A New Method for Implementing the Interval Concept of Transport Demand Modeling
Keywords
Abstract
The article examines the peculiarities of the OD-matrix generation within the framework of the interval concept of transport demand modeling, which consist in the absence of guarantees of successful distribution of known departure and arrival capacities of transport districts of the transportation object among the matrix cells, in the presence of additional, to capacities, restrictions. The possibilities of full distribution of demand on the OD-matrix in the presence of such restrictions are analyzed and two methods to overcome the existing problem are proposed. For a small number of restrictions, a two-stage matrix filling algorithm is formed. The first stage consists in the usual distribution of transport demand volumes using a random number generator, the second is in the redistribution to open cells of those values that fell into closed matrix cells using closed cycles known from the transport task. For a significant number of restrictions that arise when modeling matrices of route correspondences, or in other cases in which the modeling of transport demand using closed loops becomes impossible, has been created a new method of implementing the interval concept of demand modeling. It is based on repeated use of the basic method of distribution of transport demand without restrictions, which always leads to a complete distribution of capacities according to the OD-matrix. Restrictions imposed on the matrix are taken into account by removing from its current instance those values that are subject to restrictions. The deleted values form the conditions for the generation of the next instance of the matrix until all the departure and arrival capacities of the transport districts are distributed among such matrices. The final matrix is the sum of all the intermediate matrices created in this way. This method of forming OD-matrices allows guarantying a positive result of the generation of random matrices with restrictions and to ensure the full implementation of the interval concept of transport demand modeling.
