THE RISK ASSESSMENT OF UNDERGROUND UTILITY NETWORKS DAMAGE WITH USING FUZZY INFERENCE

ABSTRACT

Design risk is an inherent element of construction works, and one of its sources is the spatial data quality. Cases of damage to underground pipes and cables are often the causes of accidents, delays or construction stoppages. One of the directions of research on this risk is fuzzy logic. Fuzzy logic is used to describe complex or poorly defined phenomena that are difficult to describe by using probabilistic methods. The paper presents the method of risk assessment of underground utility networks damage which is based on the scheme of the fuzzy inference system. In the first stage, linguistic variables were defined and appropriate values were assigned to them (on the basis of risk factors). Affiliation functions for linguistic variables were modeled in the expert approach. In the second stage, qualitative output sets were determined by using the designed rule base. At the last stage of fuzzy inference, the resulting values were converted into numerical values. The center of gravity method was used in the defuzzification process. The developed method allows to estimating the assessment risk of underground utility networks damage for spatial data with various quality classes. This method will also be the basis for creating large-scale risk maps. The proposed solution, based on fuzzy logic, can be an alternative to conventional risk assessment methods.

Keywords: spatial data quality, cadastre, GIS