Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- Before 21 July 2026, confirm in writing with each carrier on Klang-relay services whether the rotation has been re-cut since 7 July, and restate the ETA for open bookings as a two-day band rather than the contracted transit time.
- Track Busan's published waiting figure weekly from now through the end of August; treat a move above two days as the trigger to shift 15 percent of relay volume onto a direct call.
- For any shipment with a hard in-store date before 15 August, hold a three-day buffer at the destination warehouse and set a written ceiling on what you will pay for it, so the buffer does not quietly become permanent.
- By 30 September 2026, renegotiate the port-closure clause in relay-heavy contracts so that schedule recovery after a reopening is the carrier's stated obligation, not a force majeure exemption.
Kuehne+Nagel's 7–14 July update shows Port Klang yard density at 80–85% with stable but pressure-building waits around 1.38 days, while Ningbo remained under pressure from vessel backlogs and high yard use through the Bavi closure. Busan held steady at 1.5 days but faces bunching risk from upstream delays.
The Analysis
Spread the map out and put a finger on the Yangtze estuary, then run it south-west along the coast, past the Taiwan Strait, down through the South China Sea, into the Strait of Malacca, and stop where the water narrows at Port Klang. Run the finger back north-east, past the Pearl River mouth, up across the East China Sea, and stop at Busan. Those three points are tied to the same set of strings. The 7 to 14 July update puts Klang's yard density at 80 to 85 percent with waits holding around 1.38 days, describes Ningbo as still working through vessel backlogs and high yard use left by the Bavi closure, and keeps Busan steady at 1.5 days while flagging a bunching risk from delays upstream.
The typhoon is over. That is the first thing to get straight, because most of the coverage is still written as though the weather were the subject. Bavi closed things, the closures ended, the ports are open. What those three figures describe is not a storm. It is a network that has already redistributed the storm's cost and is now deciding where to keep it.
Klang is nowhere near the typhoon. It sits two bodies of water away from where the weather happened, and it is the one reporting the highest yard density of the three. That single sentence is the entire geography of this story. Ports do not fill up because it rained on them. They fill up because the boxes that could not move somewhere else had to move somewhere, and the relay hub at the mouth of the Malacca Strait is the somewhere. Eighty to 85 percent density is a yard working hard, not a yard broken. The waiting figure of roughly 1.38 days says the same thing from the water side: vessels are getting in, slowly, and the terminal is still cycling boxes.
Read the three states together, because separately they are only three numbers. Klang is absorbing. Ningbo is holding. Busan is calm and has been warned. Those are three positions in one queue, and a queue has an order to it. Weather hits a coast, the coast pushes its backlog into the relay hubs, the relay hubs push the schedule distortion into the next port of call. This update photographs the middle of that transmission, not the end of it.
Who feels it first is not the biggest shipper. It is the one whose cargo cannot skip the relay. If your box moves on a direct deep-sea call against a contracted weekly allocation, Klang's density is somebody else's problem and your transit time is probably intact. If your box has to be broken and re-stowed, or if you feed a hub from a smaller port, then the 1.38 days is already yours, and the word stable is no comfort at all. Stable at a high density means the pressure is being held, not released.
The question worth answering is when this reaches a balance sheet. Yard density is a stock and berth waiting is a flow. A yard at 80 to 85 percent with a wait near 1.38 days is a terminal turning boxes faster than they arrive. The day the wait stops tracking the density is the day the constraint moves off the landside and onto the quay, and after that the delay stops being counted in hours and starts being counted in sailings.
This is the part the update does not say out loud, and it is the part I would pin to the wall. The trigger is not Klang. Klang is already the visible symptom, and watching a symptom tells you nothing you did not know yesterday. The trigger is Busan. A port sitting at 1.5 days with a flagged bunching risk from upstream delays is a port whose schedule has not yet broken. Bunching means vessels that should have arrived across several days arrive together. When that happens, the damage is not a longer wait at one terminal. It is a rotation that no longer matches its proforma, and a rotation error repeats on a seven-day cycle. One week late becomes three weeks late without anything else going wrong. So the number to check every week is not the highest density on the page. It is the calm port that has been told to expect company.
Step back ten years and this is a pulse, not a turning point. Nobody is dredging a new channel or laying a new hub because of Bavi, and the hierarchy of these three ports will look roughly the same in 2036. Pulses leave deposits, though. Every disruption of this shape teaches carriers to build a little more slack into the rotation and teaches shippers to hold a little more cover, and slack bought for a two-week event does not get handed back when the two weeks end. Structure does not change inside the storm. It changes in the contracts signed the month after, while everyone is still annoyed enough to pay for protection.
Here is the arithmetic I would run in front of a customer, with the assumptions stated because they are mine and not the update's. Take an importer holding 21 days of cover at the destination warehouse. Add the relay wait of about 1.4 days, and add two days for a bunching event at the next hub that may or may not happen, which is my assumption rather than a reported figure. That gives 24.4 days, roughly 16 percent more inventory-days than the plan carried. Over a pulse of a few weeks that is a small amount of working capital for a short time, and a fair price if the alternative is an empty shelf. It is a bad price if you buy it as a permanent setting. The mistake I see most often is not the buffer itself. It is a buffer nobody was assigned to take down.
One trap deserves naming before the actions. Yard density is the terminal's problem. Your ETA is a schedule problem. They only become the same problem at the moment the carrier re-cuts the rotation, and until then a density chart is a weather map of somebody else's yard. The second trap is chasing the cheapest relay option during a pulse, because the cheapest option is almost always the one with the least slack inside it, and slack is precisely what you are trying to buy.
↑ Back to the key points