OneRail uses Nvidia AI for real-time last-mile delivery optimisation
ainews ·
OneRail has introduced OmniSTAR, an AI-powered delivery platform that uses Nvidia technology to help retailers, wholesalers and distributors select how individual orders should be delivered. The system combines Nvidia's cuOpt route-optimisation engine and cuDF data-processing library with OneRail's own pricing and performance data to compare fulfilment options such as fleets, couriers and parcel carriers, choosing the lowest-cost option that meets service requirements. According to OneRail, Nvidia accelerated computing can cut computation time by up to 10 times, shrinking a 20-minute calculation to under two minutes and a week-long calculation to roughly two days. This faster processing lets the optimisation run during live delivery operations, before an order is assigned. OneRail's broader AI pipeline uses machine-learning predictions about service time, lateness, delivery success and price ranges, which feed into the optimisation decisions made by OmniSTAR. The platform is already in use with selected enterprise customers; OneRail reported that US Foods used it to identify margin-damaging delivery configurations, and an unnamed large tire distributor reportedly achieved $40 million in run-rate savings over three years. OneRail also expects OmniSTAR to exceed $6 billion in gross merchandise volume in Q4 2026 and previously worked with FedEx on a SameDay Local service.
ainews