As an emerging trend of intelligent manufacturing, collaborative robots are recently introduced and implemented in the assembly line systems with human-robot collaborative approach, which has received increasing attention and practice. In such system, workers and robots can perform tasks separately, simultaneously, or collaboratively at each station, which can significantly improve the efficiency of the production line. This study addresses an assembly line balancing problem with human-robot collaboration (ALBP-HRC) to minimize the cycle time. We first propose multiple enhancement strategies, including lower bound, upper bound and initial solution. Meanwhile, based on the existing ALBP-HRC model, we formulate an enhanced mixed integer programming model. Secondly, we present an improved simulated annealing algorithm, using multiple completion time estimation procedures for stations to solve the problem efficiently. Moreover, extensive experiments are conducted, and the results verify the effectiveness and applicability of the methods. Finally, we investigate sensitivity analysis on different parameters, revealing the managerial insights and practical references for enterprises implementing human-robot collaborative assembly line.