Click for new scientific resources and news about Corona[COVID-19]

Paper Information

Journal:   JOURNAL OF FACULTY OF ENGINEERING (UNIVERSITY OF TEHRAN)   March 2002 , Volume 35 , Number 4 (74); Page(s) 587 To 602.
 
Paper: 

A MODEL FOR THE DYNAMIC TRAFFIC ASSIGNMENT TO URBAN TRANSPORTATION NETWORK AND SOLVING IT BY GENERALIZED REDUCED GRADIENT ALGORITHM AND GENETIC ALGORITHM

 
 
Author(s):  ASGHARPOUR M.J., IBRAHIMNEGHAD S.
 
* 
 
Abstract: 

It dynamically models a traffic assignment problem. This model is a nonlinear goal programming mixed with integer variables. It belongs to a class of dynamic system optimal traffic assignment problems. Taking into consideration that these problems are known as NP-Hard problem, most of them don't have polynomial behavior from the time complexity viewpoint. Regarding this fact, this study was carried out to find heuristic algorithms for solving such problems in order to improve the efficiency of solving this kind of models. A special genetic algorithm (GA) was designed in this study, which in addition to constraints handling in decision-making space and improving the generation members, will improve the evolution process and concludes the problem solving with an acceptable speed. Therefore an efficient heuristic algorithm has been proposed to solve such problems. At this point, the results of the solution is compared and analyzed with the results of solving the exact algorithm based on the generalized reduced gradient (GRG). These results demonstrate that solving the suggested model with the GRG and GA doesn't have considerable differences in the amount of a goal objective function. Meanwhile, solving a suggested model with genetic algorithm with population size ([m]<=30) and generation cardinality ([gen]<=300) on numerous samples indicates that the time taken for solving a model is lower, compared to the GRG. If population size and generation cardinality increased from the above values, using the parallel genetic algorithm would be more efficient.

 
Keyword(s): 
 
References: 
  • ندارد
 
  Persian Abstract Yearly Visit 71
 
Latest on Blog
Enter SID Blog