- Client
- Expeditors
- Industry
- Logistics and supply chain
- Launch
- Undisclosed
- Capability
- On-premise logistics network and data centre architecture
The brief
Freight that cannot wait for the cloud
A global freight forwarder moves shipments through a network of hand-offs, each of which creates a record and a risk. The systems that hold those records carry customer, customs and commercial data that has to stay where the company controls it, and they have to answer in real time to operations running in every time zone. For Expeditors that meant a network inside its own data centre rather than in somebody else's.
The engagement covered the whole of that network: the on-premise architecture, the data layer, the operational systems that run on it, and the capacity and failover that keep it running.
The system
Architecture, data layer, operations, capacity, failover
The architecture was planned from the workload rather than from a catalogue: which systems, serving how many operations, with what tolerance for an outage. Compute, networking and storage were designed to match, inside the company's own facility.
The data layer gives the operational systems one definition of a shipment, a party and an event, so a hand-off recorded in one system is the same hand-off in the next. The operational systems sit on top of it, and capacity and failover were designed in from the start rather than added after the first incident.
Why it matters
Where the data cannot leave the building
When a workload cannot go to a public cloud, the engineering has to be better, not simpler. There is no elastic capacity to hide a mistake behind, so monitoring, failover and capacity planning carry the weight that a cloud provider would otherwise carry. The Expeditors network is the reference for Opsian's Data Centers service.
Built on
Built on the Opsian infrastructure and data services
On-premise compute, a canonical data layer and the operational systems on top of it.
- Data CentersCompute planned, built and operated on premise, with capacity, networking, storage and monitoring designed together.
- Data IntelligencePipelines and a canonical model that give every system the same definition of the things that matter.
- Automation and IntegrationOperational systems connected through adapters behind one stable interface, built to fail safely.
Outcomes
- A logistics network running inside the company's own data centre.
- One data layer under the operational systems, so every hand-off is one record.
- Capacity and failover designed into the architecture rather than added later.
- Customer, customs and commercial data kept where the company controls it.
Related

November 1, 2024Eleven11 · Logistics
On The Rock Logistics: a new brand, a new site and an AI transportation management system
On The Rock Logistics moves freight nationwide from Texas. The engagement rebuilt its brand and website, then added an AI-assisted transportation management system and digital agents for customer support.

January 15, 2026Logistics
AtlasCore Logistics: an operating system for carriers and freight forwarders
AtlasCore Logistics is Opsian's operating system for logistics businesses: loads, dispatch, tracking, documents and settlement on one record, with automation for the routing and paperwork that carriers still do by hand.

April 8, 2026Engineering
Why operations should be modeled as infrastructure
Most operations teams document with a wiki page and a Slack channel. The teams that scale model operations the same way they model infrastructure: observed, versioned, reviewed.

