From Prediction to Strategy: Machine Learning and SWOT-TOWS Analysis for Structuring Lecturer Academic Workload Using Lecturer-Semester Data
DOI:
https://doi.org/10.58812/wsbm.v4i03.3113Keywords:
Academic Workload, Lecturer Supervision, Machine Learning, SWOT-TOWS, XGBoostAbstract
Extended undergraduate study duration beyond the normative eight semesters is a significant concern in higher education management and may be associated with lecturers’ academic workloads. This study analyzes teaching and supervisory workload patterns, compares machine learning regression models for predicting students’ average ongoing study duration, and formulates workload management strategies using SWOT–TOWS analysis. Using the CRISP-DM approach and secondary data from Universitas Padjadjaran’s Integrated Academic Information System (n = 10,079 lecturer-semester observations), four algorithms—Linear Regression, Random Forest, Gradient Boosted Trees, and XGBoost—were evaluated using a 90:10 hold-out scheme. XGBoost performed best on three of five metrics (RMSE = 1.224; correlation = 0.658; R² = 0.433). Supervisory workload was the most influential predictor. The findings informed four strategies: predictive workload monitoring, discipline-sensitive supervision thresholds, gradual model implementation, and adaptive workload allocation guidelines. The study contributes a lecturer-centered predictive perspective and provides an analytical basis for academic workload management.
References
[1] J. Kenny and A. E. Fluck, “Life at the academic coalface: Validation of a holistic academic workload estimation tool,” High. Educ., vol. 86, no. 3, pp. 507–526, 2023.
[2] M. Syaifulloh, D. R. Triputra, S. B. Riono, H. Mulyanto, A. Wulandari, and S. N. Utami, “Academic productivity of private university lecturers: Analysis of the impact of workload, motivation, and institutional support,” J. La Edusci, vol. 6, no. 5, pp. 1020–1039, 2025.
[3] M. Naidoo-Chetty and M. Du Plessis, “Job demands and job resources of academics in higher education,” Front. Psychol., vol. 12, p. 631171, 2021.
[4] E. Demerouti, A. B. Bakker, F. Nachreiner, and W. B. Schaufeli, “The job demands-resources model of burnout.,” J. Appl. Psychol., vol. 86, no. 3, p. 499, 2001.
[5] M. Rahma, E. D. Laksmi, and Y. Febrianti, “Balancing Workload and Thesis Guidance: How Supervisors Provide Effective Feedback to Students,” Scope J. English Lang. Teach., vol. 9, no. 2, pp. 785–796, 2025.
[6] M. Abrar, F. Failasofah, S. Naomi, N. Fajaryani, and M. Masbirorotni, “Supervising in Silence: Challenges and Coping Strategies in EFL Undergraduate Thesis Supervision in Indonesia,” AL-ISHLAH J. Pendidik., vol. 17, no. 4, pp. 7136–7151, 2025.
[7] A. Syafi’i, A. Munir, S. Anam, and S. Suhartono, “Unleashing the power of supervisory feedback in academic writing: Strategies for timely undergraduate thesis completion,” Teflin J., vol. 35, no. 2, pp. 330–351, 2024.
[8] J. Amani, H. Myeya, and M. Mhewa, “Understanding the motives for pursuing postgraduate studies and causes of late completion: Supervisors and supervisees’ experiences,” Sage Open, vol. 12, no. 3, p. 21582440221109584, 2022.
[9] P. R. INDONESIA, “UNDANG-UNDANG REPUBLIK INDONESIA NOMOR 14 TAHUN 2005 TENTANG GURU DAN DOSEN”.
[10] P. R. INDONESIA, “Undang-Undang Republik Indonesia Nomor 12 Tahun 2012 Tentang Pendidikan Tinggi”.
[11] R. Komaladewi, P. Usmanij, A. R. Amani, and F. E. Saputra, “Exploring social media marketing: A key driver of social commerce in Indonesia,” J. Int. Counc. Small Bus., pp. 1–18, 2025.
[12] L. R. Pelima, Y. Sukmana, and Y. Rosmansyah, “Predicting university student graduation using academic performance and machine learning: a systematic literature review,” IEEE Access, vol. 12, pp. 23451–23465, 2024.
[13] A. Almalawi, B. Soh, A. Li, and H. Samra, “Predictive models for educational purposes: A systematic review,” Big Data Cogn. Comput., vol. 8, no. 12, p. 187, 2024.
[14] W. Ahmed, M. A. Wani, P. Plawiak, S. Meshoul, A. Mahmoud, and M. Hammad, “Machine learning-based academic performance prediction with explainability for enhanced decision-making in educational institutions,” Sci. Rep., vol. 15, no. 1, p. 26879, 2025.
[15] M. S. Bahri Yusoff et al., “Academic Workloads: Insights from Medical and Non-medical Lecturers’ Experience in Universiti Sains Malaysia.,” Educ. Med. J., vol. 16, no. 3, p. 119, 2024.
[16] M. Mårtensson and J. Söderström, “How does supervision shape student thesis outcomes? Expanding the theory and measurement of supervision and its impact,” J. Polit. Sci. Educ., vol. 22, no. 1, pp. 1–18, 2026.
[17] T. L. Wheelen, J. D. Hunger, A. N. Hoffman, C. E. Bamford, and T. Wheelen, Strategic management and business policy. Pearson Prentice Hall Upper Saddle River, NJ, 2004.
[18] R. Sharda, D. Delen, and E. Turban, Business Intelligence, Analytics, Data Science, and AI, Global Edition--EPub. Pearson Education, Limited, 2024.
[19] V. Kotu and B. Deshpande, Data science: concepts and practice. Elsevier, 2019.
[20] M. Bahtilla and A. I. Oben, “International students’ perception of research supervision in the Social Sciences: the case of three comprehensive universities in China,” J. Furth. High. Educ., vol. 46, no. 3, pp. 327–340, 2022.
[21] M. Neupane Bastola, “Engagement and challenges in supervisory feedback: Supervisors’ and students’ perceptions,” RELC J., vol. 53, no. 1, pp. 56–70, 2022.
[22] S. Gaftandzhieva, S. Hussain, S. Hilcenko, R. Doneva, and K. Boykova, “Data-driven decision making in higher education institutions: State-of-play,” Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 6, 2023.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Bismiarti Yudasilati, Samidi Samidi
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Facebook 
Instagram 
Twitter 
Youtube 
westscience.journals@gmail.com