Ordering policy in a supply chain with adaptive neuro-fuzzy inference system demand forecasting

Determining ordering policy has incisive impacts on the success or letdown of an organization. This research has considered reliability while developing a method for finding ordering policy for multiple supply chain stages through optimal lot sizing. Setup cost, production cost, inspection cost, rejection cost, interest and depreciation cost, holding cost, etc. are considered for each supply chain stage whereas the demand inputs in the costs are taken from an adaptive neuro-fuzzy inference system generated forecasting method. Later, a genetic algorithm has been applied to find the optimum lot size at multiple levels of supply chain network to minimize total cost. Optimal lot size, reliability and total cost are determined and the costs are accumulated to determine total minimum supply chain cost. To validate the model, a comparison with the current situation clearly indicates the superiority of proposed model over the usual company approach to ordering policy.