Thermal comfort in outdoor environments is often assessed using the mean radiant temperature (Tmrt) along with other factors. This study aims to develop a machine learning model that predicts Tmrt based on air temperature (Ta), relative humidity (RH), and wind speed. Field measurements were collected, and Tmrt was calculated using the globe thermometer method and compared with ENVI-met simulations, which achieved an R2 value of 0.916. Results show that ML model exhibits consistent performance across training, validation, and test datasets, with only minor variations in error metrics. Study points under tree shade during the afternoon exhibit lower Tmrt compared to other locations, underscoring the influence of vegetation on thermal comfort. The ML model trained for predicting Tmrt demonstrates a high level of predictive accuracy, with excellent performance across training, validation and test datasets which underscore the potential of artificial neural networks in predicting Tmrt within outdoor thermal environments. However, the study has several limitations. Future research should address these limitations by incorporating data from multiple days, different locations to further enhance the model’s robustness and applicability.