|
June 27–29, 2017
The College Park Marriott Hotel and Conference Center College Park, MD |
|
Click here to register
Existing Obsolescence Prediction Models Are Now Obsolete
Bill Fliegel (bill.fliegel@converge.com)
Bio
Bill Fliegel is a Director for end of life services Future of Obsolescence Management (FOM) at Converge, a division of Arrow Electronics. With 20 years experience in distribution to High Reliability companies, Bill has held leadership roles in Product Management, Field Sales and Global Account Management. Before joining Arrow's Zeus Aerospace & Defense team in 2012 as a Regional Sales Manager, Bill worked for Avnet and Honeywell.
Abstract
Over the last decade, institutions and organizations focused on the increasing challenges of electronic component obsolescence have turned to a range of lifecycle prediction tools to help component users act proactively to mitigate the impact of obsolescence on system sustainability.
Approaches to this challenge can be categorized into three areas:
- Lifetime buy of existing components (logistics).
- Identify + qualify substitutes (engineering).
- Redesign (engineering + logistics).
Each of these approaches has inherent shortcomings, all of which can be reduced to a common financial trade-offs calculation.
Early mitigation actions yield the highest downstream financial benefit when measured against the costs associated with shortage, counterfeits, or redesign. For example, early identification of the need for an alternate device gives engineers and suppliers more time to collaborate in the identification, application and qualification of alternate sources, which can then be added as approved sources, thus eliminating the need for re-design or a lifetime buys.
Risk prediction tools, like those developed by The Center for Advanced Life Cycle Engineering (CALCE) at the University of Maryland, aim to provide users with algorithms to help users recognize the relative lifecycle (discontinuance) risk of the parts they are using. These models depend on observed patterns related to historical lifecycle patterns for certain technologies and market characteristics that are then fit against pre-established lifecycle curves.1 The use model for tools based on these algorithms is intended to help users isolate emergent risk based on period-to-period changes in the risk profile of specific parts. Common Product Lifecycle Management (PLM) platforms may integrate these algorithms or similar predictive tools to help companies identify latent risk within their enterprise supply chain, and parse mitigation responsibility accordingly.
We call into question the current efficacy of these tools however. The semiconductor component market is going through unprecedented merger and acquisition activity. As a result, lifecycle prediction models based on historical patterns for segmented device or technology groups, source populations, market inventory trends, and pricing trends are rendered obsolete as the historically relevant “barometers” are supplanted by the economic imperatives arising from company mergers, namely the necessity to prune away underperforming product lines quickly to achieve projected expense and profitability commitments.
As a result, during this especially turbulent period, semiconductor users must rely on a rigorous, financially grounded methodology to assess and act on obsolescence events. Risk awareness “dashboards” are needed for critical components, and pre-qualified, easy to implement financial tools are required to allow supply chain decision makers to visualize, prioritize, and fund discontinued component acquisitions.
As with other industry shifts, institutions like CALCE will quickly adjust to the changing dynamics of electronic component lifecycles. But for the next five years, users must understand the inherent shortcomings in this traditional approach, and seek solutions that mitigate these shortcomings.
1 Rajeev Solomon, Peter Sandborn, and Michael Pecht, Electronic Part Life Cycle Concepts and Obsolescence Forecasting IEEE Trans. on Components and Packaging Technologies, Dec. 2000, p 6
About Coverge, a Arrow Company
Converge is a global independent distributor, in existence for 35 years and wholly owned by Arrow since 2010. Converge provides shortage, obsolete and excess component solutions to customers, through the quality processes of AS 6081 and AS 9120. Converge has established a separate end of life division called FOM to provide analytics, consultancy and to develop a supportive community around the strategic challenge of obsolescence for Converge and Arrow customers.