<p>In response to the diverse tasks in the production process, varying levels of human–robot collaboration (HRC) are necessary. This study explores the different modes of HRC to achieve automated and efficient production advantages in assembly line systems. As an effective method, HRC aims to achieve high efficiency, high quality, and low-cost production through the collaboration of humans and robots in a limited workspace. Introducing robots to execute tasks can replace humans in repetitive and stressful work, thereby improving the working environment. Aligned with the technological and economic vision of Industry 4.0, the design of production line balancing has shifted from manual configuration to HRC configuration, achieving high production efficiency and flexibility. This research aims to construct a multi-objective mixed-model assembly line balancing problem with human–robot collaboration and fuzzy processing time (MO-MALBP-HRC-FPT) model, assigning various tasks to suitable stations. The objectives of this study, which include minimizing cycle time, cost of collaboration workforce, smoothness index, energy consumption, and number of stations, are of significant importance in the context of assembly line systems. A fuzzy set-based approach is used to represent processing time uncertainty. A preemptive fuzzy multiple objective programming (PFMOP) model is proposed to plan assembly line assignments and calculate relevant operational costs. Non-dominated sorting genetic algorithm II (NSGA-II) is applied to solve large-scale real-world cases and obtain near-optimal solutions.</p>