Date of Award
Civil and Environmental Engineering
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Cellular Manufacturing (CM) is a manufacturing system in which similar parts and their required machines are grouped into manufacturing cells. The implementation of CM systems leads to increased output, decreased setup time, reduced work-in-process, reduced material handling cost as well as improved system productivity. One problem in the design of CM systems is cell formation (CF). Solving the CF problem in CM systems may lead to the organization or re-organization of manufacturing systems into manufacturing cells and to the determination of the type and number of machines required in each manufacturing cell. Many models have been developed to solve the CF problem over the last two decades. A thorough literature review reveals that most models use an indirect index such as a similarity or dissimilarity index as an objective function. The use of an indirect measure may not reflect the true state of CM systems. In this thesis, a CM system design model that uses a direct measure "productivity index" is presented. First, a 0-1 integer-programming model that maximizes the ratio of the output to the total material handling cost is developed to form part families and machine groups simultaneously. Second, a simulated annealing (SA) algorithm is developed to solve large-scale problems. This algorithm provides several advantages over some of the existing algorithms. It forms part families and machine groups simultaneously and considers production volume, selling price, and maximum number of machines in each cell. Moreover, it has the ability to determine the optimum number of manufacturing cells; so there is no need to specify the number of manufacturing cells in advance. Several problems selected from the literature are used to test the performance of the developed models. The results showed the superiority of the SA algorithm over the integer-programming model in both productivity and computational time. Furthermore, the majority of the existing models assume that each part has a unique process plan. In real manufacturing systems, however, a part can be produced using different routes and machines, which improves the productivity of CM systems. Hence, the developed models are extended to consider alternative process plans. Source: Dissertation Abstracts International, Volume: 61-09, Section: B, page: 4915. Adviser: S. Taboun. Thesis (Ph.D.)--University of Windsor (Canada), 2000.
Abduelmola, Abduelghani I., "Modeling of cellular manufacturing systems with productivity consideration: A simulated annealing algorithm." (2000). Electronic Theses and Dissertations. 2753.