Rooms with a ‘view score’: Software aids building designers

The marketing material shows natural light flooding through floor-to-ceiling windows. But in real life, neighbors can easily see inside the apartment; the view outside includes a parking lot; and the oversized windows mean costlier energy bills.

ViewScore.io, a new software tool created by researchers in the Environmental Systems Laband Housing Innovation Lab in the College of Architecture, Art and Planning (AAP), can simulate and score such scenarios, helping building designers optimize window views along with energy efficiency. The technology could also be used to inform green building standards, and to help consumers decide which home to rent or hotel room to reserve.

Leveraging surveys, interviews and machine learning, ViewScore.io offers the first systematic approach to predicting satisfaction with window views, the researchers say. Validated through New York City case studies, the software generates view scores based on 23 factors, including window size and glazing; perceived privacy; and the presence and proximity outside of ground, trees, sky, buildings, people and infrastructure.

"Architects and developers often want to make panoramic views because people like to see out, but from a sustainability perspective, windows are a weakness in buildings," said Jaeha Kim, a doctoral student in the field of systems engineering and lead developer of ViewScore.io. "With a score that captures how good window views are, or how valuable, we think we can help architectural designers improve building facades and floor layouts for human-centered design."

Kim is the lead author of several research papers published or presented recently at academic conferences that advance the analysis of how people interpret window views. Cornell co-authors include Timur Dogan, associate professor in the Departments of Architecture and Design Tech in AAP, and Katharina Kral, lecturer and housing research fellow in AAP. Their ongoing collaboration with experts from the University of California, Berkeley; New Jersey Institute of Technology; and Singapore University of Social Sciences includes regular meetings to set data collection standards and strategize about future work.

Several factors are driving researchers' interest in better understanding window views, Kim said. Americans spend more than 90% of their time indoors, making indoor environments important to physical and mental health. Sustainability concerns are pushing designers to make windows smaller, thicker or less transparent - but how those changes impact occupants is unknown. And poor views can yield up to 6% less in net rent, and often make properties more difficult to rent or sell, reducing their value and potentially wasting resources.

Research to date investigating window views has involved small samples and lacked common measurement techniques, making it difficult to compare or broaden results, Kim said. Addressing that shortcoming, the Cornell team collaborated with international experts to propose best practices for assessing window view satisfaction and to introduce an open-source database to promote collaboration and accuracy. Their findingswere published July 31 in the journal Leukos.

"There had been no holistic, systematic approach," Kim said. "To push this window view research forward, we realized we need to collaborate in a big group to scale up our empirical data collection, and that's why we think this paper is essential for the field."

The team has tested ViewScore.io's modeling in more than 30 New York City apartment buildings. "We wanted to test if our ViewScore.io protocol works for real apartments in a real urban context," Kim said. "If the simulation scores match how people assess their space in surveys, that's good, and that's what we found."

Evaluating 35 buildings with more than 10,420 total rooms, the Cornell team's analysis showed that "visual comfort" - including access to daylight, view satisfaction and privacy - increased with floor height. That shift likely corresponded to property values, raising social equity questions, the researchers said.

Another New York City case study of an apartment block in Queens similarly found privacy suffered on the ground floor and was affected by the heights of neighboring buildings. The findings suggest that uniform window patterns and floor layouts across all floors may not be the best design approach.

Those results were reported at, respectively, the International Building Performance Simulation Association conference Aug. 25-27 in Brisbane, Australia, and the 2025 Annual Simulation Conference, May 26-29 in Madrid, Spain.

Currently offered as a plug-in to simulation software used by building designers, ViewScore.io has received provisional patents and its team has completed the National Science Foundation's Innovation Corps (iCorps) program. Kim said possible future applications could incorporate view satisfaction scores into real estate listing sites such as Zillow or Booking.com, to help consumers assess property values or quality; and could help policymakers plan retrofits of older buildings to improve efficiency.

"We are not only developing our experimental protocol but pushing this into practice with more usable computational tools," Kim said. "It's my hope that this research will improve living conditions and help people."