Abstract:
For retailers, selecting several products from a diverse and extensive range of products and the amount of space that should be allocated to them are very important decisions. The purpose of this research is to apply a data mining approach to find relationships between products from a large volume of sales financial transactions, product classification, and space allocation to each category of them. In this way, a model can be presented for product classification and space allocation. The statistical population of the research is the sales data of a store called Shaqayeq in Urmia city. The research sample is also one month of sales data in the time series of sales data. This data was obtained from the aforementioned store in November 2015. 525 shopping baskets or transactions were examined considering 79 types of products. As a result of analyzing this data, products were categorized into 10 different categories, some of which were placed in more than one category. By solving the profit function and obtaining volume increase coefficients, space was allocated to product categorization.
Machine summary:
Presenting a model for maximizing profit based on classification decisions Product and space allocation with a data mining approach Manouchehr Ansari 1, Ali Heidari 2, Ali Setareh Goran Abad 3 For retailers, choosing several products from a diverse and extensive range of products and the amount of space that should be allocated to them are very important decisions.
The aim of this research is to apply a data mining approach to find relationships between products from a very large volume of sales financial transactions, product classification, and space allocation to each category of them.
In this research, by applying data mining and using association rule mining, the information of sales transaction databases is examined, and through this information, product classification and space allocation decisions are made in a way that maximizes the store's profit.
The research model, while understanding the association rules in the data, should be able to optimize profit by considering product classification and space allocation.
This goal can be categorized into the following objectives: • Discovering association rules in sales transaction data for products; • Mathematical modeling for profit and solving that model; • Allocating space on shelves to product categories.
Therefore, the presented model is superior to models such as Anderson and Amato (1974) which performed product classification and space allocation with attraction, or Corstjens and Doyle (1981 and 1983), Borin and Farris (1994 and 1995), Zofriedman (1986) and Huang, Choi and Lee (2005) who all used space attraction.
(
, 
, 
)
