C Rajendran

Transportation of goods in a supply chain from plants to customers through distribution centers (DCs) is modeled as a two-stage distribution problem in the literature. In this paper we propose genetic algorithms to solve a two-stage transportation problem with two different scenarios. The first scenario considers the per-unit transportation cost and the fixed cost associated with a route, coupled with unlimited capacity at every DC. The second scenario considers the opening cost of a distribution center, per-unit transportation cost from a given plant to a given DC and the per-unit transportation cost from the DC to a customer. Subsequently, an attempt is made to represent the two-stage fixed-charge transportation problem (Scenario-1) as a single-stage fixed-charge transportation problem and solve the resulting problem using our genetic algorithm. Many benchmark problem instances are solved using the proposed genetic algorithms and performances of these algorithms are compared with the respective best existing algorithms for the two scenarios. The results from computational experiments show that the proposed algorithms yield better solutions than the respective best existing algorithms for the two scenarios under consideration. © 2011 Elsevier Ltd.

We consider a static divergent two-stage supply chain with one distributor and many retailers. The unsatisfied demands at the retailers' end are treated as lost sales, whereas the unsatisfied demand is assumed to be backlogged at the distributor. The distributor uses an inventory rationing mechanism to distribute the available on-hand inventory among the retailers, when the sum of demands from the retailers is greater than the on-hand inventory at the distributor. The present study aims at determining the best installation inventory control-policy or order-policy parameters such as the base-stock levels and review periods, and inventory rationing quantities, with the objective of minimizing the total supply chain costs (TSCC) consisting of holding costs, shortage costs and review costs in the supply chain over a finite planning horizon. An exact solution procedure involving a mathematical programming model is developed to determine the optimum TSCC, base-stock levels, review periods and inventory rationing quantities (in the class of periodic review, order-up-to S policy) for the supply chain model under study. On account of the computational complexity involved in optimally solving problems over a large finite time horizon, a genetic algorithm (GA) based heuristic methodology is presented. © 2010 Elsevier Ltd.

This paper addresses the development of an inventory control mechanism and inventory rationing policies in a static divergent two-stage supply chain consisting of one single distributor and several retailers. The unsatisfied demand is assumed to be backlogged at both distributor's and retailers' ends. In the case of shortage at distributor, the available stock on hand is rationed among the retailers. Most of the studies in the literature treat ordering costs as negligible and assume the review period to be one unit of time. However, if there is a significant cost associated with the order placement, then the review period can be greater than one time unit. Hence, in this study, we consider ordering costs for retailers, and present a mathematical programming model which can give optimal base-stock levels and review periods and inventory rationing (in the class of periodic review, order-up-to S policy). A genetic algorithm-based heuristic algorithm is also presented for solving problems with a large time horizon. © 2010 Inderscience Enterprises Ltd.

A novel genetic algorithm (GA) is proposed in this paper for solving the machine-part cell formation problem in the presence of alternative process plans. Parent chromosomes with number of genes equal to the number of parts are generated based on the correlation value calculated using a ranking index. Crossover is carried out on the 60% of the parent chromosomes and mutation is carried out at the weakly correlated part (gene) in the chromosomes. Performance of the algorithm was tested using 20 test instances from the literature. The proposed genetic algorithm is superior in terms of solution quality for 10% of the total test instances and equal to the best solution achieved by the other algorithms for the remaining 90% of the test instances.