Learning Effects in Cell Manufacturing

Abstract

Manufacturing cells offer the promise of short lead times and lower costs. In order to fulfill that promise, both technical and human factors must be engineered properly. We report on the experience of a large automobile components manufacturing plant in its initial efforts to implement cell manufacturing. We present an alternative to classic volume-based models of manufacturing learning effects. The scrap costs experienced in six manufacturing cells over a complete model year form the test dataset for model evaluation. The new model shows a better fit to the data. Various enhancements of performance measurement and reporting systems for manufacturing in general, and cells in particular, are discussed.