NEURAL FUZZY NETWORK FOR ASSESSING TECHNOLOGICAL SOLUTIONS FOR FOREST ROADS

Abstract

The article is devoted to the problem of improving road construction technologies, in particular, technological solutions for logging roads. As you know, in road construction, the choice and justification of technological solutions for the road surface is one of the first stages of design, the efficiency of which affects further project as a whole, timing and costs of construction. The solution to such a problem is extremely difficult and, first of all, due to the many interrelated parameters, factors, as well as the uncertainties of data in the problem. The task becomes much more complicated when it is also necessary to take into account the economic indicators of road construction project. But it is in this form that it is of the greatest interest, since these characteristics are often the most important in practice. For these reasons, the problem remains completely unsolved. Therefore, requires further research, as noted, taking into account the uncertainties in the problem. Intelligent systems based on the theory of fuzzy sets, neural networks and their hybrid solutions are proposed for this class of problems, as a result of modern achievements in the field of mathematics and information technologies. Thus, the purpose of this research was to develop a neural network for evaluating technological solutions for logging roads. The result of the research was the development of an adaptive neuro-fuzzy network such as ANFIS, which allows calculating the cost of the road surface depending on the main technological and initial financial parameters. The neural network can be recommended for the design of forest roads, as well as for rapid assessment of the effectiveness of various technological solutions during competitive (tender) selection.