Human-Robot Collaboration in Mixed-Flow Assembly Line Balancing under Uncertainty: An Efficient Discrete Bees Algorithm

Abstract

In the evolving landscape of manufacturing and remanufacturing, assembly lines play a crucial role. Within the context of Industry 5.0, human workers are seen as a valuable and irreplaceable resource. Human-robot collaboration is a promising production model that combines the strengths of human workers and robots, thereby enhancing production efficiency while reducing occupational risks related to ergonomics. Despite these advancements, inherent uncertainties within assembly processes, the integration of human-robot partnerships, and the dynamic nature of market demands pose significant challenges to traditional assembly methods. To address these challenges, this research introduces a novel modelling approach through a mixed-flow assembly line balancing problem designed for uncertain environments, fostering collaboration between humans and robots. The primary goal is to facilitate efficient collaboration within a type-I assembly line balancing problem framework, where predefined assembly beats guide the workflow. In this research, the use of interval type-2 fuzzy sets capabilities was investigated to address uncertainties in the assembly process. Furthermore, the potential of pairing human operators of different abilities with robots of different models for collaborative tasks at workstations was explored, enhancing flexibility and adaptability in the assembly line. In response to the complexity of the problem, this research proposes an efficient multiobjective discrete bees algorithm that incorporates innovative operators and search strategies. Rigorously tested across diverse case studies, this algorithm consistently outperforms other comparator algorithms. This research not only offers novel perspectives on addressing assembly line balancing challenges but also provides valuable insights for the effective implementation of human-robot collaborative assembly in uncertain environments.