Sourcing strategy for a critical moulded component

A mid-sized European manufacturer decides how to source a component fitted to every unit it builds, weighing a sole-source renewal, two sole-source alternatives and a dual-source split. The log shows ...

Published content

1. node: Assumption. The two sites fail independently. They are in different countries with different inbound routes and different upstream resin suppliers, but no formal assessment of shared upstream dependency has been carried out. If they share an upstream material source, the disruption estimate for this option is materially understated.
2. node: Context. The organisation is a mid-sized European manufacturer of commercial refrigeration units, assembling in one plant and selling into the European Union and the United Kingdom. The moulded housing is fitted to every unit in the range, so a supply interruption stops the assembly line rather than reducing output. Planned volume is approximately 40,000 units a year.
3. node: Which sourcing strategy should be adopted for the moulded housing from the 2027 model year: continue with the incumbent supplier, move to an alternative single source, or qualify a second source and split volume?
4. node: Evidence. The incumbent's renewal quotation holds the unit price flat for three years with indexation capped at 3 per cent a year, which is below the rate the category has moved at over the previous three years.
5. node: Evidence. The alternative European supplier quoted 4.35 euro per unit at the planned volume, firm for two years, which is approximately 6 per cent below the incumbent's renewal price.
6. node: Evidence. The distant supplier quoted 3.41 euro per unit landed. At the planned volume that is approximately 48,000 euro a year below the incumbent's renewal price.
7. node: Decision-maker. The Chief Operating Officer, under the delegated authority for supply agreements with a committed value above 500,000 euro over the term. The Procurement Director prepared the option set and recommendation.
8. node: Scope. This decision covers the sourcing strategy and the supplier set for one component. It does not cover the design of the housing, the assembly plant footprint, or the terms of the customer contracts that depend on delivery performance. Those are governed separately and are treated here as fixed.
9. node: Evidence. Delivery performance across 2025 was 98.1 per cent on time in full, the highest of any supplier for which the organisation holds comparable data.
10. node: Evidence. A site visit on 12 May 2026 confirmed two moulding lines, one of which is idle, and that the planned volume can be absorbed without new capital investment by the supplier.
11. node: Evidence. Under a 60/40 split, a total stoppage at either site leaves the organisation able to meet approximately 70 per cent of demand by shifting volume to the other, in place of none under any single-source option once inventory cover is exhausted. The estimate assumes the incumbent can run about 10 points above its 60 per cent share and the alternative supplier can use its idle line to rise from 40 to 70 per cent; neither rate has been tested.
12. node: Consulted. Manufacturing operations on line stoppage consequences, Quality on the qualification programme, and Finance on the working capital effect of each inventory position. Manufacturing operations recorded a preference for the dual-source option before the costs were compared.
13. node: Risk. Eight weeks of inventory cover in place of three holds about 3,850 additional units, roughly 13,000 euro of working capital at the landed price, plus stock in transit on a 54-day route. The carrying cost is small against the price advantage; the case against this option rests on disruption exposure, the one-shipment correction loop and freight rates that are not held for the term.
14. node: Action. Begin the qualification programme for the second source by 30 June 2026, so that first-article inspection and both pilot runs complete before the incumbent agreement expires.
15. node: Constraint. The supply agreement with the incumbent supplier expires on 31 March 2027 and permits one extension of no more than six months. A decision that requires a new source to be in place must therefore be taken in time to complete qualification before that date.
16. node: Risk. All output comes from a single site, which was interrupted for four weeks by flooding in 2023. A three-year renewal without a break clause commits the organisation to that exposure for the full term, and the estimated annual disruption probability of 14 per cent is not in the low range in which the published analysis reports single sourcing to be preferable.
17. node: Risk. Moving wholesale to an unproven source replaces a known single-site exposure with an unknown one. The organisation would hold no delivery history at the point the incumbent relationship ends, and no fallback if early performance is poor.
18. node: Risk. A defect discovered on receipt cannot be corrected inside the replenishment cycle. The correction loop is one shipment long and there is no second source to draw on while it runs.
19. node: Risk. The full qualification cost is incurred for 40 per cent of the volume, so the blended unit cost is about 0.16 euro above the alternative European supplier as sole source, before any volume-breakpoint penalty.
20. node: Action. Commission an upstream dependency assessment covering both sites before volume is committed, to test the independence assumption on which this option's disruption estimate rests.
21. node: Constraint. Any new source must complete first-article inspection, materials certification and two pilot production runs before it may supply series parts. On the organisation's previous qualifications this has taken between three and five months.
22. node: Risk. Renewal leaves the organisation without a qualified alternative, so the response to any future price increase or performance decline is a qualification programme that takes months to complete.
23. node: Risk. The estimated annual disruption probability, 22 per cent, is the highest of the four options and the furthest from the low range in which the published analysis reports single sourcing to be preferable.
24. node: Risk. Two suppliers carry roughly twice the procurement workload: two audit cycles, two price reviews and two quality data sets, with no additional headcount planned in the procurement function.
25. node: Assumption. Annual volume remains within 10 per cent of 40,000 units across the contract period. The unit costs compared below are quoted at that volume and are not valid outside it, because all three quotations contain volume breakpoints.
26. node: Review trigger. If the upstream dependency assessment finds a shared material source, reconsider this decision before volume is committed. The disruption estimate that distinguishes this option from the single-source alternatives would no longer hold.
27. node: Assumption. The disruption probabilities used in the comparison are internal estimates, formed from each site's own interruption history and the exposure of its inbound route. They are not measured frequencies and should not be read as such. They are used to place each option against the published decision thresholds rather than to compute an expected cost.
28. node: Review trigger. Review the split in March 2028. If the second source's delivered lead time has held at or below the quoted 22 days across two consecutive quarters, consider moving it to the majority share. If it has not, consider returning to a 100 per cent incumbent position while retaining the qualification.
29. node: Research. Yu, Zeng and Zhao, "Single or dual sourcing: decision-making in the presence of supply chain disruption risks" (Omega, 2009), reports that the preferred strategy turns on the probability of disruption rather than on cost alone: single sourcing is favoured when disruption is unlikely, and dual sourcing becomes preferable once the disruption probability exceeds a threshold that depends on the cost and demand parameters. That threshold is a model result under the authors' assumptions and has not been computed for this component. The organisation's estimates for the three single-source options, 11 to 22 per cent a year, are not in the low range the analysis associates with single sourcing; the result is used to frame the comparison, not as a decision rule.
30. node: Research. Work on supply chain resilience for single and multiple sourcing in the presence of disruption risks models how a second source protects supply under disruption. The protection it credits to a second source rests on the two sources not failing together; two suppliers exposed to the same route, region or upstream material would not deliver it. This is the basis for treating the independence of the two sites as an assumption to be tested rather than a property of the option.
31. node: Research. Silbermayr and Minner, "Dual sourcing under disruption risk and cost improvement through learning" (European Journal of Operational Research, 2016), reports that the unit-cost penalty of splitting volume narrows over the life of a contract as the second source moves down its own cost curve. A first-year price comparison therefore overstates the long-run penalty of a split. The size of that narrowing is model-dependent and has not been estimated for this component.
32. node: Renew with the incumbent supplier as sole source
33. node: Move to the alternative European supplier as sole source
34. node: Move to the distant lower-cost supplier as sole source
35. node: Qualify the alternative European supplier as a second source and split volume 60/40 with the incumbent
36. node: Recommendation. Qualify the alternative European supplier as a second source and split volume 60/40 with the incumbent from the 2027 model year. The distant supplier's price advantage is real, but it comes with the highest estimated disruption probability, a correction loop one shipment long, and freight rates that are not held for the term. All three single-source options carry an estimated annual disruption probability of 11 to 22 per cent, outside the low range in which the published analysis reports single sourcing to be preferable. The recommendation accepts a unit cost about 0.16 euro above the cheapest European option, plus any volume-breakpoint penalty, and additional procurement workload, in exchange for the ability to continue supplying roughly 70 per cent of demand through the loss of either site.