Prediction of machine reconfigurability using artificial neural network for a reconfigurable serial product flow line

Reconfigurable machines (RMs) are considered to be one of the vital elements of modern manufacturing systems like reconfigurable manufacturing systems (RMSs). These machines offered customised flexibility in terms of capacity and functionality. Reconfigurable machines are assembled using some basic/essential modules and auxiliary modules. The RMTs can be reconfigured into several other configurations for variable functionality and capacity by keeping its base modules and just adding/removing or adjusting the auxiliary modules. Measuring machine reconfigurability may be considered as one of the important challenge in assessing the performance of these manufacturing systems. In the present paper, an artificial neural network model has been proposed for quantitative assessment of reconfigurability values of RMs on the product flow line. The data is generated using a developed mathematical model based on multi attribute utility theory. The ANN predictive model could thus provide a flexible and objective framework for manufacturers to evaluate reconfigurability of machines for a given product flow line. The developed approach has been demonstrated using a multi stage serial reconfigurable product flow line.