The Master of Urban Spatial Analytics (MUSA) program at PennDesign teaches students at the intersection of data science and public policy. As part of the program, graduate students from the Department of City and Regional Planning and Urban Spatial Analysis participated in the first annual MUSA/Smart Cities Practicum in the Spring of 2018. Led by faculty member Ken Steif and co-instructors Karl Dailey and Michael Fichman, students developed 5 public-sector machine learning algorithms in four cities: Philadelphia, Louisville, Minneapolis, and Providence. Their algorithms span multiple domains, including housing, transportation, and public health. The goal of the Practicum was for other communities to take this code and, according to Steif, “do something awesome with it.”
The project’s leaders and students have received positive feedback from many of the client cities, including a reflection from Providence. In response to their practicum model predicting the spatial risk of opiate overdose, Providence Fire Chief Zachariah Kenyon wrote: "The Providence Fire Department (PFD) will use the interactive map to conduct targeted outreach for Safe Stations, a program aimed at connecting people struggling with addiction to recovery services. The PFD data team will also use the code provided by MUSA students to refine the predictive models over time and to ensure project sustainability. Finally, PFD will use this project to demonstrate the need for funding to establish a data analytics team. The project has caught the attention of the state Department of Health, and the PFD, in conjunction with the city’s Healthy Communities Office, is excited to share the information and contribute to the body of knowledge about the epidemic in Rhode Island."
The practicum has received press coverage from Govtech, Inside Louisville, and Data Smart Cities. For the next Practicum, Steif hopes to involve other data-driven students from across the University.
All of the case studies and source code are available on the project website.
View Slideshow
