“The largest oil supply disruption in the history of the global market; bigger than the 1970s oil shocks.”
— International Energy Agency, on the 2026 closure of the Strait of Hormuz

The signal before the shockThe strait began closing three weeks early

The strait did not close on the twenty-eighth of February. It began closing three weeks earlier; in the price of insurance.

For a global consumer-goods company, the Strait of Hormuz is not where its containers sail; its Asia–Europe boxes run through the Red Sea and the Suez Canal. But the two are one weather system. When the United States and Israel opened an air war on Iran on 28 February 2026 and Iran moved to block Hormuz: a passage through which about a quarter of the world’s seaborne oil and a fifth of its LNG travel (CRS, 2026); the Houthis resumed attacks on Red Sea shipping the very same day, and Asia–Europe container traffic was forced around the Cape of Good Hope, adding weeks and cost to every box (Carra Globe, 2026). A retailer half a world from the Gulf watched its cost-to-serve move before a single customer order changed.

The question this raises is not whether a planning system can predict an airstrike. It cannot, and any vendor who claims otherwise should be shown the door. The question is whether the fabric was already leaning before the shock landed; and whether it kept re-planning, continuously, long after the news moved on. That is what an autonomic supply chain does, and the Hormuz crisis is an almost clinical demonstration of it.

Before, readiness, not prophecyWhat continuous scanning actually reads

The precursors to the closure were legible to anyone scanning the right signals. War-risk insurance premiums for the strait climbed from about 0.125% to between 0.2% and 0.4% of hull value per transit in the days before the strikes: a quarter-million-dollar swing for a large tanker, and a live market pricing the probability of exactly this event (Strait of Hormuz crisis, 2026). Iran tripled its oil exports between 15 and 20 February and drew down storage to de-risk its own position; there had already been a brief partial closure earlier that month, staged as a warning; and marine-traffic data showed tankers beginning to loiter rather than transit (Strait of Hormuz crisis, 2026). None of these forecast the 28 February strikes. Together they did something more useful: they moved the readiness posture.

This is the supply-chain form of what physiologists call allostasis; stability through change, achieved by shifting the setpoint before demand arrives rather than defending a fixed one. A scanning fabric that reads a rising war-risk curve does not wait for certainty; it re-weights its scenario probabilities and, where the shift crosses a governed threshold, quietly changes the plan. For our retailer, the pre-event readjustments are concrete: risk-score every lane and supplier with Gulf or Asia–Europe exposure; lift safety stock on the most exposed SKUs; pre-book alternate capacity around the Cape and hold air-freight options open; pull priority replenishment forward; extend or lock ocean-freight and bunker-fuel positions before rates move; and stage a margin-at-risk view so finance is not surprised. The plan is deliberately aimed a little ahead of where the business is going; so that if the strait closes, the enterprise has already half-moved, and if it does not, the cost of leaning was a modest, reversible buffer.

After: the cascade that does not stopOne event, a hundred continuous corrections

On 28 February the worst branch of the scenario tree became the base case. Hormuz was effectively closed; Maersk, CMA CGM and Hapag-Lloyd suspended transits; the Red Sea shut to commercial traffic as Houthi attacks resumed; and Asia–Europe boxes rerouted around the Cape of Good Hope, adding ten to fourteen days and materially higher cost per container (Carra Globe, 2026). Brent crude jumped ten to thirteen percent with analyst forecasts reaching $100–130, feeding directly into freight, fuel surcharges and input costs (Carra Globe, 2026). Around two thousand ships were left stranded in the Gulf, insurers withdrew war-risk cover, and authorities estimated six months to clear the mines (Al Jazeera, 2026).

For the retailer, the post-event response is not a decision but a continuous re-plan across every process at once; and, crucially, it does not stop when the headlines fade. When the threat level was downgraded from critical to severe on 7 June and a trickle of five to ten ships a day resumed via a southern route near Oman, with talk turning to transit tolls, each development was simply another signal that moved the plan again (CNBC, 2026). Reopening is re-planned with the same machinery as closure: buffers unwind, cheaper routes are re-booked, and the setpoint drifts back; carefully, because the risk is not gone. The loop never closes; it only changes what it is correcting toward.

The choreographySix processes, one continuous re-plan

What makes this autonomic rather than merely fast is that the six core supply-side processes do not move in sequence, waiting for one another; they move together, in a single governed loop, across every phase of the crisis. The predecessors, sensing, inventory, logistics and procurement, lean before the event; the whole system re-plans continuously after it; and material moves surface to a human rather than firing blind. Exhibit 2 sets out the choreography as a living timeline: read down each column to see what every process is doing in a given phase, and read along the bottom to see the correction loop that never stops turning.

The financial shadowEvery supply move is a P&L move, computed together

None of these supply decisions is financially silent. In an autonomic fabric the financial consequence is not reconciled after the fact in a month-end meeting; it is computed in the same loop, in near real time. Higher freight per container, bunker surcharges, the working capital tied up in bigger safety stock, the premium paid to expedite priority SKUs by air, the longer cash-to-cash cycle of a Cape routing; each flows through the shared model to landed cost, to cost of goods sold, to gross margin and EBITDA, and into the rolling forecast. This is the murmuration turn of the parent essay made concrete: a supply signal felt on one flank becomes a re-priced P&L on the other, within hours, without a reconciliation meeting. Exhibit 3 traces the bridge for a single exposed product family.

The spineHow the signal flows through the architecture

The choreography above only works because of what sits beneath it. And the honest technical story starts with a distinction most architecture diagrams blur: the signals that matter here never touch the system of record. AIS traffic, war-risk premiums, container-freight indices and news feeds do not live in S/4HANA; they arrive as external data products in SAP Business Data Cloud. The internal master and transaction tables, the supplier, the material, the open purchase order, the journal, are where a correction eventually lands, not where sensing happens. So the architecture is really two bloodstreams that meet: an external signal stream and an internal system-of-record stream, joined at the SAP Knowledge Graph, which alone knows that a war-risk spike on a Gulf lane connects to this supplier, these materials and plants, those customers, and a specific line of the P&L. Exhibit 4 traces the full spine, with the real tables named at each hop (SAP News Center, 2026)(Constellation Research, 2026).

The correction mechanismWhy the loop never closes

The feedback arc on the right of Exhibit 4 is the whole point. Every action the fabric takes: a rerouted container, a lifted safety-stock level, an alternate-source purchase order, a re-priced forecast; produces measured actuals that land in the Universal Journal and in goods-movement records, and those actuals are immediately compared with what the plan expected. The gap is the signal for the next correction. This is negative feedback in the strict, homeostatic sense: the system is not steering toward a fixed annual number but continuously closing the distance between the plan and a moving world. When demand-sensing accuracy on a disrupted lane degrades, a model-drift monitor on the loop flags it and triggers re-estimation rather than letting a stale reflex fire. And because the mechanism is symmetric, it governs the good news as well as the bad: when the strait’s threat level was downgraded and a southern route reopened, the very same loop began unwinding buffers and re-booking cheaper lanes, correcting back with the same discipline it used to correct away(CNBC, 2026).

The boundaryWhat still rises to a human

None of this is autonomy without a conscience. The moves that materially change the business: a price increase passed to customers, a large freight commitment, onboarding an unqualified supplier, anything that materially moves the P&L; are held for human approval by the Agent Hub, exactly as the parent essay’s decision-authority matrix requires. The reflexes handle the reversible, bounded majority; the exceptions rise to the people who own the trade-off. And one honest line matters above all: nothing here predicted an airstrike. The fabric read precursors and shifted posture; it compressed the reaction from weeks to hours; it kept correcting as the facts changed. That is a great deal; but it is readiness and speed, not prophecy, and it depends on the quality of the signals feeding it. AIS data can be spoofed or go dark, premium and index data lag, and a single deeply-integrated fabric concentrates dependency as much as it enables coherence. A serious operator treats those as design constraints, not footnotes.

A question for supply-chain and finance leaders: when the next chokepoint tightens, will your planning already be leaning on the precursors; or will you convene a meeting to decide what the market decided three weeks ago?

#SupplyChain   #SupplyChainPlanning   #SAP   #SAPIBP   #AgenticAI   #FPandA   #Resilience   #StraitOfHormuz

SourcesReferences

Methodological note. The Strait of Hormuz scenario is real and drawn from the cited reporting as of mid-2026; the situation was evolving at the time of writing. The retailer, its exposures and the numbers in Exhibits 1 and 3 are illustrative and directional, not a specific company or measured series. Table names (LFA1, MARC, EKKO, ACDOCA, and so on) are representative of the standard SAP S/4HANA data model, and SAP IBP, SAP Analytics Cloud and SAP Enterprise Planning capabilities are shown at representative hops; real implementations differ. This piece makes no claim that any system predicted the event; anticipation is precursor-driven posture-shifting, and its value depends on signal quality. It is a work of analysis, not investment, legal or procurement advice.