A Histogram-based Estimation Of Distribution Algorithm For Reversible Lanes Optimization Problems

The Reversible lanes optimization problem(RLOP)is a complex optimization problem in traffic management.The objective of this problem is to find an optimal direction assignment of lanes in an urban traffic network,so that the traffic capacity of urban streets could get the utmost promotion.The traditional research rarely involves this kind of problem,and most of them only focus on the structure.At the same time,there are few studies on quantitative analysis of this problem.Thus,the following work has been done in this paper.Firstly,the mathematical model of RLOP in traffic network is established.RLOP is modeled as a typical bi-level optimization problem: on the one hand,the traffic assignment problem is the lower-level optimization problem,which is based on the Wardrop principle.On the other hand,the total traffic cost optimization problem is the upper-level optimization problem.The objective of this problem is to minimize the total cost of vehicles under the optimal solution of lower-level optimization problem.Secondly,this paper proposes a Histogram-based estimation of distribution algorithm(HEDA),which is used to solve the upper-level optimization problem in the RLOP.At the same time,the traditional Frank Wolfe algorithm is used to solve the lower-level optimization problem.In addition,in the algorithm design,the Latin hypercube sampling strategy is used to initialize the population and improve its distribution,the individuals with better fitness are given higher weight in each generation of population evolution to improve the contribution of the dominant individuals to the histogram probability distribution model of contemporary population.To improve the global optimization performance of the algorithm,the learning parameters are introduced innovatively to combine the current information with the historical information in the process of updating the histogram probability distribution model.Finally,the HEDA and the comparison algorithms are applied to the simulation test on the data set.Experiments conducted on several different traffic network instances substantiate that HEDA achieves better performance than comparison algorithms on most instances,especially on large-scale network instances.