| Summary: | Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering; and, (S.M.)--Massachusetts Institute of Technology, Sloan School of Management, 1999. === Includes bibliographical references (p. 71). === A critical step in many improvement methodologies is the identification of process bottlenecks. The techniques for identifying bottlenecks that are suggested by many improvement methodologies are no more than simple heuristics. These simple heuristics may prove inadequate in a complex manufacturing environment. This thesis asserts that the use of analytical tools is often required for successful bottleneck identification. This claim is supported by two examples from a research internship conducted in ,m automotive manufacturing plant. The first example describes the process used to reduce the average cycle time of a system consisting of two production cells that shared a resource. In this system, the average cycle time is strongly influenced by the manner in which the two cells contend for the shared resource. Depending on the system parameters, reducing the cycle time of one cell may actually increase the average cycle time of the system. Two methods for analyzing this system are described, and applied to identify a low-risk, low-cost cycle time reduction opportunity. The second example describes the process used to identify the bottleneck operations in a complex serial production line. The line consisted of 31 machines performing 15 different operations. Parts were automatically processed and transferred between operations. Minimal buffers existed between operations, making common bottleneck identification heuristics infeasible. The bottleneck operations were identified through the use of an analytical software tool. The identified bottlenecks were different than those predicted without detailed analysis. Once developed, an analytical tool may not be accepted by those responsible for implementing improvements. Sources of resistance to the tools are discussed and methods are suggested for overcoming this resistance. Three benefits of implementing analytical tools for bottleneck identification are discussed. These benefits are: the correct identification of bottlenecks, the enhanced alignment of improvement teams around the identified bottlenecks, and the opportunity for the team to improve the bottleneck identification process. === by Jeff R.M. Longcore. === S.M.
|
|---|
