Loading...

Optimum Highway Alignment Using Parallel Processing

Kazemi, Farzan | 2012

736 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 43699 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Shafahi, Yusof
  7. Abstract:
  8. Optimum highway alignment problem is among the most substantial but large and complicated topics in transportation area. Infinite number of feasible solutions, numerous local optima and the constrained feature of the problem, associated with complex and mainly non-linear constrains, has put an extra effort into the problem-solving process. Moreover, the problem has a sensitive objective function to input variables. Due to the nature of the problem, conventional methods fail to find any exact solution, and so, using heuristic methods during the solving procedure is inevitable. Large-scale, complexity and sensitivity of the problem leads to a solution which is time-consuming and high-computational demanding to achieve a sub-optimal solution for a real scale problem, even by using heuristic algorithms. In this study, a parallel processing method based on master-slave architecture is proposed to solve real scale problems. In this respect, parallelization is done in synchronous and asynchronous ways, after picking up a suitable heuristic method based on the particle swarm algorithm. For assessing parallelization performance, corresponding indexes such as speedup and parallel efficiency indexes were employed. Data driven from SRTM database, which contains the elevation of ground points and some insurmountable areas, was used to solve real-world problems. New functions were defined and used as the penalty for exceeding constrains in solving procedure. The implementation of the proposed method was investigated on two real-world problems and represented it could cope with solving such problems, even in a large-scale case
  9. Keywords:
  10. Parallel Processing ; Particles Swarm Optimization (PSO) ; Trajectory Optimization ; Optimum Highway Alignment ; Asynchronous Parallelization ; Master-Slave Architecture ; Synchronous Parallelization

 Digital Object List

 Bookmark