period | TMS |
| excess_damage_claims | Incremental damage claims attributable to relay handling | Claims system |
| service_penalties | Contractual penalties triggered by relay-related service failures | TMS + contract management |
| ROD_cost | Net economic outcome: negative means relays cost more than they saved | Derived composite |
Proposed formula, pending validation pilots.