Cognitive Production Lines Reinforcement Learning and Adaptive Intelligence for Real-Time Vehicle Assembly Reconfiguration in Industry 5.0
DOI:
https://doi.org/10.5281/zenodo.20442369Keywords:
Adaptive assembly line, AI data infrastructure, machine learning, next-generation vehicle manufacturing, perceptual systems, optimal scheduling, predictive maintenance, operator support, human–machine collaboration, threat modeling.Abstract
Smart Vehicle Manufacturing (SVM) plays a vital role in the fast-growing market of electric vehicles. A distinctive feature of these New-Generation Intelligent Connected Vehicles (NGICVs) is their software architecture. SVM uses a traditional Mass Customization assembly-line production mode. Unlike simple electric vehicle architectures, NGICVs require more diverse components and parts like traditional internal combustion engine vehicles, but these part shortages lead to serious line-stopping production issues. Building SVM systems based on traditional Mass Customization assembly-line modes cannot solve this problem.
To overcome these issues, Smart Vehicle Manufacturing paradigms that adopt Adaptive Assembly Lines (AALs) are proposed. An Adaptive Assembly Line (AAL) is a fundamentally new assembly-mode paradigm that enables real-time assembly-line adaptation and mass customization of various part configurations according to real-time market demands in Smart Vehicle Manufacturing. AAL architecture drastically changes each stage of the Smart Vehicle Manufacturing assembly process. SVM systems have deployed data architecture supporting these Smart Vehicle Manufacturing paradigms. The developed data governance framework with in-depth data perception enables high-quality data generation for AI-driven Adaptive Assembly Line system performance improvement.
References
[1] Bandhana, A., & Vokřínek, J. (2025). AI-driven manufacturing: Surveying for Industry 4.0 and beyond. Operations Research Forum, 6, 145.
[2] Barua, D. A., Hossain, M. S., Rahman, M. M., & Islam, M. R. (2025). Leveraging artificial intelligence for smart production management: A mixed-method study. Scientific Reports, 15, Article 25413.
[3] Catalano, F., Faccio, M., & Ferrari, E. (2025). Manned assembly systems design considering layout, workforce strategies, and worker variability: A systematic literature review. The International Journal of Advanced Manufacturing Technology.
[4] Chaudhary, H., Sharma, P., & Singh, R. (2025). AI-enhanced modelling of queueing and scheduling systems using deep workload prediction and reinforcement learning. SN Applied Sciences, 7, Article 6755.
[5] Del Gallo, M., Mazzuto, G., Ciarapica, F. E., & Bevilacqua, M. (2023). Artificial intelligence to solve production scheduling problems in real industrial settings: Systematic literature review. Electronics, 12(23), 4732.
[6] Ding, J., Chen, M., Wang, T., Zhou, J., Fu, X., & Li, K. (2023). A survey of AI-enabled dynamic manufacturing scheduling: From directed heuristics to autonomous learning. ACM Computing Surveys, 55(14S), Article 307.
[7] El Abidine, Z. Z. (2024). The effect of learning on assembly line balancing: A review. Periodica Polytechnica Social and Management Sciences.
[8] El Machouti, S., Battaïa, O., & Dolgui, A. (2025). A hybrid genetic algorithm for multi-objective worker allocation and balancing in multi-manned assembly lines. Proceedings, 97(1), 41.
[9] Florescu, A., Popescu, D., & Oprean, C. (2024). Digital twin for flexible manufacturing systems and virtual commissioning: A review and application perspectives. Machines, 12(11), 785.
[10] Fu, B., Bi, M., Umeda, S., Nakano, T., Nonaka, Y., Zhou, Q., Matsui, T., Tilbury, D. M., & Barton, K. (2025). Digital twin-based smart manufacturing: Dynamic line reconfiguration for disturbance handling. arXiv (preprint).
[11] Geurtsen, M., van der Heijden, M. C., & Schutten, J. M. J. (2023). Deep reinforcement learning for optimal planning of maintenance in assembly lines. Computers & Industrial Engineering, 177, 109084.
[12] Hawaou, K. S., Elhoussaine, S., & Hlyal, M. (2024). Industry 4.0 and industrial workflow scheduling: A survey. Internet of Things and Cyber-Physical Systems, 4, 100119.
[13] Hlyal, M. (2025). Dynamic bi-objective balancing of assembly lines: Optimizing productivity and quality by integrating operator skill evolution. Journal of Umm Al-Qura University for Applied Sciences.
[14] Hoffmann, M. A., & Danyluk, A. (2023). Tackling industrial downtimes with artificial intelligence in data-driven maintenance. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies.
[15] Javaid, M., Haleem, A., Singh, R. P., & Suman, R. (2022). Artificial intelligence applications for Industry 4.0: A literature-based study. Journal of Industrial Integration and Management, 7(1), 1–45.
[16] Ka blinger, A., Woschank, M., & Miklautsch, P. (2022). State of the art and future directions of digital twins in production logistics: A systematic literature review. Applied Sciences, 12(2), 669.
[17] Kheirabadi, M., Battaïa, O., & Dolgui, A. (2023). Human-robot collaboration in assembly line balancing problems: Review and research gaps. Computers & Industrial Engineering, 176, 109737.
[19] Kwon, Y., Lee, J., & Kim, S. (2025). Smart manufacturing innovation through integration of digital twins and IoT energy management. International Journal of Control, Automation and Systems.
[20] Latsou, C., Mourtzis, D., & Doukas, M. (2024). A unified framework for digital twin development in manufacturing emphasizing scalability and reuse. Advanced Engineering Informatics, 60, 102194.
[21] Li, Y. C., & Wang, X. (2024). Human-robot collaboration assembly line balancing considering cross-station tasks and carbon emissions. Advances in Production Engineering & Management, 19(1), 31–45.
[22] Li, Y., & Wang, X. (2025). Optimizing assembly line balancing with carbon footprint constraints using Q-learning. International Journal of Industrial Engineering and Technology Applications and Practice.
[23] Lugaresi, G., Gadaleta, M., & Regattieri, A. (2023). Automated digital twin generation of manufacturing systems with complex material flows. Computers in Industry, 151, 103980.
[24] Ma, G., Wang, X., & Nielsen, P. (2024). Digital twin-based demand-driven production planning and scheduling: A conceptual framework. Procedia CIRP, 130, 1–6.
[25] Mao, Z., Zhang, Y., & Battaïa, O. (2023). Balancing U-type assembly lines with human–robot collaboration. Computers & Operations Research, 158, 105674.
[26] Michels, A. S., Battaïa, O., & Dolgui, A. (2024). Model and heuristics for the multi-manned assembly line balancing problem with heterogeneous workforce integration. International Journal of Production Research.
[27] Moiceanu, G., Paraschiv, G., & Dima, I.-C. (2022). Digital twin and smart manufacturing in industries: A bibliometric analysis. Sensors, 22(4), 1388.
[28] Moussa, M., Abdallah, M., & ElMaraghy, H. (2025). Industry 4.0 in automotive manufacturing: A digital twin framework for real-time monitoring and decision-making. Procedia Manufacturing, 74, 1–8.
[29] Negri, E., Fumagalli, L., & Macchi, M. (2021). A field-synchronized digital twin framework for production scheduling under equipment health degradation. Journal of Intelligent Manufacturing, 32, 1–20.
[30] Neves, M., & Neto, P. (2023). Deep reinforcement learning applied to an assembly sequence planning problem with user preferences. arXiv (preprint).
[31] Onaji, I., Tiwari, D., Soulatiantork, P., Song, B., & Tiwari, A. (2022). Digital twin in manufacturing: Conceptual framework and case studies. International Journal of Computer Integrated Manufacturing, 35(8), 831–858.
[32] Pandhare, V., Negri, E., Ragazzini, L., Cattaneo, L., & Macchi, M. (2024). Digital twin-enabled robust production scheduling for equipment in degraded state. Journal of Manufacturing Systems, 74, 841–857.
[33] Reddy, G. D., & Kumar, S. (2024). Optimizing smart manufacturing systems by integrating digital twin technology with teaching–learning-based optimization. Cogent Engineering, 11, Article 2415670.
[34] Seipolt, A., Schaefer, D., & Wende, T. (2025). Digital twins in manufacturing: A systematic literature review with retrieval-augmented generation. IEEE Access.
[35] Soori, M., Arezoo, B., & Dastres, R. (2023). Digital twin for smart manufacturing: A review. Sustainable Manufacturing and Service Economics, 2, 100017.
[36] Webb, L., Tokhi, O. M., & Alkan, B. (2024). State of the art and future directions of digital twin-enabled smart assembly automation in discrete manufacturing industries. International Journal of Computer Integrated Manufacturing.
[37] Wen, X., & Zhang, Q. (2025). An improved multi-objective harmony search algorithm for assembly line balancing under practical constraints. Computers & Industrial Engineering.
[38] Zhang, Q., Li, X., & Chen, Y. (2025). Optimizing robotic disassembly–assembly line balancing via intelligent scheduling in cyber-physical manufacturing systems. Electronics, 14(17), 3499.
Additional Files
Published
Data Availability Statement
None
