A foresight study of carbon emission control in energy supply: A case of Mexico Academic Article in Scopus uri icon

abstract

  • The purpose of this paper is to explore an integrated machine learning and expert-based strategic foresight methodology for guiding policies aimed at reducing carbon emissions. The approach is pursued to provide data-driven predictive analytics combined with structured policy road mapping. Gaussian Process machine learning models are developed to forecast carbon emissions accurately based on relevant parameters. A phased foresight framework, leveraging collective expert wisdom, is also developed to generate actionable short-, medium-, and long-term strategic plans. This work finds that the Gaussian Process models exhibited strong predictive accuracy with a 0.93 correlation, providing a means to quantify expected outcomes under proposed policy scenarios. Based on expertise, phased foresight roadmaps provide tangible strategies to guide practical implementation. Before broader socioeconomic incentives drive widespread adoption, short-term technical solutions establish the foundation. Also, innovation ecosystems ensure long-term progress for the carbon emission control in the case study. Integrated and evidence-based carbon policy informational tools as well as structured strategic plans, address the pressing need for actionable knowledge to support national and global decarbonization efforts. This methodology offers policymakers robust analytical capabilities for modeling options and scenario impacts, as well as tangible recommendations. Contributions address critical gaps preventing carbon reduction commitments from being realized. © 2025 Elsevier Ltd

publication date

  • August 1, 2025