Estimation of forest machine operational efficiency factors by regression analysis

Abstract

The paper deals with the estimation of operational factors affecting forest machine maintainability. The main goal of this study is to substantiate and test the order of operational factors estimation using correlation and regression methods. A brief description of statistical methods of operational factor analysis is presented in the first part of the paper. The second part of the paper presents the obtained multiple regression equation and determined values of beta coefficients. Operation time of forest machine, staff employment period and servicing base technological infrastructure are accepted as independent variables determining servicing time. The interaction between independent variables and servicing time is presented as a multiple equation of linear regression. Pair correlation coefficients are used as indices of close linkage among the analyzed variable quantities. The system of normal equations is used to determine the regression coefficients of the linear model. The analysis of the obtained regression equation is given in the final part of the paper. The coefficient of determination is used as the accuracy and completeness criterion of factor selection. According to the obtained value of the criterion, it was concluded that the level of completeness of factor selection is sufficiently high. The statistical significance of regression coefficients is verified using Student's test. All considered factors are recognized as significant for servicing time estimation according to the results of verification. Furthermore, operation time of forest machine is recognized as the general maintenance factor affecting the duration of technical impacts. The effects of the staff employment period and the servicing base technological infrastructure differ slightly from each other; however, the servicing base technological infrastructure factor is more significant.