Abstract: The optimal workers scheduling of assembly line considering the differences in worker assembly levels and the flow distribution between different workstations was analyzed. After determining the sort of mixed flow assembly line, the assembly worker worked in the workstation where she/he was good at and took part in the assembly work of other stations after completing the work in this workstation. An optimized mathematical model of mixed-flow assembly line workers considering the difference in workstation efficiency and flow distribution was established after the definition of the relevant optimization factor variables and constraints. The peak sharpening algorithm and multiple population migration algorithm were adopted for the adjustment of the workers and balance optimization, and then a three-layer neural network to evaluate peak cutting the location choice was used to obtain the optimal allocation of workers. Finally, an example was given to verify the listed model with Matlab, and two assembly workers were reduced and the assembly line balance rate was increased.

Key words: mixed flow assembly, worker optimization, efficiency difference, flow distribution