Ordering Policy and Inventory Classification Using Temporal Association Rule Mining

Ordering Policy and Inventory Classification Using Temporal Association Rule Mining

Reshu Agarwal (Banasthali University, Vanasthali, India)
DOI: 10.4018/IJPMAT.2018010103
OnDemand PDF Download:
List Price: $37.50


A modified framework that applies temporal association rule mining to inventory management is proposed in this article. The ordering policy of frequent items is determined and inventory is classified based on loss rule. This helps inventory managers to determine optimum order quantity of frequent items together with the most profitable item in each time-span. An example is illustrated to validate the results.
Article Preview


In any industry today, inventory optimization is considered a vital function. Excess and shortage of inventory in all levels of the supply chain can affect the availability of products and/or services to consumers. Several monitoring systems and processes can be employed to check inventory imbalances to minimize the supply and demand dynamics. To simply this 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, Pyke, & Peterson, 1988).

However, traditional ABC analysis is based on only a single measurement such as annual dollar usage. Multiple criteria can be used for classification of inventories including lead time, criticality, commonality, obsolescence and substitutability criteria, and so forth (Cohen & Ernst, 1988; Chase, Aquilano, & Jacobs, 1998). Furthermore, criteria such as the cross-selling effect as defined by Anand, Hughes, Bell, and Patrick (1997) should also be considered when classifying inventory items. A considerable number of researchers have studied the dependency of items using data mining. Various methods of data mining such as item-set mining, mining sequential pattern, association rule, classification, prediction, clustering, and regression are applied for the above areas (Han & Kamber, 2006).

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, Imielinski, & Swami, 1993). The popular approaches are Apriori and subsequent apriori-like algorithms and pattern growth methods (Agarwal, Aggarwal, & Prasad, 2001). These approaches assume all items as binary variables considering whether they are consumed or bought or not. Few researchers proposed extension of item-set mining called item-set mining with quantities considering the consumption of item along with quantity.

Recently, temporal data mining has become a core technical data processing technique to deal with changing data. Temporal association rules are an interesting extension to association rules by including a temporal dimension (Li, Ning, Wang, & Jajodia, 2001; Lee, Chen, & Lin, 2003). Kleinberg, Papadimitriou, and Raghavan (1998) developed a microeconomic framework in which the influence of associations between items could be considered for maximising the expected profit. This influence reflects the cross-selling effect between different items. Brijs, Swinnen, Vanhoof, and Wets (1999) used association rules for selecting items considering relationships among retail items by discovering frequent items-sets and discovered the profitability per set of items by identifying the cross-sales effect of product items and using this information for better product selection.

This model was later extended by Brijs, Goethals, Swinnen, Vanhoof, and Wets (2000) to enable retailers to add category restrictions. However, there is little research on how to maximize profit when the environment changes. Wong, Fu, and Wang (2005) followed their work and proposed a methodology for suggesting recommendations from analysis of item-sets and the application of the concepts of association rules and selection of maximal profit item are investigated considering cross-selling effect. Bala (2008, 2012) suggested a model for making use of consumer insight information for inventory management in retail stores. Later a study on purchase dependence association rules for retail products was suggested by Bala, Sural, and Banerjee (2010) to make inventory replenishment decisions.

As per their observation, in a multi-item retail inventory of very large number of items, purchase dependence among the items is observed frequently and when there is stock out of one item, it may result in the decline in purchase of another item. Utilizing the concept of ‘purchase dependence”, Park and Seo (2013) developed a multi-item inventory control considering dependency based on availability of items demanded over the same customer order. The authors derived expressions for additional average cost of lost sales by other items when one particular item is out of stock.

Complete Article List

Search this Journal:
Open Access Articles: Forthcoming
Volume 6: 2 Issues (2018): 1 Released, 1 Forthcoming
Volume 5: 2 Issues (2017)
Volume 4: 2 Issues (2016)
Volume 3: 2 Issues (2015)
Volume 2: 2 Issues (2014)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing