Decision intelligence for logistics handles the daily exceptions that break plans — weather, road closures, late loads, capacity gaps — by combining TMS, WMS, telematics and customer commitments, recommending reroutes, carrier swaps or holds with their cost and service impact, and executing them on-the-loop so dispatchers can override before anything commits.
Logistics decisions
| Decision | Signal | Context it needs | Authority | Outcome measured |
|---|---|---|---|---|
| Exception reroute | Storm, closure, breakdown | Live positions, delivery windows, driver hours, alternate routes | On-the-loop — dispatcher | |
| Carrier selection | Load tendered, rejection | Rates, performance, capacity, lane history | Automated within rate bands | |
| Dock & yard scheduling | Arrivals early or late | Door capacity, labor, priorities | On-the-loop | |
| Inventory positioning | Demand shift | Forecast, lead times, service targets, costs | Augmented — planner | |
| Customer promise | Order or disruption | Capacity, inventory, route feasibility | Automated with guardrails | |
| Supplier reallocation | Supplier slip | Orders, alternate capacity, contract terms | Augmented — procurement |
Example: a storm on I-35
| Step | What happens |
|---|---|
| Signal | A storm cell forms over I-35 with 14 trucks inbound. |
| Context | Delivery windows, driver hours, alternate routes via US-75, customer priorities. |
| Options | Reroute all, hold all, or reroute nine and hold five. |
| Decision | Reroute nine, hold five for 40 minutes; executes in 4 minutes unless the dispatcher overrides. |
| Outcome | On-time deliveries and cost tracked against the do-nothing forecast. |
See it live
The homepage’s oil and gas section includes a live field-fleet map: sand trucks queue at a road closure until the plan re-routes them via a county road and frac downtime drops from 9 hours to under half an hour. Open the fleet demo →
Systems and data
Execution
TMS, WMS, YMS, carrier portals and EDI 204/214/990.
Visibility
Telematics, ELD hours of service, geofences, weather and traffic feeds.
Commercial
Rates, contracts, customer SLAs and penalties.
Planning
Demand forecasts, inventory positions, network design.
Getting to on-the-loop
Most logistics teams start with the system recommending reroutes and carrier swaps for dispatcher approval. Once agreement rates and outcomes are proven, decisions move to on-the-loop: the plan executes after a short window unless someone overrides. High-value or customer-sensitive loads stay augmented. See decision autonomy.
What to measure
| KPI | Why |
|---|---|
| On-time in full (OTIF) | The customer promise |
| Cost per shipment | Margin |
| Exception resolution time | Decision latency |
| Empty miles | Asset utilization |
| Dispatcher override rate | Model and policy quality |
- Logistics value is in exceptions, not plans.
- On-the-loop execution keeps dispatchers in control at machine speed.
- Measure decision latency as closely as cost.