Field damage analysis (FDA) concept: Analysis of complex damage causes

• Main Authors: Bracke, S, Haller, S
• Format: Conference Proceeding
• Language: English
• Subjects: Algorithm design and analysis; Automotive engineering; Industries; Production; Reliability; Statistical analysis; Weibull distribution
• ISSN: 0149-144X; 2577-0993
• DOI: 10.1109/RAMS.2011.5754482

Increasing functionality and complexity of technical products is resulting in complex damage symptoms and causes during customer usage phase. Especially in the automobile industry, complex damage symptoms are often traceable to multiple damage causes. This increases the importance of structured field product observation and field data analysis. For the OEM, the goals are high levels of quality and reliability that can be attained by comprehensive analysis of product performance, design and reliability over the whole product life cycle. This includes: (1) After product launch into the market: Early and accurate identification of possible and actual damage causes by analysis of failures provides knowledge of product failure behavior. This can be attained by analyzing the occurring failures based on small amount of field damage data. The cost implications should also be assessed. (2) Product life cycle: Detailed mapping and analysis of the long-term product failure behavior and reliability These generates the basis for a targeted introduction of further actions, for example rectification of faults in the field, product optimization, or the initiation of concentrated development approaches/-strategies for failure prevention (e.g.: COP strategies). Currently no industrial standard for reliability analysis can fulfill the requirements of a comprehensive reliability analysis over the entire product life cycle. Based on these requirements the FDA concept was developed by the Department of Risk Management and Safety Engineering at the University of Wuppertal, in cooperation with OEMs of the automotive industry. Key aspects of the FDA concept are as follows: (1) Integration of several statistical and organizational methods in a comprehensive process (2) Mapping product reliability over whole product life cycle (3) Statistical analysis of failure data - for example, relating to identification of possible damage causes, production batch differences or climatic influences (4) Usage of qualitative data and information of the value added network regarding potential damage causes that can reduce the costs of the technical analysis (5) Conjunction of the determination of the regress rate and detection of the critical failures (product quality analysis) in a comprehensive approach. Currently different requirements - made by the OEMs - avoid the conjunction, e.g. sampling procedures and sample size (6) Optimization of analysis costs in terms of sampling procedure, sampling a