Accession Number : ADA292185

Title :   Artificial Intelligence Techniques for Parts Obsolescence Prediction. Phase 1.

Descriptive Note : Final rept.,

Corporate Author : STOTTLER-HENKE ASSOCIATES BELMONT CA

Personal Author(s) : Henke, Andrea L. ; Maher, Timothy P.

PDF Url : ADA292185

Report Date : 14 MAR 1995

Pagination or Media Count : 99

Abstract : The primary objective of our Phase I effort was to explore the feasibility of a Navy-wide DMSMS prediction system and develop improved methods of obsolescence prediction. In pursuit of this goal we investigated the DMSMS process at various Navy sites and identified and evaluated tools or processes currently in use. We decided to focus our efforts on micro circuit obsolescence prediction, because our study revealed that other types of parts are not nearly as significant a DMSMS problem. Furthermore, we concentrated on automating the largely manual obsolescence prediction currently performed by the MOM program. We used the artificial intelligence techniques of knowledge engineering, case-based reasoning, knowledge base development and object oriented programming to devise a solution to the obsolescence prediction problem. We also developed a preliminary design and functional description for a pro active Navy-wide DMSMS intelligent system which will be developed in a Phase II effort. We implemented both these solutions in a prototype improve their feasibility beyond a doubt. There is great potential for use of the Phase II system throughout the Navy, other branches of the military and in the commercial sector.

Descriptors :   *PREDICTIONS, *ARTIFICIAL INTELLIGENCE, *OBSOLESCENCE, INTELLIGENCE, COMMERCE, NAVY, COMPUTER PROGRAMMING, SITES, KNOWLEDGE BASED SYSTEMS, PARTS, MICROCIRCUITS.

Subject Categories : Cybernetics
      Economics and Cost Analysis
      Electrical and Electronic Equipment

Distribution Statement : APPROVED FOR PUBLIC RELEASE