AI-Enhanced Decision Support for Optimized Carbon Quota Allocation Across Industrial Sectors

To mitigate greenhouse gas emissions, governments allocate carbon quotas to industries under Emission Trading Systems. Allocating these quotas involves complex modeling challenges that require effective distribution strategies, which traditional Multi-Criteria Decision Analysis (MCDA) methods struggle to address comprehensively. These methods often fall short of encapsulating the preferences of policymakers and the distinct characteristics of different industries. In response, this paper proposes a novel framework termed the Large Language Model Carbon Quota Allocation Assistant (QAA). The QAA framework combines classical decision-making approaches with a robust expert knowledge base, enhancing the adaptability of allocation schemes to meet specific user demands and industry-specific characteristics. We validate this model through an empirical study involving the allocation of carbon quotas among six major emitting industries in Guangdong, China. The results are analyzed and compared against those derived from conventional linear combination weighting methods. Our findings substantiate the superior efficacy of the QAA, demonstrating its capability to optimize the monetary value of carbon quotas by harmoniously integrating user requirements and expert insights. This study not only underscores the potential of large language models in environmental policymaking but also sets a precedent for future research in this domain.