A Novel LambdaRank-Based Method for Assessing the Cost-Effectiveness of Power Enterprise Projects

The current evaluation methods for assessing the cost-effectiveness of the power grid projects predominantly focus on large-scale main grid initiatives that require significant investments. Nonetheless, an important deficiency exists in thorough analyses specifically addressing distribution network projects, which generally entail more modest investments. Existing evaluation systems face challenges: limited applicability of evaluation indices, complex mathematical models, and outcomes that lack rationality and scientific rigor. This research employs the LambdaRank algorithm to construct a project optimization model to address these issues. The model utilizes unsupervised learning techniques to select relevant project features, followed by applying the AHP-EMW method and fuzzy comprehensive evaluation to analyze and assess high-quality projects. A generalized post-evaluation system is subsequently established, tailored to the specific characteristics of these projects. By leveraging the LambdaRank ranking algorithm to prioritize power grid projects, the model provides a scientific reflection of both professional and economic impacts, thereby addressing the shortcomings in power grid enterprises' input-output evaluations of distribution network projects. The findings show that the proposed method enhances scientific rigor and transparency in project prioritization, leading to a more rational and systematic evaluation process.