Optimizing roadside green infrastructure: a hybrid MCDM–i-Tree Eco decision framework for high-traffic corridors

Introduction Road-transport corridors in dense metropolitan areas concentrate air pollution while providing little space for mitigation. Methods This study develops an integrated decision-support framework for roadside vegetation selection and ecosystem-service screening at Istanbul’s D-100 Göztepe Junction. Twelve criteria were examined with DEMATEL, SWARA, entropy weighting, and TOPSIS. An initial screening pool was reduced to 27 taxa with complete data across all 12 criteria; these alternatives entered the final category-specific TOPSIS analysis. The primary closeness coefficients were calculated with entropy weights, while SWARA-weighted TOPSIS served as a sensitivity analysis; DEMATEL was retained as a causal diagnostic and was not mathematically combined with TOPSIS weights. Results The entropy solution ranked Robinia pseudoacacia first among trees (CC = 0.5930), Nerium oleander first among shrubs (0.5764), and Hedera helix and Parthenocissus quinquefolia jointly first among climbers/groundcovers (0.758). Entropy- and SWARA-weighted rankings showed strong agreement for trees (Spearman ρ = 0.905) and moderate-to-strong agreement for shrubs (ρ = 0.806) and climbers/groundcovers (ρ = 0.717), although Nerium oleander and Pinus mugo were weight-sensitive. Discussion Scenario outputs increased from S1 to S3, but they are i-Tree-informed screening estimates rather than measured concentration changes or a site-calibrated aerodynamic simulation. The framework supports transparent shortlisting followed by mandatory biosecurity, toxicity, safety, airflow, and field-validation checks before procurement or planting.