Connecting the Dots: Using MIT's Campus as a Test Bed for Sustainability

• Les Norford, professor, Department of Architecture
• Jeremy Gregory, executive director of the MIT Climate and Sustainability Consortium
• Audun Botterud, principal research scientist for the Laboratory for Information and Decision Systems
• Steve Lanou, project manager in the MIT Office of Sustainability (MITOS)
• Fran Selvaggio, senior Building Management Systems engineer in the Department of Facilities 
• Daisy Green, Postdoc, Electrical Engineering and Computer Science
• You Lin, Postdoc, Laboratory for Information and Decision Systems 

MIT’s campus links more than 200 buildings that comprise 13.9 million square feet, with 8.3 million square feet dedicated to academic purposes—classrooms, labs, libraries, offices, and more—and 3.3 million in residences.

Keeping these buildings safe and operational is no small feat—and heating and cooling is one of the most critical functions that MIT’s Facilities teams take on each day. Consequently, it’s also one of the biggest areas of impact that Campus Climate Action needs to address. Why?

Today’s campus-wide building management systems simply aren’t built to respond efficiently to the unique needs of thousands of different spaces.

For example, if a classroom is only in use for a couple of hours each day, does it need to be held at a consistent, people-friendly temperature? Do certain labs need to be maintained at a specific temperature—yet within a building that doesn’t need to be heated or cooled to that extent in every space? Does a building mostly in use during standard working hours need to accommodate users who like to work late or come in early?  

Add in the unpredictability and relative extremes of New England’s seasons, the steadily increasing temperature levels we’re seeing as a planet, and the preferences of several thousand unique people, and you’ve got yourself quite a set of variables.

The result? Plenty of inefficiency and wasted resources—and a climate impact in dire need of mitigation. As Jeremy Gregory, executive director of the MIT Climate and Sustainability Consortium explains, “Our buildings are the biggest part of our carbon footprint. The inefficiencies there are a paramount problem to solve—and if we’re going to use the campus as a ‘test bed’ for change, we do well to start tackling our biggest hurdles now.”


The research-driven response

Chances are you’ve heard a lot about artificial intelligence (AI)… for better or for worse. While AI’s capacity to write a Hollywood screenplay or create a work of art for MOMA is up for debate, AI can have a powerful impact on our capacity to balance energy needs in complex environments.

If you have a “smart” thermostat in your home, you’ve experienced how sensors in different parts of your space gauge heating and cooling according to the weather outside, the rooms you’re occupying (or not), what you’re doing in those rooms at what times of day, and your personal preferences.

The smart part: your thermostat will either self-adjust to reflect those preferences or will respond according to the way you’ve programmed it to act… ideally in a seamless manner, while you’re comfortably living your life.

Joe Higgins,  Vice President for Campus Services and Stewardship, originally pitched the idea of using AI to support campus energy efficiency to students at the 2019 MIT Energy Hack. A fresh challenge was issued to MIT researchers and Facilities teams to collaborate on AI implementation within MIT’s plant—and the AI Pilot Project was born.

As Les Norford, a professor of architecture at MIT, explains, “What works in your house is possible because of the scale—a few sensors across a few rooms have many data points, but nothing like a whole campus. MIT has more buildings, more variables, and more people who demand a certain level of comfort in their environment. But sooner than later, you must tend to your own business!”

Audun Botterud, principal research scientist for the Laboratory for Information and Decision Systems, conducts research that addresses the need for a decarbonized energy grid, from energy market interactions to designing batteries that store energy more efficiently, and at a greater volume. 

Botterud’s focus makes him the perfect partner to work with a machine learning algorithm that would use data from a set of classroom spaces—for now, all in Building 66. The different needs of each classroom presented an appropriately complex environment to gauge how heating and cooling could be optimized in the face of external weather influences, occupancy needs, and the presence of different heating zones, often a wall or two away.

“This was my first time being involved in a project like this at MIT, and it’s great to be working on something very much at the applied end of the research spectrum,” said Botterud. “This is exactly how we should be using the campus as a ‘test bed’—to pioneer new solutions that we can build out across campus, and then share with the world.”

For Fran Selvaggio, senior building management systems engineer in the Department of Facilities, the first challenge was to ensure the Building Management System was ready before the AI variable was added into the mix. “Our systems wouldn’t functionally interact effectively with AI unless everything was in excellent shape—which is why the first step in innovation often begins with making sure your current systems are optimized.”

As MIT continues to learn from the AI pilot in Building 66, the plan is to scale out further and introduce more buildings into the pilot, and give the AI algorithm more data to work from to optimize MIT’s efficiency across its physical plant.

To learn more about MIT’s Fast Forward commitments, start here. To learn about MIT’s Campus Climate Action-specific efforts, head here.
 

• Brian Goldberg, Assistant Director, Office of Sustainability
• Marty O’Brien, Assistant Director, Campus Services
• Desiree Plata, Associate Professor of Civil and Environmental Engineering, Department of Civil and Environmental Engineering

With over 11,000 undergraduate and graduate students studying and researching at MIT—many of whom live in residences here—and over 17,000 faculty and staff in full-time and hybrid positions, it’s safe to say there are a lot of people who might enjoy a meal or three on the MIT campus on any given day.

Whether they’re grabbing favorites from one of our many dining spots or bringing in their own food from somewhere else, all MIT campus eaters have one thing in common: they’re putting their food waste into our system.

In fact, up to 40 percent of MIT’s trash each year is made up of food waste that ends up in landfills—and ultimately, that waste accounts for 30-40 percent of our production of greenhouse gases each year. 

Clearly, this is a problem in need of a distinctly MIT-caliber solution—but that solution would ultimately require a change on multiple levels: both in how we as a community immediately deal with the waste we produce, and how we approach the processing of waste as an institution.

As Brian Goldberg, assistant director r of the Office of Sustainability, says, “We want to find ways for our campus community to contribute to our climate action goals at every level. That means the work we do to mitigate the waste we produce as an institution, through to our impact as departments, right down to our behaviors and choices as individuals.”   

The research-driven response

MIT has committed to the goal of reducing campus trash by 30 percent by 2030. That’s no small number, but by applying all the skills MIT is known for to the challenge—rigorous research, data-driven innovation, and pioneering design thinking—the Institute is working to tackle our waste problems through a series of coordinated actions.

Goldberg calls MIT’s response a cross-campus effort to “design out waste.”

A series of waste audits were conducted in partnership with the Department of Facilities and departments, labs, and centers across campus to take a look at our waste disposal practices and the impact of waste stream contamination on our recycling efforts. An eye-opening project with the MIT Media Lab even took a creative approach to gathering actionable ways to reduce waste.

Goldberg and Marty O’Brien, assistant director for Campus Services, outlined strategies to reduce overall campus waste. Their response to food waste in particular was inspired in part by the research of Associate Professor of Civil and Environmental Engineering Desiree Plata.

Plata’s research focuses on ways to trap methane gas—including methane gas produced by cows at dairy farms and yes, by the food waste the MIT—and transform it into energy that can be used to power the regional grid. Desiree’s team at the Plata Lab at MIT is continuing to work toward a world in which the “engineered solutions of the future will incorporate environmental objectives.”

Bins have now been placed across the MIT campus to separate out food waste, which can then be reprocessed into fertilizer, compost, and energy—without the off-product of greenhouse gases caused by transporting the waste and putting it into landfills.

The entire MIT community is welcomed to monitor the Institute’s progress towards their waste impact goals in the , Sustainability DataPool which includes the Material Matters, and Campus Water Use dashboards, which you can find here.

To learn more about MIT’s Fast Forward commitments, start here. To learn about MIT’s Campus Climate Action-specific efforts, head here.  

• Adam Schlosser, Joint Program on Science and Policy
• Kerry Emanuel, Professor of Atmospheric Science,  Dept of Earth, Atmospheric, & Planetary Sciences
• Miho Mazereeuw, Associate Professor, Department of Architecture
• Janelle Knox-Hayes, Professor, Department of Urban Studies and Planning
• Franz-Josef Ulm, Faculty Director of MIT Concrete Sustainability Hub and Professor, Department of Civil and Environmental Engineering
• Bill Colehower, Senior Advisor to the Vice President for Campus Services and Stewardship
• Laura Tenny, Senior Campus Planner, Office of Campus Planning
• Brian Goldberg, Assistant Director, Office of Sustainability
  
   

Heat and flood risks are two of the most critical threats climate change poses to our planet. More specifically, there has been a global year-over-year increase in: 

• flooding from more frequent and high-volume rains;
• flooding from storm surges and rising sea levels; and
• extreme heat events.

The MIT campus and the broader region are increasingly subject to these threats—and that’s where the “Resiliency & Adaptation Roadmap” comes in. This is MIT’s plan for mitigating climate-driven risks, becoming more resilient in the face of climate change, and reducing MIT’s impact as a campus on the community around us, and on our planet. The forthcoming roadmap builds upon nearly a decade of past work.

Creation of the roadmap is led by a team including the Office of Sustainability, Campus Planning, and Campus Services and Stewardship with support from faculty, and engineering and facility staff who know every inch of the campus and how it functions; risk, insurance, and climate science experts who can help us assess what we’re facing; emergency management professionals who can help us with our response; and dedicated students who are individually and collectively driving efforts to foster a more climate-resilient campus.

Resiliency work on campus ensures that every aspect of how we plan, operate, build, and react to challenges makes the least impact on our climate. Like everything else at MIT, meeting the challenge has to begin with data. But how do we gather the data we need?

While catastrophic weather events have been happening for as long as we have records of our weather, the kinds of records kept even a couple of hundred years ago don’t give us enough concrete data to predict what’s coming next with the degree of accuracy we need to make smart decisions for our safety and wellbeing.

Beyond that, there’s a degree of difficulty in gauging local risks—like a once-in-every-100-years catastrophic event—with the most readily available global climate data.

The research-driven response

Fortunately, MIT’s own Miho Mazeereuw is on the case—and not just for MIT, but on behalf of communities worldwide. Mazereeuw heads up the MIT Urban Risk Lab, an interdisciplinary lab that uses design and technology to develop risk-reduction concepts to meet pressing climate challenges. Her work is directly tackling the lack of data needed to address heat risks in particular.

The Urban Risk lab is helping MIT implement “downscaling”, which is the term given to the process of narrowing down data inputs to understand temperature patterns within a smaller area—a process that will help the MIT team quantify and address threats to MIT, and to the communities around MIT.

To this end, the Urban Risk Lab has supported the placement of sensors across the MIT campus and on MIT vehicles, as well as other key placements in the region, to begin mapping patterns in local heat levels—input that will give us a more concrete sense of how our immediate environment is developing.

With more accurate and localized data, MIT can not only plan for more efficient cooling measures that have less impact on the energy grid, we can also develop responses to heat threats that will better serve our community, and best serve the people living and working around us.  

This is just one part of the work that’s going into the Resiliency & Adaptation Roadmap, which you can learn more about here. To learn more about MIT’s Fast Forward commitments, start here. To learn about MIT’s Campus Climate Action-specific efforts, head here.