Exact and heuristic methods of determining process parameters in a serial supply chain with lost sales and virtual backlog

Abstract

Presently, supply chain concept is a network of facilities with varying conflicting objectives, and the decision making is a complex process. The uncertainty in demand increases the complexity in inventory control mechanism. Backordering, partial backordering and lost-sales are considered in the inventory management to characterize the excess demand. In the present competitive scenario, most consumers have no patience to wait and show urgency to buy their goods failing which the management have to suffer a huge loss of supply chain members and hence the profit. There are very few research works regarding lost-sales and virtual backlog parameter in the area of multi-echelon inventory systems. A mathematical model was formulated in order to find the best base stock and review periods in the supply chain and this model could run only for a shorter time period. In order to overcome this, a Modified Gene-wise Genetic Algorithm (MGGA) supply chain model for a longer time frame which so far has not been applied in this process may help to determine best base stock levels and review periods with lost sales particularly at retailer end which minimized total supply chain cost. In this case the cost of lost sales is higher at the retailer end and in order to overcome this, another Modified Gene-wise Genetic Algorithm (MGGA) supply chain model was proposed for a longer time frame to minimize the cost of lost sales at the retailer end by considering virtual backlog at the retailer end. This model reduced the cost of lost sales at the retailer end to a greater extent but the total supply chain cost shows only marginal increase. Even though the total supply chain cost increases when compared to the previous model, the customer s satisfaction level is increased by fulfilling their demands. newline