AI in urban planning

Abstract

Urban planners in Singapore have been facing challenges such as land scarcity in the past. In recent years, climate change and sustainability are constantly brought up with regards to the land use in Singapore. Traditional methods such as manual land surveys and analytic models based on census data were conducted. However, they are either labour intensive which require a huge amount of time or fail to adjust to the dynamic environment. With the advancement of AI and technology nowadays, urban planners can tap on this potential to revolutionise their methods of planning land use configuration. In this paper, a machine learning based framework for land use configuration is proposed. A list of Point-of-Interests and their relevant information are included in a new and distinctive dataset created. Machine learning model is trained with this dataset to accurately predict and rank candidate locations based on their google ratings. An optimal site is the highest ranked candidate location.