SIMULATION OF CARS ACCUMULATION PROCESSES FOR SOLVING TASKS OF OPERATIONAL PLANNING IN CONDITIONS OF INITIAL INFORMATION UNCERTAINTY

Abstract

Purpose. The article highlights development of the methodological basis for simulation the processes of cars accumulation in solving operational planning problems under conditions of initial information uncertainty for assessing the sustainability of the adopted planning scenario and calculating the associated technological risks. Methodology. The solution of the problem under investigation is based on the use of general scientific approaches, the apparatus of probability theory and the theory of fuzzy sets. To achieve this purpose, the factors influencing the entropy of operational plans are systematized. It is established that when planning the operational work of railway stations, sections and nodes, the most significant factors that cause uncertainty in the initial information are: a) external conditions with respect to the railway ground in question, expressed by the uncertainty of the timing of cars arrivals; b) external, hard-to-identify goals for the functioning of other participants in the logistics chain (primarily customers), expressed by the uncertainty of the completion time with the freight cars. These factors are suggested to be taken into account in automated planning through statistical analysis – the establishment and study of the remaining time (prediction errors). As a result, analytical dependencies are proposed for rational representation of the probability density functions of the time residual distribution in the form of point, piecewise-defined and continuous analytic models. The developed models of cars accumulation, the application of which depends on the identified states of the predicted incoming car flow to the accumulation system, are presented below. In addition, the last proposed model is a general case of models of accumulation processes with an arbitrary level of reliability of the initial information for any structure of the incoming flow of cars. In conclusion, a technique for estimating the results of simulation the cars accumulation was proposed to optimize the transportation process, depending on the chosen criterion. Findings. The developed methodology of simulation of cars accumulation process was reflected in the dynamic models created with the participation of the author and implemented on the Belarusian Railroad on the basis of IAS SMD CT. They are designed to provide operational planning of the transportation process on the basis of methods that allow assessing technological risks. Originality. The innovative component of the work is due to the expansion of existing models of the accumulation of cars for cases of uncertainty of the initial information. Thus, the earlier deterministic models are particular cases of the proposed model. Also, unlike existing ones, the technique allows to take into account the influence of random processes in a complex manner. Due to this, technological risks can be further assessed and the necessary regulatory measures can be implemented promptly. In general, the results obtained by modeling the proposed method allow to improve the quality of output solutions in the system of shift and daily operational planning, increasing the reliability of operational plans. Practical value. To date, there is a favorable opportunity to use the proposed methodology of modeling in existing in the railway transport systems of automated operational planning for operational work, including the optimization of local railways and the solution of a number of urgent tasks of train formation.