摘要

On the basis of travel time reliability, this paper puts forward the concepts of absolute reliability and relative reliability to better assess road network; simultaneously, regional traffic road network reliability formula is deducted according to series and parallel connection of link, OD pair and road network. Finally, a bi-level programming model is established to optimize road network. The upper model, integrating travel resistance and travel time reliability in the objective function, optimizes and chooses investable links and the lower model guarantees stochastic user equilibrium. Both Monte Carlo Stochastic Simulation based genetic algorithm and Frank-Wolfe algorithm are employed as solutions. And Nguyen-Dupuis net work calculation sample shows different investment has different effect on road network optimization. Even with the same investment the effect is more remarkable in transition intervals than that of peak and flat intervals.

  • 出版日期2012

全文