If your company builds products that contain chips, you are exposed to the semiconductor supply chain whether or not you ever touch a wafer. A single unavailable component can stop a whole production line. This checklist focuses on practical steps a hardware, procurement or product team can take to understand and reduce that exposure. It deals in principles rather than market data, since conditions change often.

Start With What You Actually Use

You cannot manage risk you have not mapped. Begin with the bill of materials (BOM) for each product and ask, for every semiconductor part:

• Who is the manufacturer, and which distributors can supply it?
• Is it a standard part available from multiple vendors, or a proprietary one from a single source?
• What is its lifecycle status according to the manufacturer’s current documentation: active, not recommended for new designs, or approaching end of life?
• How long does it typically take to obtain, and how variable is that time?
• How hard would it be to replace if it vanished?

Sorting parts by how painful a loss would be, rather than by price, usually reveals that a few inexpensive chips matter far more than their cost suggests.

Classify Risk by Part Type

Not all chips present the same risk. Rough categories help focus effort.

Part type	Typical risk profile	Example mitigation
Standard, multi-source parts	Lower; alternatives often exist	Approve alternates in advance
Single-source parts	Higher; one supplier controls availability	Seek second source or buffer stock
Custom or programmed parts	High; hard to replace	Long-term agreements, redesign plan
Parts near end of life	Rising; supply may stop	Last-time buy or migration plan
Complex parts with software dependence	High; replacing requires software work	Abstract hardware in firmware early

The Checklist

1. Build and maintain a complete BOM with manufacturer part numbers. Generic descriptions are not enough to assess supply. Keep the BOM current as designs change.

2. Flag single-source and sole-fabricated parts. Ask your suppliers where the chip is manufactured and packaged when this is not clear from documentation.

3. Review lifecycle status regularly. Monitor manufacturer notices about discontinuation, and schedule a periodic lifecycle review rather than relying on surprises.

4. Qualify alternates before you need them. Approving a second-source or pin-compatible alternative takes engineering and testing time. Doing it during calm periods avoids doing it in a crisis.

5. Design for flexibility. Where practical, choose standard footprints, avoid unnecessary dependence on unique features, and keep firmware layers that isolate hardware differences.

6. Share forecasts with suppliers. Realistic demand forecasts help vendors and distributors plan. Be honest, since inflated forecasts damage trust.

7. Consider buffer inventory selectively. Stock carries cost and risk of obsolescence. Reserve it for critical, hard-to-replace parts and decide quantity from lead-time variability and the cost of a stoppage.

8. Use reputable sources and verify authenticity. Constrained supply attracts counterfeit parts. Buy from authorized channels where possible and apply incoming inspection for any parts from other sources.

9. Plan for end-of-life. For a part being discontinued, compare options: a last-time buy, a redesign, or a compatible replacement. Decide early because choices narrow as stock disappears.

10. Document a response plan. Define who decides what when a supply alert arrives, so action does not depend on someone remembering a spreadsheet.

A Hypothetical Example

A small company makes a connected sensor device with a BOM of a few dozen parts. During a review, the team finds three flagged items: a microcontroller available only from one vendor, a power chip nearing end of life, and a radio module fabricated on a single process by a single supplier.

They prioritize. For the power chip, they identify a compatible alternative and begin testing it. For the microcontroller, they cannot easily replace it without firmware changes, so they agree a longer-term supply arrangement and hold a modest buffer sized to cover the typical gap in deliveries. For the radio module, they start a design study to support an alternate module in the next board revision.

None of these actions eliminates risk, but together they convert unknown dependencies into managed ones.

Questions to Ask Suppliers

• Where are the die fabricated and where are they packaged?
• What is the manufacturer’s stated policy on notifying customers of discontinuation?
• Are there pin-compatible or functionally compatible alternatives?
• How are allocation decisions handled if supply becomes constrained?

Get answers in writing, and treat unclear answers as information.

Balancing Cost and Resilience

Resilience costs something: extra engineering, inventory or higher-priced parts. The right level depends on how costly a stoppage would be compared with the cost of protection. A consumer gadget with many substitutes may accept more risk than medical, industrial or automotive equipment where requalification is slow. Make the tradeoff explicitly instead of discovering it through a shortage.

Compliance Notes

Trade rules, export controls and sourcing requirements differ by jurisdiction and product type. Consult qualified legal and compliance professionals for your situation rather than relying on general guidance.

Frequently Asked Questions

What is a last-time buy? A final purchase of a part before it is discontinued, intended to cover expected future demand. It trades inventory cost and obsolescence risk for continuity.

Is a second source always possible? No. Custom, programmable or highly specialized parts may have no direct alternative, so redesign planning may be the only option.

How much buffer stock should we hold? It depends on lead-time variability, how critical and replaceable the part is, and the cost of stock. Model your own situation instead of copying a rule of thumb.

How do we spot counterfeit parts? Prefer authorized channels, request traceability documentation, and use incoming inspection. Details of testing methods vary, so consult industry guidance.

Should engineering or procurement own supply risk? Both. Engineering understands substitution feasibility; procurement understands suppliers and markets. Shared ownership works best.

Next Step

Export your BOM, mark every semiconductor part as multi-source or single-source, and pick the three highest-impact single-source items. Assign an owner to each with a simple question to answer this month: what would we do if this part disappeared?