Designing Animal Market Layout by Considering Consumer Purchase Behaviors

Abstract

Sales transaction data contains rich information and can support company competitiveness. However, this transaction data is initially unstructured and needs to be processed into insight for the company's decision-making. Market Basket Analysis (MBA) is a data mining technique that can be used to study consumer purchasing patterns. This paper presents a case study on using an MBA to obtain consumer buying behaviors where the result of the MBA is then used to design a proposed layout. The animal market governed by Yogyakarta Province, known as Pasty Market, was used as a case study. Pasty Market is an animal trading center with around 30,000 square meters area and 255 sellers that sell various kinds of animals such as songbirds, dove birds, rabbits, cats, dogs, iguanas, turtles, ornamental chickens, and ornamental fish as well as animal food and cage. With this enormous area and number of merchants, the layout of Pasty Market becomes crucial in customer satisfaction. Association rules result in four priority levels in proposed layout planning, where these rules are used to determine the proximity among items in the proposed layout. These four levels of priority, ordered by the confidence value, are (1) “songbirds” and “bird food” (confidence value 91%), (2) “ornamental fish” and “turtles” (confidence value 80-90%), (3) “birdcages” and “songbirds” (confidence value 70-80%), and (4) “cats” and “dog” as well as “birdcages” and “birds” (confidence value 50-70%). Association rules were then used as the basis for determining the proximity value between merchants, where the proximity rules were then used for designing a proposed layout.