Dynamic grouping of parts in flexible manufacturing systems - a self-organizing neural networks approach

Artificial Intelligence (AI) has recently been recognized as a worthwhile tool for supporting manufacturing operations. This paper reviews AI-related approaches to Group Technology (GT) and presents the Self-Organizing Map (SOM) network, a special type of neural networks, as an intelligent tool for grouping parts and machines. SOM can learn from comples, multi-dimensional data and transform them into visually decipherable clusters. What sets this technique apart from others in GT is that SOM offers the flexibility of choosing from multiple grouping alternatives. SOM can be used in a dynamic situation where quick response to changes in part designs, process plans, or manufacturing conditions is essential, and thus it can be more easily integrated into a Flexible Manufacturing System. The paper proposes a framework of an intelligent system that integrates the neural networks approach and a knowledge-based system to provide decision supporting functions.