Energy-intensive manufacturing industries are characterised by high pollution and heavy energy consumption, severely challenging the ecological environment. Fortunately, environmental, social, and governance (ESG) can promote energy-intensive manufacturing enterprises to achieve smart and sustainable production. In Industry 4.0, various advanced technologies are used to achieve smart manufacturing, but the sustainability of production is often ignored without considering ESG performance. This study proposes a strategy of edge-cloud cooperation -driven smart and sustainable production to realise data collection, preprocessing, storage and analysis. In detail, kernel principal component analysis (KPCA) is used to decrease the interference of abnormal data in the eval-uation results. Subsequently, an improved technique for order preference by similarity to ideal solution (TOPSIS) based on the adversarial interpretative structural model (AISM) is proposed to evaluate the production efficiency of the manufacturing workshop and make the analysis results more intuitive. Then, the architecture and models are verified using real production data from a partner company. Finally, sustainable analysis is discussed from the perspective of energy consumption, economic impact, greenhouse gas emissions and pollution prevention. Funding Agencies|Youth Innovation Team of Shaanxi Universities ?; Special ConstructionFund for Key Disciplines of Shaanxi Provincial Higher Education; Natural Science Basic Research Plan in Shaanxi Province of China [2022JQ-37]; Shaanxi Provincial Education Department [22JK0567]; Project of National Natural Science Foundation of China [62271390, 51905399]; Postgraduate Innovation Fund of Xian University of Posts and Telecommunications [CXJJDL2022012]