Design of Product Placement Layout in Retail Shop Using Market Basket Analysis

Only available on StudyMode
  • Download(s) : 281
  • Published : October 14, 2011
Open Document
Text Preview
MAKARA, TEKNOLOGI, VOL. 9, NO. 2, NOPEMBER 2005: 43-47

43

DESIGN OF PRODUCT PLACEMENT LAYOUT IN RETAIL SHOP USING MARKET BASKET ANALYSIS Isti Surjandari dan Annury Citra Seruni
Industrial Engineering Department, Faculty of Engineering, University of Indonesia, Depok 16424, Indonesia E-mail: surjandari.2@osu.edu, a_citra_s@yahoo.com

Abstract
Retailing is an industry with high level of competition. It is a customer-based industry which depends on how it could be aware of what the customers’ needs and requirements are. One technique most used in supermarkets is the mix merchandise. The purpose of this paper is to identify associated products, which then grouped in mix merchandise with the use of market basket analysis. This association between products then will be applied in the design layout of the product in the supermarket. The process of identifying the related products bought together in one transaction is done by using data mining technique. Apriori algorithm is chosen as a method in the data mining process. Using WEKA (Waikato Environment for Knowledge Analysis) software, the association rule between products is calculated. The results found five category association rules and fourteen sub-category association rules. These associations then will be interpreted as confidence and support to become consideration for the product layout. Keywords: layout, market basket analysis, retail

1. Introduction
Retail industry is a kind of business with high level of competition. The success of retail business is influenced by its fast response and its ability in understanding consumers’ behaviors. Retail business must focus to its consumer since retail business plays its role at the end of distribution channel. Consumer buying behaviors can be comprehended by observing how someone interacts and reacts to the marketing mix. According to Cohen (1991), company determines the decisions related to the 4P (Product, Place, Promotion, and Price) by focusing to its consumers; while each individual considers the option to buy which products under the psychological influences of culture, attitude, experience, previous usage of the products, and personal perception. Effects of both inputs (marketing and psychological) somewhat influences the customers to decide whether they will buy or not, where to buy, which brand to buy, and other choices. Aside of observing customers reactions to marketing mix, another way for understanding customers behavior is by using historical data, which is transactional data. From customer transactional

database, it can be observed customers’ shopping patterns which show associated categories or even associated products. The following study is guided by the hypothesis, that products chosen on a shopping trip in a supermarket are an indicator of the preference interdependencies between different products or brands. The bundle chosen on the trip can be regarded as an indicator of a global utility function. In other words, the existence of such a function implies a cross–category dependence of brand choice behavior. It is hypothesized, that the global utility function related to a product bundle is the result of the marketing–mix of the underlying brands. The objectives of this study were to discover the associated products, and to determine the proposed product placement layout based on the associated products. This study was carried out with several boundaries.

2. Method
This study used market basket analysis to find association rules between sets of items in transactional database and the Apriori algorithm is used in the data mining process. The goal of the association discovery

43

44

MAKARA, TEKNOLOGI, VOL. 9, NO. 2, NOPEMBER 2005: 43-47

was to find items that imply the presence of other items. For example, 90% of customers that purchase frozen pizza also buy soda. The problem of finding association rules was first introduced by Agrawal, et. al. [1], which is described as follows:...
tracking img