Content area
Abstract
This research introduces a two-phase methodology to design a Cellular Manufacturing System. In the first phase, a similarity coefficient formula is utilized to measure the similarity or relationship between the components. The result is used to form a graph in which each node represents a component and the weight on the edge between any two nodes (components) indicates the similarity between the components.
A heuristic algorithm is developed to identify the homogeneous subgraphs of the formed graph, in which, each subgraph represents a component family. Subgraphs are identified by a new edge-cutting rule which is based on the average value of the weights of cut edges.
In the second phase, the result of the first phase is used to form the machine groups (cells) such that a cost function is minimized. Two Integer Programming models are introduced to form the machine groups. An example is used to demonstrate the procedure.
