Presenting a Data-Driven Succession Planning Model in Iranian Public Organizations Using Neural Networks and Artificial Intelligence
DOI:
https://doi.org/10.63053/ijmea.91Keywords:
Data-Driven Succession Planning, Neural Networks, Artificial Intelligence, Public Administration, Machine Learning, Organizational Succession, XGBoostAbstract
Introduction: Succession planning in Iranian public organizations faces multiple structural challenges, including lack of transparency, human resource blind spots, and the absence of an objective evaluation system. The present study aimed to design a data-driven succession planning model based on neural networks and artificial intelligence in Iranian public organizations.
Methodology: This study is applied-developmental in terms of purpose and mixed (sequential exploratory) in terms of method. In the qualitative phase, using the meta-synthesis method, 85 domestic and foreign sources (2020–2026) were reviewed to identify the key components of succession planning. In the quantitative phase, data from 12,487 employees of 7 Iranian public organizations (in the period 1395–1403) were collected and analyzed using machine learning algorithms including Random Forest, XGBoost, and Deep Neural Network (DNN). To address class imbalance, the SMOTE method was used. Model validation was performed using 5-fold cross-validation.
Findings: The qualitative phase led to the identification of 7 main components affecting succession planning: (1) individual competencies, (2) performance records, (3) demographic characteristics, (4) career path, (5) organizational structural factors, (6) retention indices, and (7) leadership competencies. Quantitative results showed that the XGBoost model with F1-Score = 0.512 and AUC = 0.870 had the highest predictive power in identifying employees eligible for managerial positions. The deep neural network with F1-Score = 0.493 performed close to XGBoost, but the optimized logistic regression model with F1-Score = 0.472, due to its transparency and higher auditability, is recommended for use in the public sector. Feature importance analysis showed that "departmental membership," "long-term performance," and "tenure" are the most important predictors of promotion. The model fairness test showed that the model acts fairly with respect to gender (equal opportunity).
Conclusion: The model presented in this study can serve as an objective, transparent, and data-driven framework for succession planning in Iranian public organizations
References
References
Persian References
• Heidari, A., & Nikbakht, M. (1402). Designing a human resource planning model in Iranian public organizations. Biannual Journal of Public Administration, 14(1), 89-118.
• Rahmani, M., Hosseini, S., & Karimi, B. (1402). The role of middle managers in implementing succession planning strategies. Iranian Journal of Management Sciences, 17(68), 45-68.
• Rezaei, F., & Mousavi, S. (1402). Designing a good governance model in municipalities of Iranian metropolises. Organizational Resource Management Research, 13(2), 23-50.
• Zare, R., & Karimi, Z. (1403). Challenges of implementing succession planning in Iranian public organizations. Quarterly Journal of Public Organizations Management, 11(4), 73-94.
• Karimian, J., Rezaei, N., & Ghadiri, B. (1403). Lived experience of managers in succession planning challenges in knowledge-based companies: A phenomenological study. Journal of Development Studies, 21(68), 25-48.
English References
• ABDULRAHIM, M. K. A. M. (2025). Optimizing Human Resource Decisions: Predicting Promotions Using Data Analytics [Master's thesis, Rochester Institute of Technology].
• Ademeso, T. S., Salisu, I. I., Ibrahim, Y. S., & Rufus, P. C. (2025). Artificial Intelligence in Public Sector Human Capital Development: Opportunities, Risks, and Policy Implications. University of Abuja Repository.
• Farrell, R. (2025). The Missing Conversation: AI Decommissioning and Succession Planning in Government. GovLoop.
• Jackley, M. (2024). Top HR Challenges and Trends in the Public Sector. Oracle.
• NSW Government Workforce Intelligence. (2025). AI-powered analysis of workforce capability needs. TalentPathAI.
• r4 Technologies. (2026). Real-Time Workforce Readiness and Skills Orchestration for Government Agencies. r4.ai.
• SPC RAS & RANEPA. (2025). Software for analyzing the career trajectory of managers and officials. St. Petersburg Federal Research Center.
• TechX Media. (2025). Abu Dhabi Study Reveals AI-Powered Hiring Trends. TECHx Media.
• UK Central Digital and Data Office. (2024). Succession Select Tool: Algorithmic Transparency Recording Standard. PublicTechnology.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.
License Terms
The International Journal of Applied Research in Management, Economics and Accounting (IJMEA) publishes its articles under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
This license promotes the broad dissemination and reuse of published research while ensuring that appropriate credit is given to the original authors.
Creative Commons Attribution 4.0 International License
Unless otherwise stated, all articles published by IJMEA are licensed under the CC BY 4.0 License.
Under this license, users are permitted to:
-
Share — copy and redistribute the material in any medium or format.
-
Adapt — remix, transform, and build upon the material for any purpose, including commercial purposes.
These permissions are granted provided that appropriate attribution is given to the original authors and the source of the publication.
Attribution Requirements
When reusing or adapting an IJMEA article, users should:
-
Give appropriate credit to the author(s).
-
Provide the title of the article.
-
Identify IJMEA as the original publication venue.
-
Provide a link to the applicable Creative Commons license.
-
Indicate whether changes were made to the original material.
-
Avoid suggesting that the original authors or IJMEA endorse the reuse.
An appropriate attribution may include:
"Author(s), [Article Title], International Journal of Applied Research in Management, Economics and Accounting (IJMEA), [Year]. Licensed under CC BY 4.0."
Copyright
Copyright remains with the author(s) of the published work.
By submitting a manuscript to IJMEA and upon acceptance for publication, authors grant the journal the non-exclusive rights necessary to publish, reproduce, distribute, archive, and make the article publicly available.
The CC BY 4.0 license allows others to reuse the published work in accordance with the terms of the license while preserving the authors' copyright.
Author Rights
Authors retain the right to:
-
Share and distribute their published work.
-
Deposit the article in institutional or subject repositories.
-
Use the article for educational and research purposes.
-
Include the article in a thesis or dissertation.
-
Create derivative works based on the article.
-
Enter into separate non-exclusive agreements for additional distribution of the work, provided that the original publication by IJMEA is properly acknowledged.
Third-Party Material
Authors are responsible for ensuring that any third-party material included in their manuscript is appropriately credited and can legally be published under the applicable license.
Material that is not covered by the CC BY 4.0 license should be clearly identified, where applicable, and may require separate permission from the copyright holder.
License Statement
The following statement should appear on published articles:
© [Year] The Author(s). This article is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
For more information about the license, readers may consult the official Creative Commons Attribution 4.0 International License terms.
Transparency and Open Access
IJMEA is committed to making scholarly research openly accessible while respecting authors' copyright and promoting responsible reuse of published academic work.
The application of the CC BY 4.0 license supports the journal's open-access policy and enables researchers, educators, institutions, and other users to access, share, and build upon published research with appropriate attribution.
