Hybrid Cuckoo Algorithm Based On Simulated Annealing For Solving Bus Scheduling Problem

Since reform and opening up,China's economy has developed rapidly,people's living standards have been continuously improved,and motor vehicles have also been increasing year by year,which caused traffic congestion and environmental pollution intensified.To solve the traffic problems properly,we must vigorously develop the public transportation on urban roads,establish advanced and intelligent public transport systems,and implement public transportation priority strategy.The bus dispatch is one of the core tasks of the bus company,the key to scheduling optimization is to develop reasonable departure intervals,so as to reduce operation costs of the company and improve operation efficiency and service level.In recent years,China's public transportation undertakings have been greatly developed,and bus dispatching has also spawned many optimization methods.However,due to its complexity and particularity,this issue has not been resolved so far.The bus scheduling problem belongs to the multi-objective nonlinear optimization NP problem,and it is difficult to solve the problem by using the traditional mathematical method.However,with the development of optimization theory,a variety of swarm intelligence algorithms have been gradually applied to the optimization problem of public traffic scheduling,which effectively solve this problem.On the basis of the research results of scheduling optimization and swarm intelligence optimization algorithm,this paper focuses on the cuckoo search algorithm.As the cuckoo optimization algorithm is proposed late,domestic and foreign research results are few,and has not yet been introduced to the bus scheduling problem,so it is very important to carry out research on the cuckoo algorithm.Applying the cuckoo algorithm to the bus dispatch optimization problem has great research value.The cuckoo search algorithm has the advantages of clear theory,few parameters,easy expansion,strong global search ability,and easy implementation.The research shows that the CS algorithm has higher performance than the typical group intelligent algorithm such as genetic algorithm,artificial bee colony algorithm and particle swarm algorithm,which can get better optimization results with fewer function solving times.Considering disadvantages of the original cuckoo algorithm which was easily trapping in local optimum and has low search quality,this paper improves the search step and combines simulated annealing algorithm with it,and designs a simulated annealing-Cuckoo algorithm.The test function verifies that the hybrid cuckoo algorithm has advantages of better results and faster convergence than the standard cuckoo algorithm.In this paper,combining the characteristics of bus dispatching itself and taking into account the interests of both the bus company and the passengers,the bus scheduling model is established with the departure interval as an independent variable.The hybrid cuckoo algorithm is introduced into the bus scheduling,and the scientificalness and feasibility of the algorithm are verified by an example,it provides a new idea for solving bus dispatching optimization problems.