STUDY AND IMPROVEMENT OF WARRANTY CLAIM FORECAST METHODS

Abstract: The Vehicle Quality Information (YQI) group at Scania provides a monthly basic warranty forecast in order to predict the 12-months basic warranty level before the products have actually finished their first year in use. The presently used method for forecasting is based upon differentiation over only two parameters: responsible department for a claim and delay in claim reporting. However, warranty cost varies within a wide range depending on the complexity of the repair, thus perhaps being driven by several other parameters as well. Especially for small product series, single high costs and early repairs might cause great overestimation of the final warranty cost. This thesis work aims to improve the calculation of the warranty claim forecast and to develop a measurement of the forecast uncertainty, especially for small product series, such as the present Euro 6 engines.