Transforming Urban Planning through Machine Learning: A Study on Planning Application Classification using Natural Language Processing

Planning for sustainable urban growth is a pressing challenge facing many cities. Investigating proposed changes to the built environment can provide planners and policymakers information to understand future urban development trends and related infrastructure requirements. It is in this context we have developed a novel urban analytics approach that utilises planning applications (PAs) data and Natural Language Processing (NLP) techniques to forecast the housing supply pipeline in Australia. Firstly, we implement a data processing pipeline which scrapes, geocodes, and filters PA data from council websites and planning portals to provide the first nationally available daily dataset of PAs that are currently under consideration. Secondly, we classify the collected PAs into four distinct urban development categories, selected based on infrastructure planning provisioning requirements. Of the five model architectures tested, we found that the fine-tuned DeBERTA-v3 model achieves the best performance with an accuracy and F1-score of 0.944. This demonstrates the suitability of fine-tuned Pre-trained Language Models (PLMs) for planning text classification tasks. Finally, the model is applied to classify and map urban development trends in Australia’s two largest cities, Sydney and Melbourne, from 2021-2022 and 2023-2024. The mapping affirms a face-validation test of the classification model and demonstrates the utility of PA insights for planners. Holistically, the paper demonstrates the potential for NLP to enrich urban analytics through the integration of previously inaccessible planning text data into planning analysis and decisions.