Title: Data mining techniques and mathematical models for the optimal problem at a state public university

Authors: Lijian Xiao; Shuai Wang; Xinhui Zhang

Addresses: Department of Biomedical, Industrial and Human Factors Engineering, Wright State University, 207 Russ Engineering Center, 3640 Colonel Glenn Hwy, Dayton, OH 45435, USA ' Department of Biomedical, Industrial and Human Factors Engineering, Wright State University, 207 Russ Engineering Center, 3640 Colonel Glenn Hwy, Dayton, OH 45435, USA ' Department of Biomedical, Industrial and Human Factors Engineering, Wright State University, 207 Russ Engineering Center, 3640 Colonel Glenn Hwy, Dayton, OH 45435, USA

Abstract: This paper studies the optimal allocation problem of financial aid: the allocation of the appropriate levels of scholarships to the correct students, as observed in a state university. This research applies data mining techniques and mathematical models to solve the optimal financial aid allocation problems in three steps. First, data mining techniques, such as logistic regression, are used to determine the matriculation and graduation probabilities associated with students from various socioeconomic backgrounds and given levels of scholarship. Second, based on the responses to the different scholarship levels, an integer programming model is developed to maximise revenue over the students' course of study. Third, decision tree and piecewise linear regression methods are employed to transform the results from the optimisation model into effective policies for implementation. This research has led to a scholarship redesign, a straightforward scholarship award policy, based on a composite GPA and ACT score, been implemented.

Keywords: financial aid allocation; optimisation; data mining; logistic regression; integer programming; decision tree.

DOI: 10.1504/IJOR.2025.147786

International Journal of Operational Research, 2025 Vol.53 No.4, pp.499 - 524

Received: 08 May 2023
Accepted: 28 Jun 2023
Published online: 01 Aug 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article