Loading...

Designing a Transportation System to Minimize the Transportation Cost

Hashemi, Zahrasadat | 2013

618 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 44311 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Ghasemi Tari, Farhad
  7. Abstract:
  8. Transportation problem is one of the most important parts in supply chain management and it offers great potential to reduce costs and improve service quality. So many research efforts, are considering the optimization of the transportation systems and its related aspects. In this thesis, a transportation system according to a real case study, with the aim of minimizing total cost is considered. It is assumed a fleet of vehicles with limited capacity and different fixed and variable costs are shipping different products from a manufacturing plant to their associated nationwide distribution storages. Certain number of storages or depots with deterministic demand is available and each has a demand which is greater than the capacity of the largest vehicle kind. The costs are related to the capacity of each vehicle and the distance of the depots from the manufacturing plant. The goal is to assign a subset of vehicles to each depot, to satisfy their demands and minimize the total transportation costs. To solve the proposed problem, three different genetic algorithms are developed and are named as the genetics algorithm based on the Prüfer number, priority genetic algorithm, and clustering genetic algorithm. To evaluate the performances of the proposed algorithms, 20 test problems were generated and categorized in three classes of small, medium and large size test problem groups. Results of solving the test problems by CPLEX software and proposed algorithms were presented and compared. According to the results of this comparison, genetic algorithm based on Prüfer numbers, based on priority and clustering, and based on priority, has the lowest to highest performance in term of solving time respectively. According to the value of the objective function, genetic algorithm based on priority and clustering, based on Prüfer numbers, and based on priority, has the lowest to highest performance respectively. Also, for small and medium size problems, genetic algorithm based on priority, and for large size problems genetic algorithm based on priority and clustering have the highest performance respectively
  9. Keywords:
  10. Genetic Algorithm ; Clustering ; Transportation Network ; Prioriry ; Improving Quality of Service ; Prufer Numbers

 Digital Object List

 Bookmark

No TOC