IME department organized a seminar on “Market Basket Rules as Input to Improve Inventory Performance” and invited Dr. Pradip Kumar Bala to deliver his valuable lecture on it. Dr. Bala is associate professor at IIT Roorkee in Department of Management studies. Dr. Bala introduced the association rules, also known as market basket rules, in the context of retail items.
Market basket rules have been widely used in marketing, shelf allocation in retail stores, web designing in commerce etc; however, according to him, the works on applications of these rules in the area of operations management have been limited so far. He mentioned that recently a few works with applications of market basket rules in inventory management have been applied. According to him, the knowledge of correlated items in the form of association rules can be of tremendous importance in designing inventory replenishment policies and hence, it must be accounted for. The research incorporated in his current work makes an attempt to account for demand correlation at transaction level in retail inventory management. The cost of joint replenishment for correlated items is often found to be much less than the total cost of individual item replenishment. However, just the knowledge of the names of items with correlated demand, and a type of relationship are not always sufficient. We need to know the probabilities or probability distribution of joint demand. When the number of items increases, the complexity of the demand relationship among the items increases dramatically, making it infeasible to calculate all probabilities of joint demand in advance.
To overcome this, he suggested to make use of the association rules which are mined from the past transaction data. The impact of association rules can be used to compute the profitability measure which forms the basis of grouping items with high probability of purchase in the same group. Replenishment policy can be designed for each group separately with higher profitability of inventory operation. The research suggests a measure for profitability index and an algorithm for grouping the items based on the profitability index.
Finally the seminar was wrapped up with queries from faculty and students of the department, which churned out some interesting perspectives and left the attendees with many issues to mull over.
Contributed by:
Vivek Agarwal
MBA, 2011-13
IIT Kanpur
Market basket rules have been widely used in marketing, shelf allocation in retail stores, web designing in commerce etc; however, according to him, the works on applications of these rules in the area of operations management have been limited so far. He mentioned that recently a few works with applications of market basket rules in inventory management have been applied. According to him, the knowledge of correlated items in the form of association rules can be of tremendous importance in designing inventory replenishment policies and hence, it must be accounted for. The research incorporated in his current work makes an attempt to account for demand correlation at transaction level in retail inventory management. The cost of joint replenishment for correlated items is often found to be much less than the total cost of individual item replenishment. However, just the knowledge of the names of items with correlated demand, and a type of relationship are not always sufficient. We need to know the probabilities or probability distribution of joint demand. When the number of items increases, the complexity of the demand relationship among the items increases dramatically, making it infeasible to calculate all probabilities of joint demand in advance.
To overcome this, he suggested to make use of the association rules which are mined from the past transaction data. The impact of association rules can be used to compute the profitability measure which forms the basis of grouping items with high probability of purchase in the same group. Replenishment policy can be designed for each group separately with higher profitability of inventory operation. The research suggests a measure for profitability index and an algorithm for grouping the items based on the profitability index.
Finally the seminar was wrapped up with queries from faculty and students of the department, which churned out some interesting perspectives and left the attendees with many issues to mull over.
Contributed by:
Vivek Agarwal
MBA, 2011-13
IIT Kanpur
Comments
Post a Comment