A Local Search Algorithm Leveraging the Reasoning Abilities of Large Language Models for Solving Urban Charging Station Location Problem

To enhance the charging experience for electric vehicle users and facilitate the widespread adoption of electric vehicles, the paper introduces the urban charging station location problem (UCSLP), which has significant practical implications in the field of electric power. The UCSLP involves two conflicting objective functions: minimizing user waiting time and maximizing the revenue of charging stations. To comprehensively evaluate the solution quality in this multi-objective context, we propose a local search algorithm that leverages the reasoning abilities of large language models. Experimental results demonstrate that large language models can effectively understand the multiple objective functions of the problem and thoroughly assess solution quality. The reasoning abilities of the large language model improve the effectiveness of the local search algorithm in solving the UCSLP and have potential applications for other multi-objective optimization problems.