Modeling and Analysis of Process Complexity and Performance in Mixed Model Assembly Systems.

Abstract

Increasing global competition demands that the manufacturing industry move from mass production into mass customization production in order to provide more varieties of products and thus satisfy customer demands. It has been shown that the increase of product variety has a negative impact on manufacturing system performance. Therefore, it is essential to understand how product variety complicates an assembly system, affecting its operation performance. Such knowledge, once validated, can be further used to improve manufacturing system design and operation.

The objective of this dissertation is to develop an enhanced general methodology for modeling and analyzing process complexity for mixed model assembly systems. The following fundamental research has been conducted:

A set of complexity metrics are proposed for measuring the complexity of various elements in a manufacturing system. These metrics are proposed by constructing a linkage with the communication system framework. Unlike the existing complexity measures defined in the literature, this research is the first effort to include production quality into the measurement of how well a manufacturing system can handle the process complexity induced by the input demand variety.

A systematic method is developed for efficiently and explicitly representing complex hybrid assembly system configurations by the use of algebraic expressions, which can overcome drawbacks of two widely used representation methods: block diagrams and adjacency matrices. By further extending the algebraic configuration operators, the algebraic performance operators are defined for the first time for the systematic evaluation of system performance metrics; these metrics include quality conforming rates for individual product types at each station, process capability for handling complexity, and production cycle time for various product types. Therefore, when compared to other methods, the proposed algebraic expression modeling method also has a unique merit in providing computational capability for automatically evaluating various system performance metrics.

An integrated model is introduced for the first time to describe the effect of operator’s factors on the process operation performance. The model includes intrinsic factors such as the operator’s thinking time and experience; and extrinsic factors such as the choice task complexity induced by the product variety in mixed model assembly systems.