The three stages of Semiconductor Lifecycle Management (SLiM) were discussed last time with the key phases of Stage 1 explained in some detail. (See: Nuts & Bolts of Semiconductor Lifecycle Management, Part 1.)
The following details what is needed during Stages 2 and 3 to break out of the obsolescence mitigation trap that causes logistic and financial headaches for military/aero contractors.
Stage 1 prepares the component supply chain with partnership agreements, preparation of the key components that will be required long-term, and production and fulfillment of military/aero qualification and testing requirements. This stage lays the strategic, financial, and technical groundwork for long-term success.
Moving to SLiM Stage 2 involves wafer procurement and initial validation of the semiconductor inventory to be maintained long-term. Analysis of market trend data is required to select the appropriate devices that are expected to remain in demand by military/aero programs.
Wafer storage requirements demand facilities with controlled atmosphere and environment, including securing material from theft and tampering. During this stage, periodic controls verifying the product inventory’s manufacturability should be performed. Extended production from the wafer bank at any reliability level up to space flows can be accomplished. This requires suppliers having expertise in sustaining and upgrading manufacturing processes and the equipment to continue manufacturing for decades.
Stage 3 of SLiM adopts foundry transfer methodologies. This can be accomplished by transferring intellectual property to a different foundry when available. Alternatively, this stage may also involve reengineering and manufacturing. The challenge is designing products that are equivalent in process technology, performance, reliability, and quality to the original. When applied, this method can extend product life beyond 20 years.
Two current examples of a Stage 3 approach include a high-voltage interface chip being used in the Abrams tank and Bradley fighting vehicle as well as two FIFO memory chips that are part of the Patriot missile system. In cases like these, the key to successfully maintaining secure supply hinges on establishing relationships with a wide variety of fab foundries.
This also requires design teams experienced with the fab processes and close participation by experienced product test engineering personnel. The initial and on-going investments in plant and personnel for effective SLiM programs are not trivial. However, they far offset the mayhem resulting from chasing ever-diminishing supplies of discontinued chips under typical obsolescence mitigation scenarios.
Semiconductor lifecycle management is a long game and not one for opportunists. It takes a strategic view, with early and constant partnerships with key commercial semiconductor manufacturers. SLiM players must be primarily focused on the aerospace and defense markets. It can't be a side game.
Investments must allow a robust and successful three-stage supply chain approach as discussed. Ideally, for success it should also incorporate operational presence in the two main market areas of Europe and North America where the world's military/aero systems are designed and made. When adopted, SLiM manages secure continuity of supply for as long as customers require the components. This is where, as an industry, we need to be heading.