Superpower Daily
Smack’s $61M Funds Wrist AI for Disconnected Operators, With a...Department of War Deploys ChatGPT Mil at IL5, Built for More Th...Flower Labs Starts Endeavor 1.0 Preview With a Private Deployme...Salesforce Pushes Agents Into CRM Workflows as NIST Turns to Id...Emerson Analysis Says AI Data Centers Need Batteries for 100-MW...Skild AI’s S1 Uses One Human Video to Prompt Robots Through 10-...Smack’s $61M Funds Wrist AI for Disconnected Operators, With a...Department of War Deploys ChatGPT Mil at IL5, Built for More Th...Flower Labs Starts Endeavor 1.0 Preview With a Private Deployme...Salesforce Pushes Agents Into CRM Workflows as NIST Turns to Id...Emerson Analysis Says AI Data Centers Need Batteries for 100-MW...Skild AI’s S1 Uses One Human Video to Prompt Robots Through 10-...
Productspublished

OneRail Says OmniStar Can Cut Delivery Routing Analysis to 2.5 Minutes

The new decision layer promises to replace slow, manual delivery choices with per-order routing. Early deployments offer a first test, while its savings and volume figures remain company projections.

By 2 min read
OneRail Says OmniStar Can Cut Delivery Routing Analysis to 2.5 Minutes
OneRail Says OmniStar Can Cut Delivery Routing Analysis to 2.5 Minutes
Sign in to save and react

Listen to this story

The audio brief

About 1:21
0:001:21
Read transcript
OneRail has put OmniStar into customer use, giving retailers an artificial-intelligence decision layer for the last mile. The system chooses a carrier, a route, and a delivery mode for each individual order, rather than relying only on static routing rules or manual planning. That matters because delivery choices can be expensive and time-consuming, especially when retailers are comparing several ways to get an order from a store or warehouse to a customer. OneRail says OmniStar cuts delivery-routing analysis from about 20 minutes to 2.5 minutes. The company says Nvidia’s hardware and software make it possible to run those decisions at scale. The platform is the result of a three-year collaboration between the two companies, and deployment is already underway. One early customer is a large tire distributor, offering an initial test of how the system works in live operations. OneRail also forecasts that OmniStar will handle more than six billion dollars in gross merchandise volume during the fourth quarter. But that is a company projection, not evidence yet of repeatable savings across retailers or delivery networks. The launch therefore shows movement from development into customer use, while the larger business case remains unproven. The key thing to watch is whether the reported 2.5-minute analysis time translates into consistent, measurable efficiency gains across different retail networks—not just faster decisions in early deployments.

Story brief

3 key points

OneRail is putting OmniStar into customer use as an AI decision layer for last-mile fulfillment, choosing a carrier, route, and delivery mode for each retail order. Built with Nvidia after a three-year collaboration, the platform is designed to replace slower rule-based or manual analysis with faster per-order optimization. OneRail reports a reduction from roughly 20 minutes to 2.5 minutes and forecasts more than $6...

  1. 01

    OmniStar evaluates delivery alternatives order by order instead of relying solely on static routing rules or manual planning.

  2. 02

    OneRail reports routing-analysis time falling from about 20 minutes to 2.5 minutes.

  3. 03

    A large tire distributor is among the early customers using the platform.

OneRail has launched OmniStar, an AI platform developed with Nvidia to select a carrier, route and delivery mode for individual retail orders. OneRail says the system can sharply shorten routing analysis, positioning faster decisions as a way for retailers to manage costly last-mile delivery.

The product has already been deployed with some customers, including a large tire distributor. Its core task is narrow but consequential: weigh available delivery options for each order rather than depend on static rules or manual planning.

A decision layer for the last mile

OneRail says Nvidia hardware and software allow those decisions to run at scale. The companies began working together three years ago, according to OneRail, to bring AI into logistics operations.

OneRail’s routing-speed claim

Delivery-routing analysis
About 20 minutes→2.5 minutes
minutes

OneRail says OmniStar reduces delivery-routing analysis from about 20 minutes to 2.5 minutes.

Early customer economics are still forecasts

The company also estimates OmniStar will surpass $6 billion in gross merchandise volume in the fourth quarter. Together, those measures show the launch has moved into customer use, but they do not yet establish repeatable savings across retailers or delivery networks.

An AI addition to an existing orchestration business

The new product’s proposition is therefore not a general retail assistant. It is a faster per-order choice among delivery alternatives, aimed at giving smaller retailers access to delivery operations OneRail says can improve efficiency and compete with larger rivals.

Sources

  1. cnbc.comOneRail launches AI platform with Nvidia for retailers to make faster delivery decisions
  2. onerail.comOneRail to Transform Omnichannel Fulfillment from Shelf to Doorstep - OneRail

Related coverage

Continue reading