Spatial-temporal analysis using two-stage clustering and GIS-based MCDM to identify potential market regions
Promotion is essential in a competitive environment. Promotion to the right areas increases success and saves resources. However, due to Indonesia's vast territory and numerous regions of origin school, universities with student markets from all over the country must select target areas for pro...
Saved in:
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Success Culture Press
2021
|
Online Access: | http://eprints.utem.edu.my/id/eprint/26607/2/SPATIAL-TEMPORAL%20ANALYSIS%20USING%20TWO-STAGE.PDF http://eprints.utem.edu.my/id/eprint/26607/ http://www.aasmr.org/jsms/Vol11/vol.11.4.5.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Promotion is essential in a competitive environment. Promotion to the right areas increases success and saves resources. However, due to Indonesia's vast territory and numerous regions of origin school, universities with student markets from all over the country must select target areas for promotion to meet their objectives and save resources. Unlike for-profit businesses, besides quantity factors, educational institutions need to consider student quality factors in selecting promotional locations. This study aims to conduct a data-driven spatio-temporal analysis to identify potential regions for university promotions targets. This study uses enrollment and academic data from one private university in Indonesia for the empirical study. In Geographic Information System (GIS) environment, the origin schools' locations were geocoded, and various thematic maps were analyzed. This study integrates two-stage clustering and GIS-based multi-criteria decision-making (MCDM) to identify potential market regions. A potential region is one that consistently sends many qualified students. First, time-series clustering is conducted to groups regencies/cities based on the enrolled students' patterns over time in the university. Subsequently, the origin schools' regencies/cities were clustered using the k- prototypes algorithm based on their time-series pattern category, the consistency in sending students, average cumulative grade point average (CGPA), and dropout (DO) rate. The clusters are scored using the sum weighting method. The highest valued cluster that consists of eight regencies and 18 cities that consistently contributed high quantity and quality students were selected as the priority regions. The proposed approach's results were compared to the Simple Additive Weighting (SAW) and Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) methods for evaluation. The proposed method can assist the university management in determining potential regions for promotion purposes. |
---|