| In this work, a system-theoretic approach to bottleneck identification in Markovian serial production lines and assembly systems is introduced and analyzed. The approach is based on the sensitivity of the system production rate to the machine reliability parameters. Using this approach, definitions of bottlenecks are introduced, methods for their identification in the case of two-machine systems are developed, and their implications for production automation and preventative maintenance are discussed. Based on these results, three methods for BN identification in longer serial lines and assembly systems are developed. The identification tools derived are based on the probabilities of machine blockages and starvations. To evaluate these probabilities and, in addition, the production rate of serial lines and assembly systems with Markovian machines, three novel aggregation procedures are proposed and analyzed. Proofs of the convergence of these procedures are provided. The results obtained in this dissertation are applied to the production systems at Chrysler Trenton engine plant and Ford Rawsonville plant. |
