Forecasting Capabilities of Spare Part Production with Artificial Neural Networks Model in a Supply Chain

Gaining the competitive advantage in every supply chain is dependent on the quality of the production and the desired cost. Therefore, Iran Khodro Company (IKCO) is critical need of decreasing extra costs through effective planning and cost management in order to stay in the competitive car market. The basis of each planning is knowledge and prediction. Due to high fluctuation of the amount of the production in Ikco over time, the use of Artificial Neural Networks is appropriate method to predict the amount of the production based on non-linear and complex relationships between inputs and outputs. Artificial Neural Network, In this study, extra parameters such as fuel consumption, machinery and labor (Assembly and Transport) were employed to assess their effect in improvement structure of ANN model and training performance of generated model. The monitoring data from 2004 to 2010 are designed to provide the requirements of training and testing the neural network. Finally, with respect to MAE, MARE, RMSE and R2, suitable models were selected for the study. After performing the mentioned model, correlation coefficient (R2) and mean absolute relative error (MARE) in neural network for test achieved were 0.947 and 0.03 respectively. Results point out that artificial neural network model has more advantages in comparison with traditional methods, in predicting the productions of IKCO.