Improvement in shop floor management using ANN coupled with VSM: A case study.

In the present article, the authors have employed the Value Stream Mapping (VSM) technique for the existing shop floor process in an earthmoving equipment manufacturing unit. Thereafter, an Artificial Neural Network (ANN)-based information processing technique has been used for generating a prediction model of shop floor management. For developing the ANN-based prediction model, the production time involved in different processes has been collected for 31 working days. This collected data has been used for training and testing the ANN model. Thereafter, to validate the developed ANN model, more 7 days data has been collected and compared with the predicted values of model for the same input attributes. From the results, it has been found that the performance of the developed model is highly adequate for the prediction purpose with the MSE and MAE values for training data and testing data as 0.0008105545 and 0.0000008979 and 0.01012315 and 0.0001658978, respectively. Based on the acquired results it is evident that the proposed methodology may be significant in predicting the production time of the anticipated shop floor. [ABSTRACT FROM AUTHOR]