Optimizing Urban Park Locations with Addressing Environmental Justice in Park Access and Utilization by Using Dynamic Demographic Features Derived from Mobile Phone Data

Urban parks play a critical role in improving the quality of life of residents and creating sustainable and resilient urban environments. To make the park services more equitable for the sake of environmental justice, it is crucial to evaluate the distribution and quality of existing urban parks and consider them in locating new urban parks, in terms of both park access and utilization. By adopting a new type of information on dynamic population based on mobile phone-based tracking, we addressed the research questions of (1) where new urban parks are located when considering park access of vulnerable population as well as park utilization of all city residents, (2) how important each of different dynamic demographic and park neighborhood features is in estimating potential park utilization, and (3) how the local context of various factors (i.e., land price, available land parcels, existing urban park locations) affects the new urban park locations. In this research, we introduced a novel model for optimizing the locations of urban parks. First, we utilized mobile phone-based floating population data to estimate how easily vulnerable older adults can access parks and how frequently existing parks are used by the entire population. Second, we extracted dynamic demographic features from this data and created features related to parks and neighborhoods. These features were used to predict the potential utilization of candidate park sites using a Random Forest regression model. Third, we developed an advanced location optimization model with multiple objectives, which considers the factors of park accessibility, utilization, and construction costs. The proposed methodology was validated with an application to Ulsan Metropolitan City in the Republic of Korea.