Infill development is often used as a remedy for urban sprawl. Sometimes referred to as “land recycling,” it is the reuse and repurposing of obsolete or underutilized buildings and sites in an existing urban area. Infill development frequently results in mixed-use with development, including redevelopment and revitalization, and can lead to well-functioning neighborhoods (but not necessarily gentrification). 

While this land-use change does have appeal, there are some underlying factors that can make it challenging, including the value and size of the parcel, as well as the socioeconomic and demographic characteristics and built environment of the surrounding neighborhood. “Supervised Machine Learning Approaches to Modeling Residential Infill Development in the City of Los Angeles,” published in the Journal of Urban Planning and Development, discusses machine learning algorithms used to analyze the complexity of infill development.

Authors Dohyung Kim, Jongmin Shim, Jiyoung Park, John Cho, and Shathesh Kumar applied five machine learning algorithms to modeling residential infill development in the city of Los Angeles. They examined a comprehensive array of variables, including property characteristics, socioeconomic characteristics, and transportation accessibility, to find the interconnection between residential infill development and the characteristics of parcel and community.  Learn more about the factors involved in developing more efficient smart growth and sustainable housing policies by reading the paper in the ASCE Library at The abstract is below.


While infill development is widely accepted by cities as an alternative to urban sprawl, a very dearth of research has attempted to measure infill development and identify contributing factors to infill development. Filling this research gap, this paper models residential infill development in the City of Los Angeles by employing five machine learning (ML) algorithms. This paper attempts to identify the best-performing ML algorithms by comparing the performance of the ML algorithms. Of the five ML algorithms tested, the random forest (RF) and k-nearest neighbor (kNN) algorithms are selected as the best-performing algorithms. The RF algorithm ranks independent variables from most to least important. Overall, the ranks suggested that residential infill development in the City of Los Angeles is significantly influenced by the physical conditions of property and neighborhood rather than socioeconomic characteristics. Diverse land uses, good housing mixes, and rail transit accessibility also, importantly, contributed to the infill development. This finding suggests that the city’s planning efforts, such as the promotion of accessory dwelling unit (ADU) development and the expansion of rail transit, can create a virtuous circle for sustainable urban development.

Read the full paper in the ASCE Library: