سیستم های اطلاعات جغرافیایی در تصمیم گیری انتخاب مکان انبار

The warehouse site selection decision is not merely the question of choosing sites. It involves the comparison of the spatial characteristics of a market with the overall corporate and marketing goals of the firm. A geographic information system-aided process to the warehouse site selection decision is presented and the use of the presented process is demonstrated with a practical example. Various factors likely to affect customer service and costs are defined and subsequently integrated into an overall evaluation.

The commitment to logistics and marketing management shown recently by organizations, has stimulated a heightened awareness of the importance of warehouse site selection decisions in the formulation and implementation of effective business strategies. The appropriate number and geographic sites of warehouses are determined by customer, manufacturing and competitor locations, product requirements, types of transportation, and sales level [1] and [2]. From a policy viewpoint, warehouses should be established in a logistical system if they can render service or cost advantages. Warehouse site location analysis, particularly with the increased availability of computer-based techniques, can provide invaluable information to assist warehouse and marketing management with their decision-making process. Computerized maps can display hidden relationships between customers or competitors and territories. “Geographic information system (GIS) is a group of procedures that provide data input, storage and retrieval, mapping and spatial analysis for both spatial and attribute data to support the decision-making activities of the organization” [3]. In this sense, GIS can support logistic and marketing managers to evaluate placement options for warehouse stores, based on costs (transport, labor, utilities, etc.), and on customers and competitors demographics. Typical location analysis problems can be characterized as very complex and data intense. Modeling and analysis techniques must be employed to effectively deal with such complexity and data intensity in order to identify the best alternatives. According to Ballou and Masters [4] a number of commercial software products exist that are specifically designed to address the location analysis problem. A variety of different approaches and models have been suggested to assist in the evaluation and measurement of location decisions. Decision support systems (DSS) incorporate both data and models. They couple the intellectual resources of individuals with the capabilities of the computer to improve the quality of decisions and to support managerial judgment. The aim of this paper is to develop a geographic decision support system for the warehouse site selection process, enabling the manager to use quantitative and qualitative criteria in order to classify alternative warehouses or visualize the best one.

Decisions on warehouse location can be sup- portedbyavarietyofmodelsbasedonquantitative and qualitative factors. A GDSS for the evaluation of warehouse location was developed in order to support the managers decision. The model allows the user to input unlisted factors which are considered most relevant in a particular situation, using information available from previous aca- demic and empirical work, as well as his own ex- perience. Numerical values are grouped by the managerintofourclassesinorderforallthefactors to be scored on the same scale (0 } 4), according to the user ' s needs. These evaluated factors are used formodelapplications.Usingpairwisecomparison, the weights of the above factors are calculated. ` What if a scenarios can by applied since the alter- native sites can be seen on one screen and the elasticityoffactorweightsandratingscanbeeasily tested. One important advantage of the model is that, once the relevant factors and their weightings have been set up, the model can be used for site assessmentbypersonnelwhohavelittleknowledgeof site location theory. This model could be ex- tended to allow the input of more quantitative models, the results of which can be assessed by the managers.