To solve the electric vehicle problem with simultaneous delivery and pickup, a mathematical model was set up with the lowest total cost as target to solve the model by simulated annealing-ant colony algorithm considering the limitation of vehicle capacity and electric power. Firstly, a mathematical model was built for the location and routing according to the limitation of simultaneous delivery and pickup and the constraints of time window during the actual distribution. Secondly, drawing-back and Gaussian mutation design were added to improve the hybrid simulated annealing-ant colony optimization. The model was solved by this mixed algorithm, and the proposed algorithm was compared with ant colony optimization (ACO), Tabu search (TS) and adaptive large neighborhood search (ALNS) to confirm its superiority. Last, comparison was made with the distribution strategy of separate delivery and pickup. Based on the performance of example checking in different scales, the experimental results show that the algorithm and distribution strategy proposed achieve lower cost.