Traverse captures the operational knowledge inside your commercial teams. The knowledge that used to walk out the door, live in inboxes, or exist only in one scheduler's head - now permanent, queryable, and compounding into better decisions on every barrel.
Commercial teams carry years of accumulated knowledge about which assets are reliable, which routes move commodity efficiently, which signals matter. That knowledge exists in inboxes, spreadsheets, and the heads of people who leave.
Traverse captures it. The infrastructure layer maps every asset, every connection, every grade. The operations layer gives schedulers inventory projection, demand forecasting, replenishment paths, and voyage planning. The intelligence layer monitors the market, classifies signals, and correlates events to the assets that matter.
Each system compounds. The graph gets richer with every verified signal. The forecast improves with every actual. The routes get faster with every constraint added.
A refinery fire breaks. A pipeline shuts. A tanker diverts. These events reach Traverse before they reach the desk - ingested from RSS feeds, X/Twitter, and webhooks, classified by LLM into event type and severity, and correlated against your asset network automatically.
Your analyst verifies in three stages: confirm the signal, associate the affected assets, group with related events in the same geography and timeframe. Every decision is logged. Every link is auditable.
The knowledge that used to live in one trader's inbox is now part of the organisation.
Signals classified, geolocated, and matched to your asset network - automatically. Your analyst verifies, links, and logs. The rest of the organisation sees the same picture.
Four workbenches unified under a single control tower interface. Each slides in from the right - full-window, no context switch. Your schedulers work the same way they already work, with the data they already know, structured for the decisions they make each day.
Truck via road network. Rail across 50,000km of OpenRailwayMap-derived track. Barge through the Rhine, Netherlands, and Belgian waterway graph - built from IENC S-57 charts at 298,892 nodes. Vessel across the maritime network. Pipeline via the connected asset graph.
One API call. Five engines. Every route returned is a route that can actually move commodity - constrained assets flagged, pipeline graph edges validated, waterway access checked before the result reaches your scheduler.
12,868 assets modelled as a connected graph in PostGIS. Every terminal, refinery, tank farm, pipeline segment, rail loading facility, and jetty - with operator, geometry, access flags, and grade compatibility.
Ask how commodity gets from Rotterdam to Ingolstadt. Traverse walks the graph across pipeline, barge, and rail - and returns every viable route with constraints surfaced.
We'll load your assets, run your routes, and show you what your team's knowledge looks like when it's permanent.
