Table of contents
Anyone that works in the electronics industry knows that component shortages are a persistent reality. Shortages have periodically cycled between capacitors, discrete semiconductors, and integrated circuits, with certain sectors like the automotive industry being hardest hit. One risk-reduction strategy that companies are starting to embrace is creation of design variants, where designs are built around alternate components and are ready to build at a moment’s notice.
Because a given design could contain dozens or hundreds of components, which specific components should receive swap priority when creating variants? Creating variants with the goal of reducing supply chain risk requires an up-front investment in time and design effort. Design teams should determine which components in their designs carry the greatest supply chain risk, and these should be prioritized as part of variant creation and procurement.
Focus on Components That Create Value
There is one supply chain strategy that is useful for reducing supply chain risk in moderate volume or high-volume production: create variants of a design and manage production of variants. The approach is not new, but it is becoming the norm rather than the exception.
Every component plays its unique role in a product, but not all components create the same level of value. Certain components are absolutely critical to executing primary functions in a product, and they have very few or no replacements. If companies want to take a supply chain risk mitigation strategy, they will have to take a hard look at components that have few replacements in order to maintain sustained production of their product.
Unfortunately, it can be very difficult to balance the need to use value-creating components with risk reduction. It can also be incredibly time consuming to create a large number of variants that address every possible at-risk component in a given design. This means a design and procurement team should prioritize which components create the most value in an end product yet carry the highest risk.
In my opinion, the value hierarchy for component swaps in variant creation would be ranked as follows:
- Specialty ASICs (unique function/footprint)
- Specialty ASICs (common function/footprint)
- Embedded processors (within a given product family)
- Embedded processors (unique products)
- Memories
- Common ASICs
- Discrete semiconductors
- Passives
Obviously, it’s not possible to create a set of design variants that addresses every component group in this list. Designers should instead triage the highest value/highest risk components when using variants to sustain production and reduce supply chain risk.
Which Components to Triage First?
Anyone who looks at this list would probably put embedded processors (CPUs, MCUs, etc.) somewhere near the top as they are the main value-creating components in many of the consumer devices we enjoy today. However, as we saw over the course of 2020 and 2021, many of the commercial and automotive-grade ICs that were persistently low on stock were specialty ASICs, not embedded processors.
The main reason specialty ASICs claim the top value creation spot is because they perform major functionality that is not always replicated in another component like an embedded processor. They also are not easily replaceable outside the same product family. This means these components can create the most value through their unique functionality, but they also carry the most risk. If possible, these components are top priority for identifying suitable component swaps for design variants.
As we look down the above list we start to see that many of these component groups come in more common packages and even with common pinouts. Passives are usually the lowest risk (except the 2018-2020 era) as there can be millions of these parts in stock with major distributors at any given time. They are also among the smallest cost drivers in terms of parts costs.
Passives tend to have low cost, high inventory, and common packaging. This makes passives among the lowest risk parts in a BOM.
Financial Risk for Parts vs. Assemblies
Depending on where components are being procured and a product is being manufactured, there can be additional costs that incurred when procuring parts for a design. Overseas procurement of critical parts involves some or all of the following costs:
- Shipping
- Insurance
- Customs duties
- Import tariffs
- Packaging
- Additional time for logistics management
The increased landed costs for some components reduces the value they could potentially provide. If these costs are excessive, as is often the case when shopping in the gray market, it’s worth looking at a design variant that uses local parts that do not carry the same additional costs.
BOM Management
Each variant involves changes to the BOM so that procurement managers can track what parts will be used in various assembly runs. How should the BOM for a product be managed once variants are created, and production is being planned around a group of variants?
Individual BOMs for each variant are best separated based on alternates that cannot be interchanged between devices. Suppose you have two design variants for a product, and each uses a different FPGA. If these FPGAs have different footprints and pin connections, then each variant should have its own BOM. Furthermore, if you’re kitting your own components, or using an inspection and kitting service, then each should have its own kit as well.