Track 2 Session 6

10:30 to 11:30 a.m. Thursday April 8, 2010

Quantification of Risks Arising from Low Data Availability in Extended Warranty and Preventive Maintenance

Due to the increasing demand for extended warranties and the increasing popularity of preventive maintenance for investment goods, the importance of dependable reliability figures has increased. Reliability figures tend to be more dependable when more failure data are used to obtain them. The low availability of data, however, is a key problem in reliability analysis in the investment goods industry. Reliability figures obtained from only a small amount of failure data can be given only with uncertainty. This uncertainty needs to be assessed. In this presentation a method will be shown that uses the parameter confidence and a Monte Carlo simulation to quantify the risks that arise from this uncertainty. Furthermore, it is shown how this method can be integrated in warranty prognosis and the optimization of preventive maintenance for investment goods.

Key Words: Warranty Prognosis, Preventive Maintenance, Parameter Confidence

Patrick Werner

wbk Institute of Production Science, Karlsruhe Institute of Technology

Germany