Due to the numerous human casualties and major industrial accidents occurring annually, the issues of occupational safety and health of underground coal mine workers come to the forefront. This article applied an approach to the assessment of production risks based on the principles of fuzzy logic. This method makes it possible to effectively balance decisions in the field of industrial safety and competently build a control system. The relevance of the work is conditioned by the scientific and industrial problem of occurrence and spreading of methane and coal dust explosions along mine workings. In such accidents a large number of people die. And it is difficult to exclude such accidents, as it is impossible to solve the problem of degassing completely.The study was based on analyses of data on dangerous coal mines in Russia collected from public sources. Expert analysis made it possible to classify the key risk factors threatening the safety of mining: explosion hazard of gas-dust mixture; risk of mining shock; probability of sudden release of coal-rock mass and gas; threat of water mass and pulp breakthrough into underground workings; danger of endogenous and exogenous fires; risk of rock collapse; hazards associated with blasting operations; hazards resulting from human factor. When processing qualitative indicators, a fuzzy logic system was used to minimise the uncertainty factor. Modelling allowed to identify nine categories of risks of different degrees. The developed methodology demonstrates high potential as an effective risk management tool for building a comprehensive safety system at coal mining enterprises. Its implementation can significantly increase the level of protection of coal mine workers.