Schedule reliability by carrier, from real arrivals.
ForProcurement and routing desks choosing between near-identical sailings
Two sailings look identical on paper: same lane, same transit, a few dollars apart. One of them habitually arrives four days late. The schedule never told you that. The arrivals did.
Schedule reliability by carrier, from arrivals we saw
SchedulesMCP scores on-time performance per carrier and lane from arrivals we actually observed through the tracking layer, the same one that follows containers across 240+ carriers. Every score shows its sample size, and thin lanes are labelled early-data. Carriers publish their own reliability claims; this one is measured from where the ships really docked.
One sailing, not three near-identical rows
Carriers share vessels. Three of them often sell slots on the same hull, which is why most schedule tools show you the same sailing three times. SchedulesMCP groups by the physical vessel: one card, every way to book it, with the detail a desk decides on.
- Transit time and transshipment ports and legs.
- Cut-offs, so you don’t miss the gate.
- Emissions, when that’s part of the call.
Compare carriers before you book
Enter the lane by port name.Sort the sailings by reliability, transit or cut-off.Open the one you want and book it with the carrier that sells it.
Type “Ningbo to Rotterdam”, not CNNGB to NLRTM. Names resolve to the codes the data uses, so you find the lane on the first try. The same comparison is an MCP tool (compare_sailings, get_lane_reliability) and a REST call, so an assistant or your TMS gets the numbers your desk sees.
Questions about schedule reliability by carrier
What is schedule reliability? The share of sailings that arrive when the schedule said they would. SchedulesMCP measures it per carrier and lane from arrivals observed through the tracking layer, and shows the sample size next to the score.
Why not use the carrier's published on-time figure? A carrier reports on its own network in its own way. A score built from observed arrivals compares every carrier on the same lane by the same rule.
What happens on a lane with few observed arrivals? The score is labelled early-data, so you can see it rests on a small sample and weigh it accordingly.
Can I get reliability through the API or an AI assistant? Yes. The get_lane_reliability and compare_sailings MCP tools and the REST API return the same scores behind the same key.