Iranian Sustainable Building Scientific Association International Journal of Urban Management and Energy Sustainability 2538-1628 7 1 2026 01 01 Assessing the Influence of Climate on a Framework for Urban Neighborhood Site Selection under a Sustainable-Development Approach Using Artificial Neural Networks and Exploratory Factor Analysis (Case Study: Tehran City, Iran) 274 286 738954 10.22034/ijumes.2026.738954 EN Zoherh Rezaiekavanroodi Department of Civil Engineering, Fir.C., Islamic Azad University, Firoozabad, Iran 0009-0005-6255-8629 Yahya Kazemi Department of Civil Engineering, Fir.C., Islamic Azad University, Firoozabad, Iran 0009-0001-4299-9559 Journal Article 2026 01 12 Site selection for urban neighborhoods is a multi-criterion, spatially explicit problem that becomes markedly more sensitive under a changing climate, where urban heat island intensity and outdoor thermal comfort strongly condition livability and sustainability. This study develops and demonstrates an integrated framework that couples Geographic Information Systems, exploratory factor analysis and artificial neural networks to assess how climatic factors with an emphasis on UHI and thermal comfort shape a sustainable-development-oriented framework for neighborhood site selection. A pool of climatic, environmental, social, economic and physical–access indicators are assembled from GIS layers at the neighborhood scale. EFA reduces the correlated indicators to a small set of interpretable latent factors. The factor scores then serve as inputs to a multilayer-perceptron ANN trained to predict a neighborhood sustainability suitability index, which is validated with R², RMSE and MAE and interrogated with a sensitivity analysis. In the demonstrative case, EFA retained four factors, climate & thermal, access & mobility, socio-economic, and physical form explaining a large share of the variance, with the climate & thermal factor. The ANN reproduced the reference suitability index with high accuracy on the held-out set, and the sensitivity analysis ranked UHI-related variables as the most influential. Neighborhoods were classified into five suitability tiers. All of these values are synthetic and illustrate the framework’s behavior. Coupling EFA with ANN inside a GIS pipeline provides a transparent, reproducible and transferable framework for climate-sensitive, sustainability-oriented neighborhood site selection, and foregrounds UHI mitigation as a first-order siting criterion. The framework is ready to be populated with empirical Tehran data to produce validated, city-specific results. https://www.ijumes.com/article_738954_9fcf04a1ebc0797612f4162cab197cd0.pdf