3d Geomarketing Segmentation: A Higher Spatial Dimension Planning Perspective

Abstract

ABSTRACT:Geomarketing is a discipline which uses geographic information in the process of planning and implementation of marketing activities. It can be used in any aspect of the marketing such as price, promotion or geo targeting. The analysis of geomarketing data use a huge data pool such as location residential areas, topography, it also analyzes demographic information such as age, genre, annual income and lifestyle. This information can help users to develop successful promotional campaigns in order to achieve marketing goals. One of the common activities in geomarketing is market segmentation. The segmentation clusters the data into several groups based on its geographic criteria. To refine the search operation during analysis, we proposed an approach to cluster the data using a clustering algorithm. However, with the huge data pool, overlap among clusters may happen and leads to inefficient analysis. Moreover, geomarketing is usually active in urban areas and requires clusters to be organized in a three-dimensional (3D) way (i.e. multi-level shop lots, residential apartments). This is a constraint with the current Geographic Information System (GIS) framework. To avoid this issue, we proposed a combination of market segmentation based on geographic criteria and clustering algorithm for 3D geomarketing data management. The proposed approach is capable in minimizing the overlap region during market segmentation. In this paper, geomarketing in urban area is used as a case study. Based on the case study, several locations of customers and stores in 3D are used in the test. The experiments demonstrated in this paper substantiated that the proposed approach is capable of minimizing overlapping segmentation and reducing repetitive data entries. The structure is also tested for retrieving the spatial records from the database. For marketing purposes, certain radius of point is used to analyzing marketing targets. Based on the presented tests in this paper, we strongly believe that the structure is capable in handling and managing huge pool of geomarketing data. For future outlook, this paper also discusses the possibilities of expanding the structure.

Within cluster pattern identification: A new approach for optimizing recycle point distribution to support policy implementation on waste management in Malaysia

“…k -Means can be considered the most popular clustering algorithm due to its simplicity. Previous studies have adopted the k -means algorithm for different applications, such as geo-marketing (Azri et al, 2016b), database organization (Azri et al, 2016a), wireless sensor framework (Azri et al, 2019) and location-allocation problem (Kim et al, 2018). However, the drawback of k -means is that it is a NP hard problem (García et al, 2018; Tîrnăucă et al, 2018; Wang et al, 2019).…”

Chicago area Chick-fil-A stores as an analogue to GTA site selection

“…The mapping of data can allow spatial trends to become more apparent to a retail practitioner, and the QSR sector has leveraged this information to gain insight into spatial market trends (Anselin, 2003;Church, 2002;Murray, 2010). GIS can help answer spatial questions concerning where customers live, their purchasing power, where competitors are located, and the potential impact on store market share (Aboulola, 2018;Suhaibah et al, 2016). Trade area creation models can be orchestrated using GIS programs such as ArcMap which produce quantitative results as well as visual maps which help practitioners more easily understand spatial trends (Aboulola, 2018;Bas and Gulersoy, 2018;Suhaibah et al, 2016).…”

“…GIS can help answer spatial questions concerning where customers live, their purchasing power, where competitors are located, and the potential impact on store market share (Aboulola, 2018;Suhaibah et al, 2016). Trade area creation models can be orchestrated using GIS programs such as ArcMap which produce quantitative results as well as visual maps which help practitioners more easily understand spatial trends (Aboulola, 2018;Bas and Gulersoy, 2018;Suhaibah et al, 2016).…”

Determination of the Representative Socioeconomic Level by BSA in the Mexican Republic

“…However, the process runs the risk of overlapping clusters, making geospatial analysis inefficient. Suhaibah et al (2016) use a variant of the k-means algorithm, called k-means ++; this variation prevents overlapping of the clusters, allowing the existence of boundary clusters. k-means ++ achieves an adequate initialization in a primary set of centers for k-means through a random seeding.…”