Worldwide competition in global economies has posed significant challenges to companies wanting to fulfill the continuously changing requirements of market place. Supply Chain Management (SCM) attempts to reduce supply chain risks and uncertainty thereby improving customer service and optimizing inventory levels, business processes and cycle times, resulting in competitiveness, customer satisfaction and profitability as cited by Van Der Zee and Van Der Vorst (2005). In SCM, determining optimal facility location is a very crucial problem. To optimize logistical network configuration such as factories, warehouses, Distribution Centers (DC), and retail outlets requires locations that maximize supply chain performance as quoted by Simchi-Levi et al. (2003).
Facility Location Selection (FLS) is a nonrecurring, cross-functional decision-making problem. Location selection is a problem faced not only by entrepreneurs but also by designers, city planners and administrators. The location selection decision has been treated as a three-stage process in literature, where Stage 1 deals with selection of a geographical region or state; Stage 2 deals with the selection of a particular locality; and Stage 3 deals with the selection of the final site. Thus, location selection is a
multi-stage, multi-decision problem where the choice of location in each stage is made such that it directs the final decision of location to be either equal to or as close as possible to the global optimum.
FLS factors are stated and classified as quantitative and qualitative. Numerous attributes of facility location must be considered during location selection process such as quantitative criteria like material costs, labor costs, etc., and qualitative factors like political environment, proximity to markets and customers, etc. The general procedure for making location decision usually consists of the following steps: decide on the criteria that will be used to evaluate location alternatives; select the criteria that are important; develop location alternatives; and select the alternatives by evaluation. There are many approaches available to select the best location, and cluster analysis is one of the recent techniques that is a very useful classification tool as cited by Ye et al. (2011). Cluster analysis, which divides data into groups that is meaningful, useful, or both, has long played an important role in a wide variety of fields such as, psychology, biology, statistics, pattern recognition, information retrieval, machine learning and data mining. Kaufman and Rousseew (1990) report that clustering algorithm is ideally suited for locating similarity in large and manageable data, particularly in the absence of prior information about the data. Also, unsupervised algorithms are found to be a better tool for dealing with incompleteness and ambiguity in data which is an acceptably common case in the problem of location selection as described by Jain and Dubes (1998). Sergo et al. (2007) used clustering analysis in a capacitated location-routing problem and confirmed the potentialities of investigation opportunity for applying two distinct scientific areas, cluster analysis and operational research. Further, Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method is another emerging approach for multi-criteria decision making for evaluation of alternatives.
|
