Abstract: Seeing the fact that the losses and costs caused by bad bicycles continue to grow in the operation of Shared Bicycle System and forecasting the distribution of bad bicycles in the site network, a damaged bicycle recycling network model is established based on cost threshold analysis, a recycling route assignment model is built with the goal of minimizing recycling costs and the actual case solved as well using the hybrid algorithm of tabu search and genetic algorithm. The results show that the method can solve the problem of distribution prediction and path optimization in the recycling process of damaged bicycles in real-situations. In the example analysis, compared with the traditional recycling method, the established model has a cost saving of 8.72% and an excellent application value, which not only reduces the operating cost of the recycling process, but also improves the recycling efficiency, which has certain reference value for decision-making optimization of shared bicycle operators.

Key words: bike sharing, damaged bicycle recycling network, recycle routes, genetic algorithm, tabu search algorithm