Comprehensive evaluation of air pollution at Zhenjiang port based on the un-weighted TOPSIS method

Conventional TOPSIS approaches for comprehensive air pollution assessment are often constrained by their reliance on pre-assigned weights and high sensitivity to outliers. To address these limitations, an Un-weighted TOPSIS (UW-TOPSIS) method was applied to evaluate the air quality of Zhenjiang Port from September 2021 to September 2024 based on six criteria pollutants (PM_2.5, PM₁₀, SO₂, NO₂, CO, and O₃).Meanwhile, AHP-TOPSIS (incorporating expert-derived weights) and EW-TOPSIS (using entropy weight determination) were also employed for comparative analysis. Model performance was quantitatively evaluated against normalized Air Quality Index (AQI) scores using the relative error (ε). Analysis revealed that Zhenjiang Port's pollution was predominantly driven by PM₁₀ (37.93% of days) and PM_2.5 (36.75% of days), exhibiting a distinct seasonal pattern of winter highs and summer lows. Methodologically, AHP method yielded a combined weight of 0.6282 for PM, showing a pronounced bias, whereas the EW produced a more balanced weight structure (0.3334 combined) but was highly sensitivity to gaseous pollutants, particularly NO₂ and CO. Using July 2023 (low pollution) and January 2024 (high pollution) as representative periods, UW-TOPSIS demonstrated superior stability. During the low-pollution period, 68.75% of days exhibited ε < 0.06 under UW-TOPSIS, outperforming AHP (54.84%) and EW (48.39%); during the high-pollution period, UW-TOPSIS maintained stability with only 32.3% 32.3% exceeded ε = 0.1, significantly outperforming AHP (64.5%) and avoiding the extreme error peaks (up to 0.463) observed in the EW method. Ultimately, AHP's reliance on subjective weighting amplified errors during shifts in pollution composition, while the EW method proved excessively sensitive to outliers, yielding volatile evalution outputs. By eliminating explicit weight assignment in favor of bounded weight constraints, UW-TOPSIS substantially enhanced stability and robustness. These findings confirm its reliability of UW-TOPSIS under multi-pollutant conditions, presenting a robust framework for developing composite pollution indices and the evaluating regional air quality management efficacy.