The fast progress of technology has made a product's life cycle shorter and shorter. Transferring experience gained from pilot runs to mass production rapidly has thus become an important issue for enterprises. Neural networks are one of the learning models widely applied to implement this task. However, neural networks generally require a great amount of training data to establish the learning model, which is difficult to collect in the early stages of a manufacturing system. Therefore, in this paper, for cases when the collected data is insufficient, a procedure proposed by Li et al. (Li, D.C., Wu, C.S., Tsai, T.I. and Lin, Y.S. Using mega-trend-diffusion and artificial samples in small data set learning for early flexible manufacturing system scheduling knowledge. Comput. Oper. Res., 2009, 34, 966-982), called mega-trend-diffusion, is applied to make the forecast more precise. This paper takes the multi-layer ceramic capacitor as an object of study, and applies the procedure to the pilot runs of production to create a robust model of the process to shorten the lead-time for mass production. The results reveal that it is possible to rapidly develop a model of production with limited data from pilot runs.
All Science Journal Classification (ASJC) codes