Leveraging Street Level Predictive Modelling with Green Infrastructure for Urban Resilience
S.R.H.W. Tew (TU Delft - Architecture and the Built Environment)
D. Cannatella – Mentor (TU Delft - Urban Data Science)
C. Forgaci – Mentor (TU Delft - Urban Design)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Urban resilience is an evergrowing issue as cities face environmental, social and infrastructural challenges. Green infrastructure is currently being used as one of the main approaches in achieving urban resilience by providing multifunctional benefits, such as improving microclimate, enhancing biodiversity, and increasing accessibility to nature. However, through this process the evaluation and optimisation of green infrastructure at the street level remains a methodological challenge, particularly in the case of integrating multi dimensional indicators into predictive models using nature based solutions.
This study develops a machine learning based framework in order to assess and optimise green infrastructure in urban areas at a street level scale. Through the use of publicly available spatial datasets on environmental, biodiversity and morphological factors, the project constructs a comprehensive dataset incorporating factors such as urban heat islands, green infrastructure distribution, and green space accessibility.
Modelling techniques such as gradient boosting regression and random forest regression are employed as a regression technique to predict urban resilience related targets. These will provide insights into how resilient different areas of Rotterdam can be. By focusing on the study at a street level approach, this study aims to offer a comprehensive understanding of green infrastructure effectiveness at a high resolution, providing urban planners with data driven recommendations for optimising and implementing green infrastructure solutions to areas in need.
The findings of this research aim to contribute by bridging gaps in green infrastructure assessment by integrating geospatial data, network based spatial analysis, with machine learning, to assist in decision making to make urban areas more resilient.