Co-EvoDT: Design of a Co-evolving Digital Twin Framework for Media Digital Transformation in the Metaverse through Integration of Multi-Agent Reinforcement Learning and LSTM-based Sequential Forecasting

Document Type : Original Article

Authors

1 PhD Student in Media Management, Department of Media Management, Se.C., Islamic Azad University, Semnan, Iran. Email: seyedehtahereh.mousavi@iau.ac.ir

2 Corresponding Author. Associate Professor, Department of Industrial Management, Se.C., Islamic Azad University, Semnan, Iran. Email: fa.faezy@iau.ac.ir

3 Associate professor, Department of Media Management, Se.C., Islamic Azad University, Semnan, Iran. Email: ab.danaei@iau.ac.ir

10.22091/jemsc.2026.14030.1306

Abstract

Digital twins in the metaverse have created new opportunities to redefine digital transformation in media; however, the absence of co-evolutionary models that simultaneously optimize infrastructure resource-allocation policies, content production, and load forecasting remains a major barrier to fully realizing these opportunities. To address this gap, we propose the Co-EvoDT model and implement it within a simulated metaverse environment using a design-driven, quantitative research approach. The proposed architecture combines sequential load forecasting via an LSTM predictor (prediction horizon = 1, sequence length = 8), multi-agent reinforcement learning trained with the policy-based REINFORCE algorithm to enable co-evolutionary training of content, infrastructure, and user digital twins, and a runtime rolling-update mechanism that appends the LSTM’s normalized output to the state vector used by the control policies. Performance was evaluated using Quality of Experience (QoE), latency, cost, and RMSE of the load predictor. Simulation results show that a policy trained with the combined predictive signal outperforms a random policy by simultaneously improving multiple metrics: a relative increase of ≈93% in QoE, a relative latency reduction of ≈29%, and a cost reduction of ≈27%. Forecast RMSE was reduced by ≈43%, ≈53.2%, and ≈88% compared to Naive, ARIMA, and Exponential Smoothing baselines, respectively. Reward-curve convergence analysis and parametric sensitivity experiments further corroborate the stability and robustness of the learned policies under variations of key system parameters. The principal innovation of this research is the operational integration of multi-agent reinforcement learning, sequential forecasting, and co-evolutionary population-gradient update mechanisms within digital twins to enable proactive, coordinated resource management in the metaverse. The results indicate that Co-EvoDT can concurrently enhance user experience and macro-level system performance.

Keywords

Main Subjects

References

AlBalkhy, W., Karmaoui, D., Ducoulombier, L., Lafhaj, Z., & Linner, T. (2024). Digital twins in the built environment: Definition, applications, and challenges. Automation in Construction, 162, 105368. https://doi.org/10.1016/j.autcon.2024.105368 
Alvi, M., Dutta, H., Minerva, R., Crespi, N., Raza, S. M., & Herath, M. (2025). Global perspectives on digital twin smart cities: Innovations, challenges, and pathways to a sustainable urban future. Sustainable Cities and Society, 106356. https://doi.org/10.1016/j.scs.2025.106356 
Benaben, F., Congès, A., & Fertier, A. (2025). A prospective vision of the evolution of immersive technologies: Towards a definition of metaverse. Technovation, 140, 103154. https://doi.org/10.1016/j.technovation.2024.103154 
Bersani, M. M., Braghin, C., Cortellessa, V., Gargantini, A., Grassi, V., Presti, F. L., Mirandola, R., Pierantonio, A., Riccobene, E., & Scandurra, P. (2022, March). Towards trust-preserving continuous co-evolution of digital twins. In 2022 IEEE 19th International Conference on Software Architecture Companion (ICSA-C) (pp. 96-99). IEEE. https://doi.org/10.1109/ICSA-C54293.2022.00024 
Chow, Y. W., Susilo, W., Li, Y., Li, N., & Nguyen, C. (2022). Visualization and cybersecurity in the metaverse: A survey. Journal of Imaging, 9(1), 11. https://doi.org/10.3390/jimaging9010011 
Cong, A., Jin, Y., Lu, Z., Gao, Q., Ge, X., Li, Z., Rongzhou, L., Xinying, H., & Hou, L. (2025). Transfer learning-based physics-informed DeepONets for the adaptive evolution of digital twin models for dynamic systems. Nonlinear Dynamics, 1-28. https://doi.org/10.1007/s11071-025-11158-4 
Fan, S., Tong, H., & Wang, S. (2025). A system dynamics-based hybrid digital twin model for driving green manufacturing. Systems, 13(8), 651. https://doi.org/10.3390/systems13080651 
Feng, B., Wang, Z., Yuan, L., Zhou, Q., Chen, Y., & Bi, Y. (2025). Towards safe motion planning for industrial human-robot interaction: A co-evolution approach based on human digital twin and mixed reality. Robotics and Computer-Integrated Manufacturing, 95, 103012. https://doi.org/10.1016/j.rcim.2025.103012 
Hady, M. A., Hu, S., Pratama, M., Cao, Z., & Kowalczyk, R. (2025). Multi-agent reinforcement learning for resources allocation optimization: A survey. Artificial Intelligence Review, 58(11), 354. https://doi.org/10.1007/s10462-025-11340-5 
Haimes, Y. Y. (2018). Risk modeling of interdependent complex systems of systems: Theory and practice. Risk analysis, 38(1), 84-98. https://doi.org/10.1111/risa.12804 
Li, Z., Ji, Q., Ling, X., & Liu, Q. (2025). A comprehensive review of multi-agent reinforcement learning in video games. IEEE Transactions on Games. https://doi.org/10.1109/TG.2025.3588809 
Mousavi, S. T. , Faezy Razi, F., & Danaei, A. (2025). Medavers: A digital transformation model for media in the metaverse. Interdisciplinary Journal of Management Studies, 19(1), 221-246. https://doi.org/10.22059/ijms.2025.395001.677621 
Novikov, R.Yu., & Zohrabyan, E.P. (2023). Digital transformation of media: challenges and opportunities. Journal of Digital Economy Research, 1(4), 102–125. https://doi.org/10.24833/14511791-2023-4-102-125 
Olcay, K., Tunca, S. G., & Özgür, M. A. (2024). Forecasting and performance analysis of energy production in solar power plants using long short-term memory (LSTM) and random forest models. IEEE Access. PP(99), 1-1. https://doi.org/10.1109/ACCESS.2024.3432574 
Rammel, C., Stagl, S., & Wilfing, H. (2007). Managing complex adaptive systems—A co-evolutionary perspective on natural resource management. Ecological economics, 63(1), 9-21. https://doi.org/10.1016/j.ecolecon.2006.12.014 
Rezaeenour, J., & Karimian, R. (2024). Identifying metaverse developments in digital libraries based on library theory. Knowledge Retrieval and Semantic Systems, 11(39), 67-108. https://doi.org/10.22054/jks.2023.76141.1617 
Rezaeenour, J., & Karimian, R. (2025). Identification of digital library indicators in metaverse environment. The International Journal of Metaverse & Virtual Transformation (IJMVT), 1(2), 74-94. https://mvt.artahub.ir/article_227497_ed969c594c314ad19dc784f65f9cf800.pdf 
Salamattalab, M. M., Zonoozi, M. H., & Molavi-Arabshahi, M. (2024). Innovative approach for predicting biogas production from large-scale anaerobic digester using long-short term memory (LSTM) coupled with genetic algorithm (GA). Waste Management, 175, 30-41. https://doi.org/10.1016/j.wasman.2023.12.046 
Tong, X., Bao, J., & Liu, T. (2024, August). Co-Evolution DTs: Achieving value-added cognitive digital twins across the entire lifecycle. In 2024 IEEE 20th International Conference on Automation Science and Engineering (CASE) (pp. 3140-3146). IEEE. https://doi.org/10.1109/CASE59546.2024.10711333 
Tong, X., Bao, J., & Tao, F. (2024). Co-evolutionary digital twins: A multidimensional dynamic approach to digital engineering. Advanced Engineering Informatics, 61, 102554. https://doi.org/10.1016/j.aei.2024.102554 
Jafari, M., Akhavan, P., & Akbari, A. H. (2026). Enhancing supply chain agility and performance through big data analytics: the role of digitalization and top management support. International Journal of Productivity and Performance Management, 1-22. https://doi.org/10.1108/IJPPM-06-2025-0557 
Tavakkoli-Moghaddam, R., Akbari, A. H., Tanhaeean, M., Moghdani, R., Gholian-Jouybari, F., & Hajiaghaei-Keshteli, M. (2024). Multi-objective boxing match algorithm for multi-objective optimization problems. Expert Systems with Applications, 239, 122394. https://doi.org/10.1016/j.eswa.2023.122394 
Yavari, M., Marvi, M., & Akbari, A. H. (2020). Semi-permutation-based genetic algorithm for order acceptance and scheduling in two-stage assembly problem. Neural Computing and Applications, 32, 2989-3003. https://doi.org/10.1007/s00521-019-04027-w 
Send comment about this article
CAPTCHA Image

Files

Share

How to cite

Statistics