Loading...

Modeling and Solution Algorithm for Multi-Path Railroad Blocking Problem under Demand and Travel Time Uncertainty

Mohammad Hassani, Reza | 2017

629 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 49639 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Shafahi, Yousef
  7. Abstract:
  8. The railroad blocking problem emerges as an important issue at the tactical level of planning in freight rail transportation. This problem consists of determining the paths of each shipment which minimize the total cost subject to the key constraints such as limitation on travel time, limitation on number of dispatched trains, and limitation on the number of cars classified at each station. The multi-path railroad blocking model is developed and then a new heuristic algorithm based on the Lagrangian relaxation technique to solve the realistic rail network is introduced.To consider the uncertainty which is inherent in the demand and travel time (as a supply indicator of rail system), a two-stage stochastic programming with recourse is presented, and a solution algorithm based on the L-Shaped method modifying with a new initial solution is introduced.Since the solution of the blocking problem is as an input for the operational decisions, the robust formulation for the blocking problem is presented to ensure the stability of the blocking solution subject to uncertainty inherent in real-world problem. The robust model combines integer variables and nonlinear functions, and it causes the robust model to be difficult to solve for the large rail network. Therefore a branch-and-cut algorithm is used to solve this model. The computer simulation can be used to test the results of analytical models in an environment close to real conditions, and this simulation is able to simulate the train movements and station operations. The output of simulation shows that as the number of trains increases, the total cost and total delay of robust and stochastic solutions rapidly decrease in comparing those of deterministic solution.Iranian railway network as a case study to test the proposed models is selected. The application of the simulation shows that stochastic and robust solutions cause the reduction of 48 and 17 percent in total cost in comparing that of the deterministic model
  9. Keywords:
  10. Optimization ; Robust Optimization ; Computer Simulation ; Stochastic Modeling ; Railway Transport ; Uncertainty ; Pailroad Blocking Problem

 Digital Object List

 Bookmark