A machine-learning assisted multi-cluster assessment for decarbonization in the chemical fiber industry toward net-zero: A case study in a Chinese province

China has pledged to achieve carbon neutrality by 2060, requiring deep decarbonization in its manufacturing sector, aligning with sustainable development goals such as climate action and responsible production. Notably, China's chemical fiber industry contributes over 70% of global production, facing challenges in net-zero transition due to differences in enterprise scale and energy efficiency. This study proposed an assessment framework for the decarbonization pathway for this type of manufacturing industries, use the chemical fiber industry as a case study. A hybrid model based on machine learning was introduced to predict the industry's energy consumption, while multiple-cluster standards were established to assess energy efficiency improvement potential. Monte Carlo simulation was employed to analyze the carbon trading impact on industry decarbonization. Using a Chinese province's chemical fiber industry as a case, results suggest its carbon emissions could reach 1.58 × 107 tCO2 by 2030, and energy efficiency enhancements could reduce emissions by approximately 22.6%. Achieving carbon neutrality would cause the industry to reduce profits by approximately 10%∼15% on higher-priced emissions trading system (ETS), unless additional carbon reduction techniques are adopted. This assessment framework can be applied to study decarbonization transitions in other manufacturing industries. •Industry energy consumption and CO2 emission prediction based on a hybrid model assisted by machine learning•Comprehensive analysis of the industry's energy efficiency improvement potential based on a multi-benchmark energy intensity benchmarking scheme•Comparative analysis of carbon trading strategies under different decarbonization scenarios based on Monte Carlo simulation