Methanol production technology development pathways under multi-agent stochastic evolutionary game: a complex network perspective

The low-carbon transformation of the methanol industry is a pivotal issue under the global carbon neutrality drive. This study proposes an integrated analytical framework combining stochastic disturbance, dynamic game and multidimensional analysis, and develops a tripartite stochastic evolutionary game model (government, fossil-based methanol producer (FMP), renewable methanol producer (RMP)) within a complex network framework to analyze the low-carbon transition of the methanol industry under carbon neutrality. Using Gaussian noise to simulate policy shocks and two-dimensional Monte Carlo simulation, the study traces technology evolution from 2025 to 2055. Results show a three-stage transition pattern under symmetric conditions: policy response, industrial adjustment and stable development period, with government carbon tax strategies playing a central guiding role at an adoption proportion of approximately 98%; while asymmetric settings cause bifurcation, highlighting the importance of initial technological configuration. System resilience is nonlinearly influenced by noise intensity, and stochastic disturbances accelerate technology adoption by altering risk preferences—specifically, FMP's CTM adoption proportion decreases from 16.56% to 8.41%, while RMP's CHTM adoption proportion declines from 16.55% to 8.91%. Government revenue grows consistently—rising from 1.57 × 1016 USD to 4.92 × 1016 USD, FMP revenue declines—from −7.51 × 109 USD to −1.72 × 1010 USD, and RMP revenue shifts from negative to positive—surging from −4.11 × 109 USD to 3.45 × 1010 USD, demonstrating the role of technological learning and scale effects. Tornado diagrams identify key evolving drivers—FMP output coefficient declines from 16545.70 to 10544.76 (2025–2035), whereas RMP counterpart increases from 1272.38 to 21064.34 (2045–2055), offering a quantitative foundation for policy and pathway selection.