<p>According to the SDG Report 2025, by 2040, over 2 billion urban residents may experience temperature increases of at least 0.5&#xa0;°C, and nearly 36% of the global urban population could be exposed to average annual temperatures of 29&#xa0;°C or higher. Additionally, the WHO reports that 9 out of 10 people worldwide breathe air exceeding recommended limits. However, many regions lack sufficient ground-based monitoring infrastructure, making comprehensive environmental assessment challenging. In this context, the present study aims to develop an index by integrating thermal and air pollution parameters using an objective weighting framework. The parameters considered include DTL, NTL, NO₂, PM_2.5, SO₂, and CO. DTL and NTL were derived from MODIS, while NO₂, SO₂, and CO were obtained from Sentinel-5P data. Five weighting methods, including statistical and machine learning approaches (EWM, PCA, Autoencoder, RFR, and XGBoost), were applied to estimate parameter importance. While machine learning models captured complex nonlinear relationships, variability among methods was observed. To address this, a two-stage CRITIC-based ensemble approach was employed to derive stable and unbiased weights. The results indicate that NO₂, PM_2.5, and DTL were the most influential parameters. Using these weights, TPSI was computed for both summer and winter seasons from 2019 to 2023. The study area predominantly experienced moderate stress conditions, with higher stress observed in central urban regions and during winter. Sensitivity analysis (± 10% variation) confirmed the robustness of the model. Overall, the proposed TPSI provides a reliable and scalable framework for assessing the combined effects of temperature and air pollution using satellite data, particularly in data-scarce regions. The approach can support urban planners and policymakers in identifying stress-prone areas and enhancing sustainable environmental management.</p>