Research On Traffic Network Optimization Based On Space-Time Accessibility

The traditional traffic planning is based on mobility,which aims to improve the operational efficiency of motor vehicles and reduce congestion,instead of putting the travel subject in the first place.In addition to ensuring the balance of supply and demand of traffic resources,the urban traffic system needs to ensure the service level of the transportation system,to allow the traveler get a larger trip utility with reasonable travel cost.Under the constraint of resources and environment,it is a new requirement for traditional transportation planning to improve the ability of obtaining job opportunities and public services.As a derived requirement,travel is generated from the needs of completing the activity.Individual space-time accessibility takes the activity-based travel as an analysis unit,which aims to meet the needs of completing the activity.Compared to the space accessibility,the individual space-time accessibility emphasizes whether the needs of performing activities are satisfied,which highlights the idea of people-oriented.On the basis of the individual spacetime accessibility,the optimization design of traffic network can provide more reasonable suggestions for decision makers in deciding road construction.This paper uses the space-time network to show the individual movement trajectory and to express the individual space-time accessibility.Here,the individual's physical moving trajectory is expressed as the individual space-time path,and it is influenced by the factors such as travel time budget,activity-performing time and time-dependent traffic network.If the individual space-time path is feasible,which means he or she can finish the activity-based travel under the travel time budget,with the allowed travel speed in the traffic network.This paper constructs a traffic network optimization design model based on space-time accessibility.Taking the activity-based travel as the analysis unit,the model aims at maximizing the overall space-time accessibility of the network.The optimization strategy is to select the best segments from the roads set,within the construction capital budget.The decision variable includes the road selection variable and the road construction variable,and the constraints consisits of the space-time flow balance constraint,the activity performing constraint,the construction budget constraint,and the coupling constraint between space-time arcs and physical links.The Lagrangian relaxation decomposition method is used to solve the network design model.The Lagrange relaxation technique is used to absorb the "difficult constraint" in the original problem of network design into the objective function,and we can get the Lagrangian relaxation problem.In the relaxation problem,the objective function and the constraint condition can be divided into two groups according to the variable type,so the relaxation problem is decomposed into two kinds of subproblems by the Lagrangian decomposition technique,with a reduction in the difficulty of solving the original problem.Compared with the original problem,the Lagrange relaxation problem expands the feasible field of solution,which leads to the gap between the objective function value of the relaxation problem and the original problem.To narrow the gap,we solve the Lagrangian duality problem,update the Lagrange multiplier of "difficult constraint",shrink the feasible region of solution of Lagrange relaxation problem,narrow the gap of the target function value between the relaxation problem and the original problem,and force the solution of the dual problem to approach the optimal solution of the original problem gradually.Finally,the paper uses numerical examples to analyse the performance of the space-time accessibility.By conducting the example,the influence of input parameters such as investment budget and travel time budget on network accessibility is analyzed,and the proposed model is applied to specific network examples.