The Sole Source That Stopped the Line Twice

The exposure is concentrated in one sterile-barrier component with no audit history and a relaxed receiving gate. Controls that do not depend on the sourcing decision:

1. Quantify the risk properly. Two stoppages in six months at two days each is roughly $180,000 in contribution loss, plus expedite freight and possible backorders. That number is the budget available for mitigation, and it dwarfs a 5% film price increase. Purchasing should be told this explicitly.
2. Buffer strategically, not accidentally. Safety stock sized on the supplier's demonstrated lead-time variability, not on average lead time. If the supplier is the constraint, a bonded or segregated inventory of 6-8 weeks of film may be the cheapest insurance available - but only with shelf-life and storage controls validated for a sterile barrier.
3. Put the supplier on a formal SCAR with a dated containment and corrective action: allocation agreement in writing, named resin source and sub-tier visibility, capacity commitment, and a delivery performance recovery plan with weekly reporting.
4. Reinstate the audit - a focused process audit of the film extrusion and handling, plus change-control expectations, before any further volume commitment.
5. Re-examine skip-lot. Reduced inspection was justified by 12 months of clean history; that history is now contradicted by two delivery failures and no capability evidence. Without process capability data, skip-lot is an assumption, not a conclusion. Tighten the gate and collect data to justify relaxing it again.
6. Validate the 5% offer. Priority allocation sounds like a contractual priority, not a capacity one. Get it in writing with remedies.

Both are right about their own constraint, and the decision is a risk-versus-time trade, not a vote.

Analysis to put in front of the decision-maker:

• Cost of the status quo: $180,000 in six months of stoppages, plus the tail risk of a missed sterile-barrier supply. Annualized, this is the number the dual-sourcing project must beat.
• Cost and time of dual sourcing: nine months, change control, validation of the sterile barrier (seal strength, integrity, aging), and a permanent split that reduces leverage and volume at the incumbent, possibly raising unit price on both.
• Total cost of ownership, not piece price. Include freight, buffer inventory carrying cost, expedite premium, quality risk, and the cost of regulatory revalidation. A 5% increase on film is likely a rounding error next to two days of line stoppage.

Recommendation shape:

1. Do not dual-source as a panic move with a single unqualified alternative. Begin qualification now, phased - it is a nine-month insurance policy being written today for next year's risk.
2. Simultaneously invest in the incumbent with a formal improvement plan and the allocation agreement. Supplier development is faster to first benefit than a new source.
3. Split the strategy by purpose: the second source exists for resilience and surge, not to chase price. Expect to pay a premium for that resilience and budget it as insurance.
4. Set decision gates at months 2, 4, and 6 on the incumbent's delivery and audit performance; if gates are missed, accelerate the second source.

Judgement point: the wrong answer is 'dual-source because sole source is bad.' The right answer is a risk-based, time-phased plan with explicit gates and a quantified cost of failure.

The scorecard is a lagging rear-view mirror with no predictive content. OTD 88% and 1,200 PPM tell you what happened; they do not tell you whether next month will be short.

Redesign:

• Add leading indicators: order confirmation accuracy, capacity utilization reported by the supplier, sub-tier resin allocation status, days of finished-goods inventory at the supplier, open corrective actions and their ageing, and process capability on the critical film attributes (thickness, seal strength) rather than PPM alone.
• Stratify by criticality. 220 suppliers should not be measured with one template. A sole-source sterile-barrier component deserves monthly review and a named relationship owner; a commodity carton supplier does not.
• Make the scorecard actionable. Every red metric needs an owner, a date, and a consequence - including, for chronic performers, a documented exit path.
• Tie it to the audit cycle. Four years without an audit of a sole-source critical supplier is the actual finding here. Audit frequency should be driven by risk, not by history.

Receiving inspection:

• Skip-lot is defensible only when the supplier's process is demonstrably capable and stable. With no capability data, revert to a tightened sampling plan and use the data collected to build the case for relaxing it again.
• Where skip-lot is retained, keep a periodic verification sample so the assumption is tested rather than assumed.
• Track escapes caught at receiving as a supplier metric, not just a quality department cost.

Test: can someone read the scorecard on Monday and know which supplier to call this week?