The right resources at the right time!
Many electronic parts have life cycles that are shorter than the life cycle of the product they are a component of. Life cycle mismatches caused by the obsolescence of electronic parts can result in high sustainment costs for long life systems. In particular, avionics and military systems often encounter part obsolescence problems before being fielded and nearly always experience part obsolescence problems during their field life. Successful obsolescence management strategies and plans should be analyzed before taking any decision for action as the goal is to reduce systems’ life cycle costs by minimizing the time needed to resolve obsolescence cases.

Obsolescence of products occurs due to future part innovation especially in areas where technology develops rapidly. Today, innovation cycles in all areas, from the smallest daily objects up to complex electronic circuits, are significantly faster than in the past.

In order to avoid obsolescence a wide variety of specific and detailed actions can be taken at all stages of product development and production to minimize life cycle costs.

A comprehensive obsolescence management includes management of contracts. Suitable obsolescence management clauses on both sides of the supply chain with suppliers and customers are needed. Here is an example clause according to the “SD-22 - Diminishing Manufacturing Sources and Material Shortages (DMSMS) Guidebook,” published by the Defense Standardization Program Office (DSPO):

A well-planned obsolescence management strategy measures three dimensions: the impact of obsolescence (which parts are affected), the probability of obsolescence occurring to specific parts and the cost associated with the resolution chosen. Costs occur when alternate sources are needed to procure parts from a different manufacturer, a redesign of a system is unavoidable or because of other parameters (see figure below).

The importance of a vital mix of the semiconductor supplier chain is evident. One good way to get the right parts in time and in the designated quality is to have a variety between original manufacturers, franchise distributors, catalog distributors and brokers.

To successfully operate obsolescence management, the ability to forecast obsolescence events for parts that are needed to manufacture or sustain a product is essential. Past trends are a valid indicator of the future. As a consequence algorithms developed by using data mining based forecasts can be helpful for obsolescence forecasting. Generally, a distinction can be made between forecasting for parts with evolutionary parametric drivers and for parts without evolutionary parametric drivers.