Article Preview
TopIntroduction
Inventory is usually one of the biggest numbers on their balance sheet, therefore, effective inventory control and management is a vital function to help insure the continued success of distribution and manufacturing and companies. The effectiveness of inventory control is typically measured by how successful a company is at reducing inventory investment, meeting its customer service goals, and achieving maximum throughput and cost containment. Several monitoring systems and processes can be employed to check inventory imbalances to minimize the supply and demand dynamics. To simply these monitoring systems and process items/materials/products are classified into different groups. Traditionally, ABC analysis has been based on the criterion of dollar volume (Silver et al, 1988).
However, traditional ABC analysis is based on only single measurement such as annual dollar usage. It has been pointed out that other criteria can be important; among these are lead time, item criticality, durability, scarcity, reparability, stock ability, commonality, substitutability, the number of suppliers, mode and cost of transportation, the likelihood of obsolescence or spoilage, and batch quantities imposed by suppliers. Several methods have been developed to perform multi-criteria ABC analysis that can be quite easily implemented today. Multiple criteria can be used for classification of inventories including lead time, criticality, commonality, obsolescence and substitutability criteria etc. (Cohen & Ernst, 1988; Chase et al., 1998). However, the criteria such as cross-selling effect defined by Anand et al. (1997) should also be considered when classifying inventory items. Further, one of the most popular data mining techniques is association rule mining. The patterns discovered with this data mining technique can be represented in the form of association rules (Agrawal et al., 1993). Recently, clustering has become a core technical data processing technique to deal with similar data. Data mining adds to clustering the complications of very large datasets with very many attributes of different types. This imposes unique computational requirements on relevant clustering algorithms. A variety of algorithms have recently emerged that meet these requirements and were successfully applied to real-life data mining problems. Broder et al. (1997) defines clusters as maximal connected components of some pair-wise similarity of transactions, thus suffers from the breakdown of the transitivity of pair-wise similarity. Guha et al. (2000) proposed the common neighbors of two transactions as a measure of pair-wise similarity. Wang et al. (1999) method does not use any notion of pair-wise similarity. They cluster transactions that contain similar items. The difference is that clustering emphasizes the dissimilarity of clusters.