MODELING THE LEVEL OF DISSATISFACTION OF THE NEEDS OF RESIDENTS OF SMALL TOWNS IN THE DECISION SUPPORT SYSTEM FOR WATER SUPPLY IN EXTREME CASES

The problem of ensuring water supply to small settlements in extreme cases is considered. It is stated that the authors previously formulated and solved the problem of creating a decision support system, which allows to calculate the optimal route of movement of a water tank with data on the number of residents in each district and the distances between districts. Using a number of parameters (average service time, tank volume, average percentage of the population going out for water, restrictions on the volume of water dispensed, etc.), the system allows you to make a schedule (schedule) of the movement of the tank, as well as give recommendations on increasing the number of these tanks and their optimal distribution by districts, to determine the places of their best location in each district for the maximum satisfaction of all its residents. Currently, a mathematical model has been developed for calculating a special coefficient that allows to assess the level of dissatisfaction of residents' needs in the provision of drinking water. The proposed coefficient contains three components, namely, the ratio of the volume of the recommended dispensing of the maximum possible number of liters per person to the calculated one; assignment of real "time on the road" to optimally calculated; the average relative distance from the calculated center of the district. The created model is added as an additional module to the existing decision-making support system, examples of the system's operation during route calculation and the position of tanks supplying the city of Toretsk, Donetsk region are given. The examples show that the reason for the biggest dissatisfaction is the insufficient number of cars, in second place is their not very good location; available time has almost no effect on the result. Increasing the number of machines to two reduces the first component of the coefficient to unity. The results indicate the possibility of reducing the indicator from 1.305 to 1.087.