Enhancing Fairness and Efficiency in Teacher Placement based on Staff Placement Model: An Intelligent Teacher Placement Selection Model for Ministry of Education Malaysia

Shamsul Saniron, Zulaiha Ali Othman, Abdul Razak Hamdan
dx.doi.org/10.17576/apjitm-2025-1401-09

Abstract

Efficient workforce management is critical for large organizations such as the Ministry of Education Malaysia (MOE), particularly in processing thousands of teacher placement applications. The current Teacher Placement Selection (TPS) system considers personal merit, environmental factors, and staffing requirements across 20 attributes but has been criticized for its perceived lack of fairness. This study proposes an Intelligent Teacher Placement Selection (ITPS) system based on a Staff Placement Model (SPM), expanding the attribute set to 27 by incorporating personal, staffing position, placement type, and human factors to enhance decision-making fairness. The effectiveness of ITPS was evaluated using five machine learning algorithms: J48, Decision Tree, Na?ve Bayes, Random Forest, and K-Nearest Neighbors. A dataset of 4,484 teacher placement applications from the first session of 2020 was analyzed, comparing Teachers Placement Committee (TPC) and ITPS outcomes. Results indicate that J48 achieved the highest accuracy, improving from 71.28% in TPC to 95.74% in ITPS. Further validation using 2,562 applications from the second session of 2020 demonstrated ITPS ability to approve more placements (1,582) than the actual decisions (1,560). A T-test comparing TPC and ITPS yielded a high p-value (p > 0.05), confirming no statistically significant difference between the models while validating ITPS reliability. In conclusion, the ITPS model based on SPM enhances efficiency and fairness in teacher placements at MOE, demonstrating strong potential as a decision-support tool for optimizing workforce allocation.

keyword

Teacher placement, Staff Placement Model, classification techniques, data mining, J48

Area

Information Systems