The work aims to propose a whale algorithm based on hybrid optimization (MAIWOA) to solve the time optimal value of the manipulator in the process of two-point movement, so as to improve the movement efficiency of the manipulator in warehouse storage, packaging and production. Firstly, the initial population was optimized by chaos algorithm; Secondly, a learning mechanism of tick function and bounding contraction was proposed to help the algorithm jump out of local convergence; Then an improved reverse tree topology was used to improve the diversity of population exploration; The improved convergence factor and adaptive weight mechanism were adopted to balance the global exploration ability in the early and middle stages and the convergence in the later stage. Results showed that the convergence speed and accuracy of this algorithm were improved significantly. The algorithm was applied to the optimal solution of joint motion time of quintic polynomial interpolation, and good results were obtained. The motion time of joints 1, 2 and 3 was reduced from 20 s to 9.642 5, 9.251 5 and 10.787 s respectively, and the efficiency was increased by 51.79%, 53.74% and 46.07% respectively. It is concluded that applying MAIWOA to the trajectory planning of general manipulator can improve the execution efficiency of manipulator in production line.