A New Decision Making Method for Product Development Based on Multiple Neural Network

Product design is a knowledge intensive activity and unstructured decision-making process; designers must make correct decision according to a great amount of uncertain information for a successful product development. That leads to the unsuccessful rate of new product development (NPD) higher than 30%. This article is a contribution to a new decision-making method based on multiple neural networks for NPD from the early stage of product development. First, we describe an evaluation index system which considers every aspect for NPD. Then, how to apply the rough set theory to simplify the index system is described. Thirdly, according the complexity of NPD decision-making, the technology of multiple neural network (MNN) is proposed to support NPD decision-making, and the MNN structure of NPD decision-making is described. Finally, the core of MNN application for NPD decision-making is highlighted