PULLMAN, Wash. - Cities are built around critical public facilities, some more desirable for neighbors than others.
People want to live near hospitals and parks. Living near a wastewater plant or a landfill is another matter.
A study from Washington State University, published in the journal Urban Sustainability, introduces a new method of analyzing the socioeconomic fairness of where such facilities are located. Researchers conducted an analysis of Spokane County, Washington, and the city of Spokane that could be replicated in other communities to guide planning decisions.
The analysis found that some "desirable" facilities in Spokane, such as hospitals and parks, are located both centrally and fairly, in terms of providing broad access for residents across socioeconomic differences. But all of the "undesirable" facilities studied, such as the wastewater plant and landfill, were located in neighborhoods where people have less education.
"Fairness, for us, means paying attention to the nonuniform distribution of socioeconomic conditions of people, and trying to avoid cases where a certain group or demographic is not disproportionately affected," said corresponding author Assefaw Gebremedhin, Berry Family Distinguished Associate Professor in WSU's School of Electrical Engineering and Computer Science and director of the Northwest Institute for Cybersecurity Education and Research (CySER). "If it is the same for everybody, then it's fair."
A host of factors influence where public facilities are located, and many of those decisions were made long ago. But as more attention is paid to socioeconomic disparities, and as communities continue to grow, the desire to fairly distribute the positive and negative effects of public facilities has also grown.
The study builds on the field of network science, which examines how communities and organizations are connected. A key element of network science is centrality - whether residents have roughly equal distances or time to travel to important services or the maximum distance traveled in critical situations is minimized.
The WSU project applied a new holistic measure of fairness, built on 13 variables from U.S. Census data, including income, education levels and other demographic characteristics. Other studies of the fairness of public facility locations have tended to rely on a single socioeconomic metric, such as income or race.
The researchers then used a cluster analysis to map distances between residents and facilities.
"The fairness metric introduced in our paper builds upon classical measures of centrality, which use distance or time as a measure to be minimized, and considers the heterogeneity of a particular neighborhood to ensure that facilities are fairly and optimally placed given the demographic and socioeconomic characteristics of the neighborhood," said Stephanie Kane, a co-author of the new publication and assistant vice provost in WSU's Office of Institutional Research and a doctoral student in WSU's Interdisciplinary Studies program.
James Crabb, a researcher and program coordinator in the CySER Institute, was the co-first author of the paper along with Kane.
"This project was an interesting data analysis and coding challenge," he said. "We had to combine Census data with travel distances from Google Maps and then use it within a Python library to construct and visualize the graph. While we were able to rely on pre-existing code libraries for most of the project, we did have to write custom Python code to combine everything and even had to code one of the centrality measurements from scratch."
In the analysis of Spokane County, researchers developed four demographic clusters: high income, high education; high income, low education; low income, low education; and low income, moderate education.
The undesirable facilities used in the study were four Superfund environmental cleanup sites and the city's wastewater treatment plant. The 22 desirable sites analyzed included fire stations, parks and a hospital.
Researchers found that both of the low-education clusters were the closest to undesirable facilities. In the more complicated analysis of desirable facilities, many were found to be ideally located in terms of both centrality and fairness.
Kane said the technique could be adapted in other communities and be modified with additional data to evaluate the siting of specialized facilities.
"City planners can use this technique repeatedly as city or regional demographics change over time, in order to ensure that new facilities are well placed and that undesirable facilities do not disproportionately impact residents," she said.
