Resumen
The article deals with the interrelated processes of cargo transportation on the main road transport network. The problem of distribution of available vehicles was stated. The flows of incoming orders are stochastic, but have no features of the simplest. On the specified territory, the orders for long distance transportations appear and are repeated with random periodicity during a fixed period. Each order has its time window. Vehicles of one carrier are placed on a transport network in random order, according to the latest run performed. To execute the orders, motor-vehicle trains take the cargo at the location point, or in the absence of loads, move to the nearest transport point, where such orders appear. The typical situation, when even if there are enough vehicles, clients are denied transportation or vehicles have to stand idle or travel unloaded, was analyzed.The simulation modeling procedure was developed. With the help of the random number generator, the set of coordinates of the points of departure and delivery of random order cargo and the points where vehicles are primarily located, as well as time windows, transportation volumes, and periodicity of orders are obtained. The service is provided according to one of three strategies: without a no-load run and waiting, with a no-load run, with full forecast of the upcoming process. The number of refusals due to the absence of transport or its being engaged was calculated. The parameters for several cycles were calculated. The order handling strategy is implemented based on the correction of decisions of subjects of transportation process at obtaining additional information about previous iterations. The decisions of subjects are limited to the carrier?s intentions. The procedure is applied in order to research the activity of the transport enterprise in the south-eastern territory of Ukraine during the agricultural cargo transportation during the harvest period. The indicators of the incoming flow service were found to have a fluctuating character. Three strategies were compared. The advantages and disadvantages of the application of no-load run, expectations, forecasting, and vehicles distribution by the volume of existing work were identified