MANAGING THE CAPACITY OF INTERMEDIATE POINTS IN AN EXTENSIVE TRANSPORT NETWORK

Abstract

Relevance. A special case of a transport problem with intermediate points, when the throughput capacity of these points is not specified, is considered, which is important for practice. Purpose. The problem of finding an unknown distribution of the throughput capacity in intermediate points, which minimizes the total transport costs, is formulated. Method. Two methods for solving the problem are proposed. The first one implements an iterative procedure for improving the initial plan for dual model of original problem. The computational scheme at each iteration is a two-step one. At the first step of iteration, a coordinating problem is solved, the solution of which sets the next set of values for throughput of intermediate points. In the second step, this set is used to solve the original transport problem. The resulting solution is tested for optimality. If it is not optimal, then transition to the next iteration is performed. To implement proposed computational scheme, the Nelder-Mead zero-order optimization method was used. Tasks. Statement of problem for controlling the throughput capacities of intermediate points in triaxial transport problem. The developed method has high performance due to the use of the structural decomposition of problem. The method has low computational complexity, which is a significant advantage for practical application. Results. The constructive possibility use of this method has been proved, taking into account a large number of transport-type restrictions. In order to simplify the technology for solving transport problems at each iteration of the algorithm, their dual models are introduced. Due to the fact that computational complexity of proposed method grows rapidly with an increase in the number of intermediate points, a simple alternative approximate method for solving the problem is proposed. Conclusions. The proposed method solves problem of calculating the throughput of intermediate points in "production - delivery - consumption" system.