Optimising and modeling the mixed-model two-sided disassembly line balancing problem with human-robot cooperation restrictions

This study examines human-robot cooperation in a two-sided disassembly line balancing problem, emphasising its importance in enhancing line flexibility and efficiency in the industry. A mathematical model is applied to optimise the cycle time, followed by a simulation model to account for the uncertainty arising from the factors: skill level of resources, the quality of End-of-life (EOL) products, and the number of running workstations. Statistical analysis through design of experiment and ANOVA shows that the factors and their interactions significantly affect the KPIs: time in the system, cycle time, and average resource utilisation. Then, the response optimiser composite desirability was used to determine the level of factors that simultaneously minimise cycle time and TIS while maximising average resource utilisation under different running conditions. The distribution with the shortest disassembly time consistently achieved the highest composite desirability in all examples. Although increasing the number of workstations reduces or maintains cycle time, it does not always lead to proportional gains. A tailored dashboard was developed to support decision-makers in making informed choices regarding workstations and operational factors, thereby optimising disassembly line performance for diverse EOL product needs.