Machine Learning & Predictive Analytics in Environmental Protection
Thursday, July 9, 2020
1:00 PM - 2:30 PM Eastern
Thank you for joining us for the second EE2020 webinar! A recording of the proceedings is now available at the link above. Interested in continuing the conversation? Contribute to a discussion thread on the E-Enterprise Community Inventory Platform and share your questions and ideas about ways that environmental agencies can put machine learning to work.
The July edition of the EE2020 webinar series focused on applications for Machine Learning and Predictive Analytics in environmental protection.
The webinar included:
Information about the basics of machine learning and predictive analytics;
Case studies in using predictive analytics to help agencies target resources for environmental inspections where they can do the most good; and
A discussion on machine learning, implementation challenges, use cases, and collaborative development opportunities for the environmental protection community.
Webinar Agenda
1:00 - 1:05
Welcome & Logistics
1:05 - 1:35
Using Machine Learning to Target Inspection Resources
Michael Greenstone, Professor in Economics at the University of Chicago
The University of Chicago Energy and Environment Lab (E&E Lab) has partnered with EPA’s Office of Enforcement and Compliance Assurance (OECA) to determine whether new data analytic techniques can help agencies better target their resources for facility inspections. This presentation will introduce some basic principles of predictive analytics and describe the team’s work to develop a model that uses machine learning to predict the likelihood that an inspector will find severe violations at a RCRA large quantity generator. A nationwide field test benchmarking the performance of this model relative to status quo targeting practices suggests that applying machine learning to inspection targeting has significant potential to strengthen targeting practices. This presentation accompanies the release of predicted likelihoods of non-compliance for large quantity generators nationwide by OECA and the E&E Lab to support state and regional FY2021 inspection planning.
1:35 - 2:05
Improving Environmental Compliance with Artificial Intelligence
Daniel E. Ho, Professor of Law at Stanford University
Elinor Benami, Postdoctoral Scholar at University of California, Davis
The Regulation, Evaluation, and Governance Lab (RegLab) at Stanford University has been partnering with U.S. EPA's Office of Enforcement and Compliance Assurance (OECA) to bring machine learning and artificial intelligence into environmental enforcement. We will showcase three examples of cutting-edge methods to use artificial intelligence to improve environmental compliance. First, we show how risk models can identify facilities likely to be in “significant noncompliance” for the National Compliance Initiative. Second, we will show how high-resolution satellite imagery can be used to identify relevant unpermitted concentrated animal feeding operations. Third, we will show how water quality sensor data can be used to investigate the likely cause of pollutant spikes. This presentation is oriented towards regulators and/or NGO participants interested in a high-level and accessible overview of how artificial intelligence can be used to improve environmental monitoring and compliance.
2:05 - 2:30
Interactive Discussion
This facilitated discussion will offer webinar attendees an opportunity to discuss machine learning applications with the panel of speakers. Bring your questions about machine learning and predictive analytics. Share your business challenges to learn how these technologies might support your agency’s mission. Talk about your own experiences using machine learning. Propose ideas for collaborations or partnerships that could help environmental protection agencies build new capabilities using machine learning and predictive analytics.
Featured Speakers
Michael Greenstone is the Milton Friedman Distinguished Service Professor in Economics, the College, and the Harris School, as well as the Director of the Becker Friedman Institute and the interdisciplinary Energy Policy Institute at the University of Chicago. He previously served as the Chief Economist for the White House Council of Economic Advisers, where he co-led the development of the United States Government’s social cost of carbon. Greenstone also directed The Hamilton Project, which studies policies to promote economic growth, and has since joined its Advisory Council. He is an elected member of the American Academy of Arts and Sciences, a fellow of the Econometric Society, and a former editor of the Journal of Political Economy. Before coming to the University of Chicago, Greenstone was the 3M Professor of Environmental Economics at MIT. Greenstone received a Ph.D. in economics from Princeton University and a BA in economics with High Honors from Swarthmore College.
Daniel E. Ho is the William Benjamin Scott and Luna M. Scott Professor of Law at Stanford Law School, Professor of Political Science, and Senior Fellow at the Stanford Institute for Economic Policy Research. He serves as Associate Director of the Stanford Institute for Human-Centered Artificial Intelligence and is Director of the Regulation, Evaluation, and Governance Lab (RegLab). He received his J.D. from Yale Law School and Ph.D. from Harvard University and clerked for Judge Stephen F. Williams on the U.S. Court of Appeals, District of Columbia Circuit. Ho served as president for the Society of Empirical Legal Studies (2011-12), co-editor of the Journal of Law, Economics, & Organization (2013-16).
Sarah Armstrong is a Research Fellow in the University of Chicago’s Energy and Environment Lab working with the Environmental Protection Agency in Washington, DC. Her research supports several social policy research collaborations between Urban Labs and the EPA that leverage econometrics and data science methodologies to improve environmental outcomes. Sarah previously served as the Syed Babar Ali Foundation Fellow in Lahore, Pakistan, and has completed internships with the National Economic Council, the U.S. Department of the Treasury and the Urban Institute. Sarah graduated from Georgetown University with a master's degree in mathematics and statistics, and from Texas A&M University with a bachelor's degree in economics and political science.
Elinor Benami is a Postdoctoral Scholar in the Agricultural and Resource Economics department at the University of California, Davis, following her Ph.D. from the Emmett Interdisciplinary Program in Environment and Resources (E-IPER) at Stanford. Her research employs both quantitative and qualitative methods to address how governance shapes social and environmental outcomes in agricultural and industrial production systems. Her work spans geographies, and she has significant experience evaluating systems that affect people and the planet across the global tropics with a focus on Brazil, Uganda, Kenya, and Indonesia as well as the United States.