Notices arrive as documents
Port omissions, blank sailings and rotation changes land as PDFs, web pages and emails, each in its own format.
Commercial intelligence for ports & terminals
Corridor Twin is a living model of the trade corridors behind your volumes. It reads carrier notices, partner feeds and market news as they happen, shows your commercial team where untapped demand sits and why, and lets you test a move before you commit to it.
“Voyage 214W will omit Port A; cargo will be relayed via Port B on the next feeder.”
TEU a year through your terminals
Illustrative network, names and figures
Shipping lines, forwarders, cargo owners, inland operators and authorities each hold part of the picture. Commercial teams stitch it together by hand, and by the time the analysis is ready, the cargo has moved.
Port omissions, blank sailings and rotation changes land as PDFs, web pages and emails, each in its own format.
Forwarders, carriers and inland partners can't hand over commercially sensitive data, so whole stretches of each corridor stay out of view.
Volumes, network plans and customer records live in separate tools, so nobody sees where untapped demand actually sits.
When a service changes or a trade route shifts, the window to respond is short, and spreadsheets take longer than that.
Most of this comes down to one problem: getting the data at all. Industry-wide data platforms have stalled on exactly that. Corridor Twin starts where it needs nobody's permission: your own data.
How it worksSix workspaces in four steps. Each answers one question, and hands its answer to the next.
Where should we look?Every corridor in your network, ranked by the opportunity you have not captured yet. Filter by trade lane and by your footprint.
| # | Corridor | Lane | Opportunity | Δ qtr |
|---|---|---|---|---|
| 1 | Shanghai → Rotterdam | Asia → Europe | 74 | +6 |
| 2 | Busan → Los Angeles | Asia → North America | 69 | +2 |
| 3 | Singapore → Jebel Ali | Asia → Middle East | 66 | +4 |
| 4 | Rotterdam → Santos | Europe → Latin America | 61 | −1 |
| 5 | Antwerp → Lomé | Europe → Africa | 57 | +3 |
Read it as a sentence: dense, growing flows and plenty of headroom drive this corridor's score; thin inland reach holds it back. Each part is a value from 0 to 100, times a weight your team can see and change.
Weekly relay to two secondary ports.
Three departures a week to an inland hub.
Shared forecasts with a forwarder.
Carrier notices and partner feeds become structured events your team can trust, with the source one click away.
Web notices, PDFs and pasted text are read by the AI into structured events: carrier, service, voyage, ports and dates.
Every fact carries the exact passage it came from, and code checks each quote against the source.
Ports and services are matched to your corridors, and the effect is sized by what the affected sailings carry.
A named analyst reviews each extraction. Only reviewed notices change the model.
Please be advised that voyage 214W will omit Port A due to congestion. Cargo for Port A will be relayed via Port B on the next available feeder.
The hard part is rarely the model. It is getting the data at all: it sits with many parties, in many countries, under rules that limit what may move. So Corridor Twin starts with your own records, through a copy that carries their patterns and none of their secrets.
Terminal and customer data never leave your environment. When partners join later, their data goes into a clean room and never leaves it either.
A model writes a synthetic copy: the same totals and the same patterns, but no real shipper, carrier or cargo inside it.
Commercial teams in every country work on the copy, without a legal approval for each dataset.
Accuracy: are its totals close enough to the truth for a commercial team to rely on them? Privacy: can any record in it be traced back to a real one? A release check tests both, and a named person signs off.
Commercial decisions need an owner and a trail. Corridor Twin keeps each part of the work where it belongs.
The AI reads notices, news and plain-language questions, and always shows where each statement came from. It drafts; it never decides.
Scores, scenarios and effects follow rules anyone on the team can read and challenge. Same inputs, same answer, every time.
Reviews and approvals carry a name and a reason, and they are kept, so every decision has an owner and a history.
When you are ready to bring shipping lines, rail operators or forwarders in, their data joins yours in a clean room. It works like a ballot box: everyone drops their data in, nobody can open it, and only the count comes out. The partners agree in advance which questions may be asked.
Partners join through the clean room when you are ready: step 3 of getting started. Getting started →
Find the corridors worth chasing, and build the case for each with numbers anyone can check.
See how service changes and new levers move volumes across your network before they happen.
Know which customers a change touches, and tell them first, with an advisory someone approved.
Pick the partners a move needs, and share data with them at the level of detail you choose.
Start with your own data, in your own environment. Widen it at your own pace.
Connect your terminal data inside your environment, starting with two or three terminals. It is checked before anyone works with it, and it never leaves.
Bring in your commercial teams in every country, on one shared twin. Your team sets the scoring weights and the scenario rules.
Bring in partners and customers when you are ready, through the clean room: partners add their feeds, customers see their own network.
Corridor Twin is available by invitation. Tell us about your organisation and what you would like to explore, and we will get back to you to arrange access and a guided walkthrough. Most teams start with two or three terminals in one region.