Cost consequence of failure in failure mode and effect analysis

Abstract

PurposeThe purpose of this paper is to investigate the possibility of including the cost consequence of failure in the a priori risk assessment methodology known as failure mode and effect analysis (FMEA).Design/methodology/approachA model of the standard costs that are incurred when an electronic control module in an automotive application fails in service was developed. These costs were related to the Design FMEA ranking of the level of severity of the failure mode and the probability of its occurrence. Monte Carlo simulations were conducted to establish the average costs expected for each level of severity at each level of occurrence. The results were aggregated using fuzzy utility sets into a nine-point ordinal scale of cost consequence. The criterion validity of this scale was assessed with warranty cost data derived from a case study.FindingsIt was found that the model slightly underestimated the warranty costs that accrued, but the fit could be improved with adjustments dictated by actual usage conditions.Research limitations/implicationsCost data used in the simulations were derived from government and academic surveys, analyses, and estimates of the manufacturing cost structure; and nominal costs for various quality issues experienced by Tier 2 automotive electronics supplier. Specificity is lacking. The sample size and the type of the failure modes used to validate the model are constrained by the number and type of products which have had demonstrable performance concerns over the past three years, with cost data available to the authors. The power of the validation is limited. The validation is considered a screening assessment.Practical implicationsThis work relates the characterization of risk with its potential cost and develops a scaling instrument to allow the incorporation of cost consequence into an FMEA.Originality/valueA ranking scale was developed that related severity and occurrence rank scores to a cost consequence rank that keys to a cost of quality figure (given as percent of sales) that would accompany a realization of the failure mode.