Smart Urban Planning

A Systematic Review of Predictive Models of Urban Use and Occupation

Abstract

ABSTRACT

Objective – This study aims to conduct a systematic literature review on the use of software, applications, and predictive models applied to urban planning. The research seeks to identify how geoprocessing and machine learning have been integrated to optimize territorial management, enhance public policy efficiency, and promote sustainable development in urban contexts.

Methodology – The research adopted a systematic literature review methodology, following a rigorous protocol for identifying, selecting, and analyzing scientific publications. The process included: (i) formulation of the research question; (ii) definition of inclusion and exclusion criteria; (iii) selection and analysis of full-text articles from databases such as CAPES, Web of Science, and Scopus; (iv) evaluation of study quality; and (v) synthesis and dissemination of results. The temporal scope covers publications from 2015 to 2025, focusing on the application of geoprocessing and machine learning in urban planning.

Originality/relevance – This study fills a theoretical gap by systematically integrating geotechnology and machine learning in urban planning, highlighting innovative applications in different contexts. Its academic relevance lies in consolidating existing knowledge, identifying emerging trends, and proposing new directions for future research. The scarcity of studies that address these technologies in an integrated manner in complex urban environments highlights the novelty of this research.

Results – The main findings indicate that combining geoprocessing and machine learning significantly enhances predictive capabilities in urban planning. Reviewed studies demonstrate improvements in environmental risk management, accessibility to public services, infrastructure optimization, and demographic and socioeconomic pattern analysis. The integration of these technologies facilitates the identification of risk areas, infrastructure gaps, and sustainable development opportunities.

Theoretical/methodological contributions – This research demonstrates how predictive models based on machine learning can be integrated into spatial analyses to improve the understanding of urban dynamics. Methodologically, it highlights the use of advanced techniques such as AHP and clustering algorithms, which enhance the accuracy of spatial analyses and strengthen data-driven public policy formulation.

Social and environmental contributions – The study indicates that applying geotechnology and machine learning can improve equity in access to essential public services, such as healthcare and education, while supporting policies that promote social inclusion and reduce inequalities. Environmentally, these tools aid in the sustainable management of natural resources, disaster mitigation, and the planning of more climate-resilient cities. The study emphasizes the importance of data-driven governance to foster fair, inclusive, and sustainable urban development.