Understanding Public Space Use in Market Square
The Pittsburgh Downtown Partnership (PDP) is collaborating with Metro21: Smart Cities Institute, and Carnegie Mellon’s Computational Design Laboratory (Code Lab) to better understand how Pittsburghers use Market Square, in order to inform future planning decisions.
The project, led by Code Lab researcher Javier Argota Sánchez-Vaquerizo, is based on an urban analysis toolkit that uses computer vision and machine learning techniques to collect and analyze anonymous data about the movement of people and vehicles in public spaces, with unprecedented accuracy.
Argota, a Spanish architect, recently completed his MS at CMU’s School of Architecture with the support of the Fullbright Foundation, and developed key aspects of the toolkit as part of his master’s thesis, advised by Code Lab co-director Professor Daniel Cardoso Llach. The toolkit is co-developed by Urban Data Eye, a company that Argota formed to develop the technology further.
At the end of the of the project in November 2018, the PDP expects that the insights produced by this experimental project will help maintain Market Square as a vibrant hub of downtown activity, and perhaps make visible new planning opportunities. By better understanding how people use Market Square, for example, we may be able to better inform programming decisions —which the partnership tackles 200 days a year.
Project Update (December 2018)
During the project implementation phase of the project, four cameras were placed around Market Square covering a 50,000 sq foot area. These cameras tracked 12 categories ranging from pedestrians to tents. The cameras were deployed for 5 weeks, collecting over 3000 hours of video and detecting 250 million items.
- Weather Affection: Pedestrian traffic levels are lowered during periods of rain, but days after a rain episode the square sees an increase in pedestrian traffic by 20-30%.
- Entrances: Forbes Street Axis saw 2-3 times more activity than any other entrance. Conversely, the Market Street entrance saw the least amount of activity.
- Events: Consecutive events at the square bring more people to the area. Additionally, people stay three times longer during evening hours.
- Spatial Allocation: The project found that where furniture and trees are placed has a direct impact on where pedestrians congregate throughout the square.
Pittsburgh Downtown Partnership
Computational Design Laboratory, CMU
Urban Data Eye
Javier Argota Sánchez-Vaquerizo
Project Lead, Visiting Researcher, Code Lab, CMU School of Architecture
Daniel Cardoso Llach, Ph.D.
Advisor, Co-Director, Code Lab, CMU School of Architecture
Molly W. Steenson, Ph.D.
Advisor, Senior Associate Dean for Research, College of Fine Arts, CMU
Pittsburgh Supercomputer Center