"Cornell’s experience was, in fact, part of a national trend which saw a shift in research funding patterns away from the lone wolf investigator model. That historical shift underpinned a broader move toward collaborative interdisciplinary science, which you will recognize in my past and which is what I believe led me here." Read more about Lynden Archer, Cornell Engineering Dean
Powerful Cornell software informs NYC’s transportation planning
By Syl Kacapyr
A powerful software developed at Cornell Engineering is helping to inform transportation planning in New York City, resulting in cleaner air and healthier communities. Now the engineers behind the software are looking to bring their digital tool beyond the Big Apple.
The Cornell University Post-Processing Software, or CU-PPS, takes New York City’s travel demand data – for instance, how many cars pass over the George Washington Bridge during any given time of day – and merges it with a complex model that can estimate how much pollution is being produced from any given source.
The result is an inventory of air quality for the entire city, which is required biannually by the Environmental Protection Agency (EPA) as part of the 1990 Clean Air Act Amendments. The software solves a major challenge for transportation officials, because calculating transportation inventory data is a complex process, according to Oliver Gao, professor of civil and environmental engineering and lead developer of CU-PPS.
“Transportation officials are good at designing roads and bridges, but you can’t ask them, ‘if this car is on this road at 55 miles-per-hour under these specific weather conditions, how many grams of emissions are coming out?,” said Gao. “And what kind of fuel is the car using, what is the age of the car and how aggressive is the driver? The EPA can model that, but they don’t understand how transportation is distributed in the city. So neither side can do this on their own.”
But the CU-PPS software can, adds Gao. The data it produces tells city officials which areas of the city have cleaner transportation and which do not. For areas of the city that don’t meet EPA ambient air-quality standards, officials are legally obligated to develop an action plan to mitigate the air quality and protect public health.
“The software integrates the EPA’s state-of-the-art emission model and New York City’s best-of-practice, activity-based travel demand model for emissions inventory estimation at a finely resolved link-by-link scale,” says Gao. “The software has gone through rigorous testing and evaluation procedures by the inter-agency consulting groups and has been in official use in the New York City metro area.”
A NEW TOOL
When the Clean Air Act Amendments were passed in 1990, city officials were faced with the challenge of producing an emissions inventory of airborne pollutants such as sulphur dioxide, carbon dioxide, carbon monoxide and volatile organic compounds – a process known as conformity assessment. Larger cities hired entire staffs just to collect traffic data and conduct transportation modeling, but still had no way to estimate the emissions that traffic was producing. The EPA provided models, but melding the traffic data into the models is no easy task.
The standard solution to completing the rigorous work became one that would ultimately cost some cities millions of dollars each year: hiring consulting agencies.
“When you hire consultants, they have a tendency to do what the clients ask them to do, which means most of the time they’ll come out with a pleasing estimate that’s not always accurate,” explains Gao. “Or sometimes the EPA will not accept the report because the agency didn’t use the most up-to-date methodology.”
Among those searching for a better way to complete its required conformity assessment was the New York Metropolitan Transportation Council (NYMTC), a regional council of governments overseeing transportation planning in New York City, Long Island and the lower Hudson Valley. During a presentation to NYMTC officials in 2006, Gao showed how software could be used to complete the assessment and better serve the 12 million people living in the region.
Soon after, Gao won a request for proposals to build the software. He collaborated with Johannes Gehrke, a former professor of computer science at Cornell, and worked closely with the EPA and New York State Department of Environmental Conservation to ensure the software used the latest methodologies and standards. Gao and Gehrke also worked to develop an intuitive, user-friendly, web-based interface – something that would distinguish the software from others that had attempted to serve the same purpose.
The result was the first version of CU-PPS, which NYMTC still licenses today. Under the agreement, the Gao Research Group provides continuous maintenance and support for the software, and hosts it on a Cornell server.
“With our software, all they have to do is upload data from their models and they get tables they can use for their transportation conformity report,” says Gao. “It’s a tremendous reduction of workload. There are people there that love the software because it makes their life so easy.”
The software has also minimized staff training time, according to Sangeeta Bhowmick, manager of model application and development at NYMTC.
“All the required inputs files are clearly labeled which makes it easy for quality assurance and quality control, and reduces human error,” says Bhowmick. “It is also much faster than what we had in the past and the result view page provides professionally formatted output reports.”
REAL WORLD IMPACT
While CU-PPS makes transportation conformity cheaper and more manageable, Gao says it’s the impact on public health that is most important.
“This software is like a gatekeeper to help enforce and implement the Clean Air Act,” explains Gao. “It translates frontier research into tools that practitioners can use in their planning of transportation systems for the protection of our environment and public health.”
For instance, says Gao, the software can be used to produce impressive reports that help cities compete for millions of dollars in federal transportation funds. It can also be used to determine the best way to allocate those funds and help quantify the reduction in emissions for any proposed transportation policy.
“New York may need to decide between enacting congestion pricing, building a bridge or adopting new transportation system technologies. With this software, they can use it as a tool to quantify the environmental benefit each choice would bring to a community,” says Gao.
CU-PPS is a tool that can help engage and educate the general public as well. Gao was invited to a public hearing organized by a Long Island community that raised concerns about billboards placed on the Long Island Expressway. The billboards were designed to draw additional tourism to their municipality, which some worried would raise carbon dioxide levels because of the increased traffic. The CU-PPS software can be used to help community members quantify and weigh the environmental impact of the increased traffic versus the economic benefits.
In a project supported by Cornell’s Atkinson Center for a Sustainable Future, Gao’s collaborators used maps produced by the software to show Staten Island residents the ebb and flow of air pollution in their borough. The maps were part of a social experiment designed to study how people respond to learning about environmental hazards in their community. For most, it was their first time seeing data related to the air they breathe.
One of the members of the Gao Research Group who maintains CU-PPS is Mohammad Tayarani, a postdoctoral associate who has spent the past six years examining the link between transportation, public health and climate change.
“Working on CU-PPS provides me with the opportunity to work on a real-world problem for one of the most populated urban areas in the world. I can see any effort on this software will contribute to mitigating climate change,” said Tayarani.
BEYOND THE BIG APPLE
Seeing the value CU-PPS brought to New York City, Gao says he wants to bring his software to other cities faced with the same challenge of producing conformity assessment reports for the EPA.
“My research is having a real-world impact, and so now I want to expand this impact. If we can do it for New York, we can do it for Chicago, Houston, any city,” says Gao, noting that CU-PPS would have to be customized for each city.
Leo He, a Ph.D. student in the Gao Research Group who helps maintain the software, also sees CU-PPS’s value as a tool that can provide finely-resolved data to researchers and the general public.
“It can simulate travelers' and residential exposure to traffic-related emissions and corresponding health impacts,” says He. “Certain areas with high pollutant concentration draw more attention and need corresponding transportation policies, but the fine output also helps researchers to find interesting research topics and objects.”
Bhowmick agrees, noting “the ability to download the intermediate files is a huge advantage of the application and gives the curious researcher a better understanding of internal calculations.”
Gao encourages those interested in learning more about CU-PPS and other research projects to visit the website for the Center for Transportation, Environment, and Community Health – a multi-university research center directed by Gao.
As cities address their aging infrastructure and begin to incorporate new technologies into their transportation planning, Gao stresses that it is a pivotal time for policy makers to consider how the decisions they make affect the health of city residents.
“It’s important to have smart cities, but for me, it’s also important to have healthy cities,” says Gao.