A cost-benefit model for production vehicle testing

Abstract

In 1995 a record 17.8 million vehicles were recalled in the USA. In addition to these recalls, warranty costs for non-recallable defects, and the cost of lost future sales due to dissatisfied customers, contribute to the cost associated with poor quality. Although pre-production testing has more leverage, post-launch vehicle testing may provide valuable information in an economical fashion. By field testing the first production vehicles off the line, automotive manufacturers may detect manufacturing-related and component-interaction defects that would not be discovered by pre-production testing. We present a cost-benefit model for production vehicle field testing. Using data on defects and warranty costs for past vehicle programs, we can estimate the average warranty cost, the likelihood of a recall, the number of vehicles recalled, failure rates, and severity of defects for future vehicle programs, as well as the warranty costs avoided by a particular production vehicle test program. This cost-benefit analysis allows us to determine if a post-production vehicle test program is worthwhile, and if so, it allows us to choose the best such program.