| With the rapid development of the Internet sharing economy,the shared bicycle system has become the most typical sharing economy model in the form of Internet advocacy,based on the low-carbon economy and the characteristics of stop and go,the shared bicycle system effectively solves the "last mile" problem of users’ travel in urban transportation.With the rapid development of the shared bicycle industry,related problems have also emerged,for example,the wanton parking of shared bicycles seriously affects the appearance of the city,and the unreasonable layout of shared bicycles makes it difficult for users to pick up and return the bicycles,and they are far away from users,to a certain extent,reduce the utilization rate of users for shared bicycle transportation,in addition,the initial investment cost of shared bicycles is high,and the operating cost is high,and a single fixed fare is adopted,the fixed location of picking up and returning bikes also limits the number of shared bicycle users,making it difficult for shared bicycle companies to make profits.therefore,this paper has a positive significance to study the location problem and pricing strategy of the shared bicycle system.this paper mainly conducts research and analysis on the following aspects:(1)First,study the reasonable layout planning of shared bicycle placement points,based on the robust optimization theory,a model for the location selection of shared bicycles is established when the cost of shared bicycles is determined,design a simple example to verify and use Lingo18.0 programming to solve,and obtain the optimal placement point of shared bicycles among the known alternative placement points;considering that the cost of shared bicycles is uncertain,the location model of shared bicycles is established.in the established cost uncertainty location model,the cost is an indefinite number and only exists in the objective function,that is,using the robust optimization method to introduce the box uncertainty set to describe the uncertainty of the cost,using the dual theory to transform it into a deterministic robust peer-to-peer model that is easier to solve,and designing an immune algorithm to solve it.it is calculated that the optimal placement point of the shared bicycle is selected from the given known node positions,so as to meet the travel needs of users to the greatest extent.(2)Secondly,from the perspective of revenue management theory,the problem of pricing strategy of shared bicycles is studied.This paper analyzes the pricing principles,profit conditions,influencing factors in the pricing process and pricing research methods of shared bicycles.The shared bicycle pricing strategy is regarded as a Leader-Follower problem,and a two-level programming model is constructed to describe the mutual interests between shared bicycle companies and users.The model takes the decision-maker of the shared bicycle enterprise as the leader and the user’s travel behavior as the follower.Taking the maximization of the revenue of shared bicycle companies as the upper-level function,the user travel always chooses the transportation mode with the smallest generalized cost as the lower-level function according to their own conditions cost.(3)Finally,an example is verified and simulated by the shared bicycle transportation mode from Lanzhou West Railway Station to Liujiabao Square.the upper-level planning function is solved by particle swarm algorithm,the lower-level planning function converts the demand change problem into a fixed demand problem and then uses the Frank-Wolfe algorithm to solve it,when the convergence of the optimal particle value of the upper-level planning function is stable,the passenger flow allocated by the lower-level function also stabilizes,through repeated iterations,the optimal fares of shared bicycles from Lanzhou West Railway Station to Liujiabao Square are obtained,and the sensitivity analysis method is used to verify the feasibility of the results.the approximate expression form of the lower reaction function is obtained by Taylor expansion,after solving,the optimal fares of the two methods in this road section are approximately equal,the feasibility of solving the bilevel programming model based on particle swarm optimization is verified. |
