OneRail has launched OmniSTAR, an AI-powered delivery optimisation platform built with Nvidia technology to help retailers, wholesalers and distributors decide how individual orders should be delivered.
The platform evaluates fulfilment options such as owned fleets, couriers and parcel carriers, then selects an option that meets the required service level while aiming for the lowest delivery cost. OneRail combines Nvidia’s cuOpt optimisation engine and cuDF data-processing software with its own delivery pricing and performance data.
According to OneRail, the system can reduce some computation times by as much as 10 times. A calculation that previously took around 20 minutes can be completed in under two minutes, while a calculation that took about a week can be reduced to roughly two days.
The platform can rerun delivery scenarios as conditions change, including traffic, road blockages, driver availability, fuel costs, weather and shipping conditions. OneRail says this enables more fulfilment options to be compared during live operations.
OmniSTAR is already deployed with selected enterprise customers. OneRail said US Foods used the system to identify delivery configurations that were reducing margins and subsequently adjusted pricing and delivery patterns. The company also reported that an unnamed large tyre distributor achieved $40 million in run-rate savings over three years.
The development highlights how AI optimisation is moving beyond forecasting into real-time logistics decisions, where delivery cost, service requirements and changing operational conditions must be balanced continuously.
Leave a Reply