ABSTRACT
This chapter describes the application of neural networks to forecast warranty performance in the presence of “warranty growth” phenomena. Section 9.1 introduces the phenomena of warranty growth. Section 9.2 discusses warranty performance forecasting using log-log plots and dynamic linear models. Sections 9.3 and 9.4, respectively, provide RBF and MLP neural network methods for forecasting the warranty performance. To optimize the network parameters, training and testing errors are minimized through planned experimentation. Section 9.5 discusses the results obtained. The key points are summarized at the end of the chapter in Section 9.6.
