.png){kind=icon}
.png){kind=icon}
Return
A data-driven decision support system for selecting managers based on their performance evaluations over time
.png){kind=icon}
PRE
AI
DOI：10.1016/j.engappai.2025.110874.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
One of the pillars of organizational success is having capable and efficient managers; therefore, selecting competent people for managerial positions can play a significant role in the future of the organization. This paper proposes a method for selecting managers within the organization using data-driven methods based on their performance-evaluation information over time. To do this, optimal function and suitable algorithm for aggregating the performance evaluation data are determined, and the individual's performance in each vacant management position is considered based on mean absolute error (MAE), mean squared error (MSE), and root mean square error (RMSE) measures. The proposed approach has been implemented in the Yazd University as a case study, which showed reliable results. Three managerial positions have been considered in this case study, i. e., 1) budget and planning manager, 2) financial and administrative affairs manager, and 3) student affairs manager. Among several rival aggregations and predicting algorithms, the data averaging aggregation method with gradient boosting machine algorithm showed the best results. Finally, the selected individuals have been assigned to each management position. The experts validated the output of the proposed model, and their satisfaction scores for the three managerial positions were 100 %, 80 %, and 80 %, respectively. The achievements of this research include identifying the best algorithm for predicting the performance of personnel in managerial positions, identifying proper people in the organization, and using the performance evaluation information to select capable managers.
Keywords:
Data mining
Machine learning
Manager selection
Performance evaluation
Job description
Performance forecasting
Journal
IF:
8
Papers:
5.4K
Citations:
3.5W