Software Service Demand Management for Providers Supported by Machine Learning and System Dynamics

• Ednardo da Rocha UFC
• Alberto Sampaio Lima UFC https://orcid.org/0000-0003-0696-5148
• José Neuman de Souza UFC https://orcid.org/0000-0002-2980-8877
• José Antão Beltrão Moura UFCG

Abstract

Demand management in software services provided by vendors faces challenges such as the unpredictability of request arrivals and the need to maintain strict service level agreements (SLAs). This article proposes a hybrid model that integrates System Dynamics (SD) and Machine Learning (ML) to forecast demand and simulate the evolution of the service backlog. The algorithms Random Forest, MLP, and Gradient Boosting were evaluated, with Random Forest demonstrating the lowest predictive error. The results show that the proposed model enables the analysis of operational scenarios based on historical data, anticipates bottlenecks, and supports decision-making regarding resource allocation. Its modular and reproducible structure makes the solution applicable to real-world service management contexts, fostering proactive and evidence-based actions.

Keywords: Demand management, Software service, Machine learning, System dynamics

References

Akbari, M. and Do, T. N. A. (2021). A systematic review of machine learning in logistics and supply chain management: current trends and future directions. Benchmarking: An International Journal, 28(10):2977–3005.

Beal, A. (2001). Introdução à gestão de tecnologia da informação. Beal Educação e Tecnologia, 2.

Bezerra, T. R., Moura, J. A. B., Lima, A. S., and Souza, J. N. d. (2024). Decision-making support in it services sourcing management through a system dynamics model. IEEE Transactions on Network and Service Management, 21(4):3813–3828.

Chen, S., Pan, Y., Chen, L., and Wu, D. D. (2020). A system dynamics model for capacity planning of component supply in complex product manufacturing. IEEE Systems Journal, 15(1):8–16.

Choi, J.-h., Lee, Y., and Kwak, Y. H. (2020). A socioeconomic ripple effect analysis of integrative national construction standards codification efforts: system dynamics approach. IEEE Transactions on Engineering Management, 69(6):2959–2975.

Faezipour, M. and Ferreira, S. (2016). A system dynamics approach for sustainable water management in hospitals. IEEE Systems Journal, 12(2):1278–1285.

Fenner, G., Lima, A. S., Souza, J. N. d., Moura, J. A. B., and Bezerra, T. R. (2020). Business-driven support for infrastructure as a service capacity management through system dynamics simulations. IEEE Transactions on Network and Service Management, 18(2):2063–2076.

Kohlrausch, G. S. (2013). Proposta para implantação e gestão de um processo de desenvolvimento de software baseado em métodos ágeis. Master’s thesis, Universidade do Vale do Rio dos Sinos (UNISINOS).

Li, X., Kuang, H., and Hu, Y. (2020). Using system dynamics and game model to estimate optimal subsidy in shore power technology. IEEE Access, 8:116310–116320.

Monteiro, L. H. A. (2002). Sistemas dinâmicos. Editora Livraria da Física.

Perea, E., Grossmann, I., Ydstie, E., and Tahmassebi, T. (2000). Dynamic modeling and classical control theory for supply chain management. Computers & Chemical Engineering, 24(2-7):1143–1149.

Polo-Triana, S., Gutierrez, J. C., and Leon-Becerra, J. (2024). Integration of machine learning in the supply chain for decision making: A systematic literature review. Journal of Industrial Engineering and Management, 17(2):344–372.

Villate, J. E. (2016). Dinâmica e sistemas dinâmicos. Universidade do Porto, Creative Commons Atribution Sharealik, ISBN 978-972-99396-1-7.

Ye, H., Liang, L., Li, G. Y., Kim, J., Lu, L., and Wu, M. (2018). Machine learning for vehicular networks: Recent advances and application examples. IEEE Vehicular Technology Magazine, 13(2):94–101.