Modeling and Analysis of Multiproduct Multistage Manufacturing System for Quality Improvement

Shichang Du,Rui Xu,Lin Li

DOI: https://doi.org/10.1109/tsmc.2016.2614766

2018-01-01

Abstract:The ability to produce multiple types of products using the same manufacturing system plays an essential role in the success of a manufacturing enterprise. Meanwhile, most complex manufacturing systems include many stages, called multistage manufacturing systems (MMSs). MMS has a cascade property, which means the product quality is not only affected by the current stage, but is also related to the outputs of the upstream stage. Multiproduct MMSs (M(3)Ss) have been widely applied in industry. Thus, this paper is devoted to modeling and analyzing steady-state M(3)Ss for quality improvement. The discrete Markov model for single-product-multistage system is extended to novel Markov models for multiproduct-two-stage systems and multiproduct-multistage systems by calculating state transition probabilities of six key manufacturing parameters to obtain an acceptable product quality probability. Based on the developed models, product sequence analysis is carried out to obtain the best product sequence under Bernoulli conditions and Bernoulli relaxation conditions, and bottleneck analysis under Bernoulli conditions is conducted to identify the machine and/or parameters whose improvement can lead to the largest quality improvement. The applicability of the proposed models is validated through numerical experiments and a case study using real-world data.