Knowledge Base Approach to Integrated Fmea

Only available on StudyMode
  • Download(s) : 131
  • Published : October 20, 2008
Open Document
Text Preview

If you want snapshots, use a spreadsheet.
If you want continuous improvement, use a Knowledge Base Approach

Artificial Intelligence, Continuous Improvement, Corrective Action ,Risk Priority Number


Integrated Failure Mode and Effects Analysis (IFMEA) is an interdisciplinary methodology for product and process improvement. The methodology employs the fundamentals of artificial intelligence and knowledge mine acquisition to develop a comprehensive decision making environment. The benefits of IFMEA include identification of controls and elimination of potential failures.


To compete in today’s marketplace, designers and manufacturers must eliminate, or at least decrease, the impact of all severe malfunctions and possible failures from their products and manufacturing processes. Moreover, modern standards and regulations (QS-9000, GMP, FAR) require designers and manufacturers to formally demonstrate that all potential malfunctions are analyzed, controlled, and their risks have been minimized (Chrysler Corporation 1995, CCH Inc. 1996). Failure Mode and Effects Analysis (FMEA) is a systematic set of activities—crossfunctional team work—intended to identify, investigate and apply better control and corrective actions to minimize a risk of potential concern.

The real objective of FMEA may be expressed as follows: not only to avoid risk but also to recognize it, price it, minimize it, and maybe even to sell it (Rafetery 1994, Bluvband 1989).

Up-front time devoted to comprehensive FMEA, at the stages when products/processes changes and improvements can be easily and inexpensively implemented, will obviate late change crises. Integrated Failure Mode and Effects Analysis (IFMEA) is based on the understanding that artificial intelligence (AI) can be applied to product/process improvement only if accompanied by a customized knowledge base that organizes, not hinders, the improvement effort. Most AI systems seek to mimic human intelligence by making sense out of ambiguous data, or by finding similarities and differences between various situations. Since the source of improvements are verbal ideas and system behavioral rules, which later evolve into analysis and implementation plans, a specialized Knowledge Base (KB) must be blended into AI systems for expediting and enhancing improvement efforts. This Knowledge Base-AI blend must support the following:

CONTINUOUS IMPROVEMENT. All products or processes must be always improved due to improvements in competitors’ products or evolution of customers’ needs. Besides, a company should be committed to continuous improvement so that its products/processes will remain cost-effective. Before making decisions on how to improve, it is necessary to review previous suggestions and analyze their advantages and drawbacks—to avoid the ones that failed, and to consider the ones that were not implemented.

DECISION MAKING. The best choice (selection of Corrective Action) can be made if exists good knowledge infrastructure, considerable analysis and judgment capabilities, based on integrated criteria such as risk estimation: the higher the risk, the higher is the priority of dealing with the subject. Thus, KB usage is an ultimate way to construct and quantify the appropriate Risk Priority Number (RPN) for most productive decisions in design, support and maintenance.

EFFICIENCY THROUGH SIMILARITY. Frequently, the knowledge gained while improving one product or process can be applied to improve another. A KB organizes knowledge so that one can quickly access it and port it to other applications.

KNOWLEDGE MINE. Ideas, recommendations, experience, success—these are acquired constantly, not just during the weekly quality meetings or brainstorming sessions. A knowledge mine is a mechanism for constantly updating, reviewing, and evaluating experts’ input into the KB.

tracking img