Design and analysis of complex configurations of manufacturing systems.

Abstract

Many manufacturing systems traditionally have been designed as dedicated ones using configurations based on serial flow lines in order to produce a few product models at high volumes for a relatively long period of time. These systems are not suitable for product variety or quick changeover, and new concepts and methods in manufacturing systems have been proposed and implemented to overcome these problems. The configurations of manufacturing systems, however, have been simple or symmetric ones, although more complex or asymmetric configurations can be useful for product variety and change. Furthermore, many existing methods for system design and analysis mainly concern simple configurations, and cannot be applied to complex configurations. Therefore, further research on configuration design and analysis is desirable, in particular, on non-simple and non-symmetric configurations that can alleviate problems caused by increasing product variety and frequent change. This dissertation concerns the design and analysis of complex configurations for product variety and change. First, this dissertation presents new methods for system throughput evaluations in complex configurations, providing (i) a method for the reduction of system state-space, (ii) the analysis of the effect of a configuration structure on state-space, and (iii) a procedure for throughput evaluation. Second, this dissertation proposes novel methods for balancing workload in asymmetrically configured systems designed with delayed product differentiation. Third, the dissertation establishes a framework for reusability studies by (i) defining system reusability, and (ii) proposing an assessment method. Fourth, the dissertation presents an approach for enhancing manufacturing system reusability by designing system configurations for multiple generations of products. This research enhances the fundamental understanding of system configurations, and also helps engineers to efficiently evaluate and systematically design complex asymmetric configurations in terms of productivity, balancing and reuse. In particular, this research will help engineers to select appropriate configurations from many alternative configurations beyond simple symmetric configurations.