The prediction of coal and gas outburst risk can effectively prevent underground coal mine accidents. Due to the overlapping, coupling and complexity of coal and gas outburst in the development process, coal and gas outburst is generally Gaussian distribution or nearly Gaussian distribution, especially when the sample data in the research area is not accurate or the information is insufficient, the traditional evaluation model and method have certain limitations. To further improve the scientific and accurate prediction of coal and gas outburst risk level, a coupling model of coal and gas outburst risk assessment is established based on Monte Carlo stochastic simulation (MCS) and triangular fuzzy number (TFN) theory. Firstly, the index weight measurement value is determined by using expert opinions and AHP method. Then, the risk level and risk importance of coal and gas outburst risk assessment index are quantitatively described by using fuzzy semantics with five-level classification standards. Finally, the confidence interval of the comprehensive risk value of the coal mines to be evaluated in the research area is established. The research results show that after 20,000 simulation experiments with the coupling model, the calculation results have converged, and the confidence interval value of the system comprehensive risk simulation value of each coal mine is 95%, which can provide relevant decision support for the prevention and control planning of coal and gas outburst.