FORECASTING VOLUMES OF ACCUMULATED WASTE

Abstract

Problem statement. The task of assessing the volume of accumulated waste in the territory of the Dnipropetrovsk region, which is constantly increasing, occupying a larger area and causing damage to the environment, is under consideration. To solve this forecasting problem, it is necessary to create a regression mathematical model for statistical evaluation and analysis of the factor variables’ influence to the total volume of accumulated waste. The purpose of the article. Creation of a mathematical model for forecasting assessment of possible volumes of accumulated waste in the territory of the Dnipropetrovsk region by adjusting the volumes of factor variables. Methodology. Analysis of the dynamics of changes in the volumes of generated, utilized, incinerated, removed waste and capital investments and costs for waste management and establishment of trends in their changes based on descriptive statistics. Application of correlation analysis methods to establish the most statistically significant relationships between factor variables and the resulting feature. The use of regression analysis methods to obtain the coefficients of the regression mathematical model and statistical indicators that explain the probability of the significance of these coefficients. Scientific novelty. A multiple regression mathematical model was developed, which takes into account the factor variables affecting the process of waste accumulation in the territory of the Dnipropetrovsk region. Practical value. The developed regression mathematical model makes it possible to estimate the required amount of investment and forecast the amount of current costs to adjust the total amount of accumulated waste. Conclusions. A mathematical model was created to analyze the volume of accumulated waste in the territory of the Dnipropetrovsk region. Based on this model, the volumes of accumulated waste were calculated. The average value of the relative error of the calculated data is 1.03 %, while the maximum value of the error is 1.97 %, which confirms the adequacy of the developed mathematical model.