In view of the challenges in quantifying carbon emission data due to fuzziness and randomness in the evaluation process of near-zero carbon ports,high-frequency indicators are screened using literature frequency statistics.Combined with the characteristics of port energy activities and the expert consultation method,an evaluation system is built,covering four aspects:safeguard mechanisms,operational management,business activities,and carbon reduction effectiveness,with ten secondary indicators and twenty-two tertiary indicators.The cloud model theory is introduced,utilizing three numerical characteristics of expectation,entropy,and hyper-entropy to determine weights and convert qualitative evaluations into quantitative descriptions.An empirical analysis is conducted on three typical inland ports in Anhui Province.The results show that the Ma’anshan Port specialized dry bulk terminal scores the highest at 89.6 points.However,its entropy and hyper-entropy values are relatively high,indicating a “high score with instability”.The Hefei Port International Container Terminal and the Wuhu Port International Container Terminal score 84.9 and 85.8 points,respectively,and are primarily constrained by hardware-related emission reduction efficiencies such as energy structure.The results of this study verify the effectiveness of the cloud model in handling non-deterministic evaluations,reveal the bottlenecks in the differentiated development of inland ports,and provide a quantitative basis for formulating precise emission reduction strategies.