Main Article Content
In the procurement decision, the future price of the raw material plays an important role as it affects the budget and procurement plan. Inaccuracy in raw material price prediction affects the performance of procurement activity and budget plan. Low Ash Metallurgical (LAM) coke is one of the important raw materials used by various alloy manufacturing company. In India, this coke generally imported from the foreign market and followed a long-term contract with the supplier. The LAM coke price is highly volatile in nature. Hence, accurate forecasting of the LAM coke price is very important for any alloy manufacturing company to manage its budget for the procurement. The intelligent models such as Artificial Neural Network (ANN) and Adaptive Neuro-fuzzy Inference System (ANFIS) can learn the past pattern through their self-learning and adapting capability. In this study, the discrete wavelet transformation has been used to extract the past LAM coke price pattern and resultant series applied to train the ANN and ANFIS model to improve the prediction accuracy. From the study, it has found that the prediction error for both the model is less than 5%, hence those models can be used by the alloy industry to predict the LAM coke price.
Artificial Neural Network (ANN) Adaptive neuro-fuzzy inference system (ANFIS) Wavelet analysis Discrete wavelet transformation