ORC IAP Seminar 2018

1/29/18 | 9:30am-4:30pm | 34-101
Operations Research for Social Good

Description: The applied nature of operations research makes it an important force for good in alleviating poverty; promoting accessible housing; improving health outcomes; and other endeavors that benefit large groups outside academia and industry. In this seminar, we will engage with a wide range of researchers and practitioners tacking these and other topics via data science, optimization, and other contemporary OR methods. 

Date: Monday, January 29th, 2018
Time: 9:30am-4:30pm 

Video: https://www.youtube.com/watch?v=ob6KhbWv3Kc&list=PL6nqfd-VvqxF5sZC6XHORh...





Michael Johnson

Professor and Chair, Department of Public Policy and Public Affairs, UMass Boston 

Community-engaged operations research: Localized interventions, appropriate methods, social impact
Community-engaged operations research is an extension of multiple OR/MS traditions to support participatory research, localized impact and social change. It applies critical thinking, evidence-based policy analysis, community participation and decision modeling to local interventions. It emphasizes the needs, voices and values of disadvantaged and marginalized populations. It rests on a foundation of meaningful engagement with communities. Through a survey of current scholarship in two complementary areas of inquiry, ‘community operational research’ (referring to work by primarily European researchers) and ‘community-based operations research’ (referring to work by primarily American researchers), I develop principles for community-engaged OR, present critical questions that represent opportunities to expand the impact of this work, and discuss current projects whose methods and applications have the potential to enrich research and practice in operations research, management science and analytics.

Dr. Michael P. Johnson is Professor in the Department of Public Policy and Public Affairs at University of Massachusetts Boston and serves as department chair.  He is also an instructor in the masters in Urban Planning and Community Development program at University of Massachusetts Boston. Dr. Johnson received his Ph.D in operations research from Northwestern University in 1997 and his B.S. from Morehouse College in 1987.

Dr. Johnson’s research interests lie in data analytics and management science for housing, community development and nonprofit service delivery. His methods enable non-profit and public organizations, especially those serving disadvantaged and vulnerable populations, to develop programs and policies that jointly optimize economic efficiency, beneficial population outcomes and social equity.  Current research projects include: planning models and case studies in foreclosed housing redevelopment and community revitalization; strategy design for vacant property management in distressed neighborhoods, shrinking cities and declining regions; community-based data collection, analysis and sharing for local development, and community-based operations research.  

His work has appeared in a variety of journals, edited volumes and conference proceedings. His most recent book is Decision Science for Housing and Community Development: Localized and Evidence‐Based Responses to Distressed Housing and Blighted Communities (John Wiley & Sons, 2016). He was editor of Community-Based Operations Research: Decision Modeling for Local Impact and Diverse Populations (2012, Springer). He is currently co-editing a volume of the European Journal of Operational Research on the topic “Community Operational Research: Innovations, Internationalization and Agenda-Setting Applications”, and is lead author of a book, “Supporting Shrinkage: Better Planning and Decision-Making for Legacy Cities”, under contract to SUNY Press.

More information on Prof. Johnson’s research is available at https://works.bepress.com/michael_johnson/



Arthur Delarue and Sebastien Martin

PhD Students, MIT, Operations Research Center

8 months on a school bus

Many problems faced by US school districts present interesting challenges for operations research practitioners. In this talk, we focus on the two major problems of school transportation and bell time choice, as part of a collaboration with Boston Public Schools (BPS), the oldest and one of the largest school districts in the nation (with 126 public schools and over 56,000 students). For the problem of school transportation, which involves delivering students to school every morning and back home every afternoon using a fleet of specialized vehicles (around 700 hundred school buses for Boston), we identify a natural multistage decomposition of the problem and propose integer programming formulations and efficient heuristics for each stage to compute school bus routes at the scale of the district. For the problem of bell time choice, we develop an integer optimization model with multiple objectives that provides both the versatility and simplicity that is required to facilitate public policy decision-making . Above all, the talk highlights interesting aspects of a collaboration with a public organization and describes the process of implementing OR solutions at the scale of a major city.


Arthur Delarue is a PhD student at the MIT Operations Research Center, working on mixed-integer optimization and its applications to real-world transportation problems under the supervision of Prof. Bertsimas.

Sebastien Martin is a PhD student at the MIT Operations Research Center, working on tractable large scale optimization with a focus on routing optimization and machine learning, under the supervision of Profs. Dimitris Bertsimas and Patrick Jaillet.


Andrew Therriault

Chief Data Officer, City of Boston 

Saving the world with data - The case for civic data science
The emergence of data science over the past decade has had an effect on almost every field, but its importance to the non-profit, advocacy, and government sectors has been especially profound. With a wide variety of data and limited resources which beg for optimization, these types of civic organizations can benefit tremendously from data-driven tools and analysis. What's more, the real-world challenges these organizations take on create opportunities for data scientists, analysts, and engineers to do incredibly exciting work and have a much greater impact than you can get anywhere else. This presentation will introduce the audience to the landscape of civic data and technology, present real case studies of data science applications, and discuss how anyone interested in getting more involved can find their place in using data for good.
Andrew Therriault joined the City of Boston as its first Chief Data Officer in 2016, after serving as Director of Data Science for the Democratic National Committee. He received his PhD in political science from NYU in 2011 and completed a postdoctoral research fellowship at Vanderbilt, and more recently served as editor of "Data and Democracy: How Political Data Science is Shaping the 2016 Elections" (O'Reilly Media). Therriault leads Boston’s Analytics Team, a group that is a nationally-recognized leader in using data science to improve city operations and make progress in critical areas such as public safety, transportation, citizen engagement, and public health.


LUNCH BREAK (not provided)

1:45pm-2:30pm *Talk canceled*

Edoardo Airoldi

Associate Professor of Statistics, Harvard 

Near-optimal design of social network experiments
Classical approaches to causal inference largely rely on the assumption of “lack of interference”, according to which the outcome of an individual does not depend on the treatment assigned to others, as well as on many other simplifying assumptions, including the absence of strategic behavior. In many applications, however, such as evaluating the effectiveness of healthcare interventions that leverage social structure, or assessing the impact of product innovations and ad campaigns on social media platforms, or experimentation at scale in large IT companies, assuming lack of interference and other simplifying assumptions is untenable. Moreover, the effect of interference itself is often an inferential target of interest, rather than a nuisance. In this talk, we will formalize technical issues that arise in estimating causal effects when interference can be attributed to a network among the units of analysis, within the potential outcomes framework. We will introduce and discuss several strategies for experimental design in this context centered around a judicious use statistical models, which we refer to as “model-assisted” design of experiments. In particular, we wish for certain finite-sample properties of the estimator to hold even if the model catastrophically fails, while we would like to gain efficiency if certain aspects of the model are correct. We will then contrast design-based, model-based and model-assisted approaches to experimental design from a decision theoretic perspective.

Edoardo M. Airoldi (airoldi@alumni.harvard.edu) is an Associate Professor of Statistics at Harvard University, where he has been directing the Harvard Laboratory for Applied Statistical Methodology & Data Science since 2009. He holds a Ph.D. in Computer Science and an M.Sc. in Statistics from Carnegie Mellon University, and a B.Sc. in Mathematical Statistics and Economics from Bocconi University. His current research focuses on statistical theory and methods for designing and analyzing experiments on large networks, and on modeling and inferential issues that arise in analyses that leverage network data. His work has appeared in journals across statistics, computer science and engineering, including Annals of Statistics, Journal of the American Statistical Association, Journal of Machine Learning Research, Proceedings of the National Academy of Sciences, and Nature. He is the recipient of several research and young investigator awards including an Alfred Sloan Research Fellowship and a Shutzer Fellowship from the Radcliffe Institute of Advanced Studies. He delivered an IMS Medallion Lecture at JSM 2017 in Baltimore.



Hamsa Bastani

Goldstine Postdoctoral Fellow, IBM Research

Mechanism design for social good: Analysis of medicare pay-for-performance contracts
Medicare has sought to improve patient care through pay-for-performance (P4P) programs that better align hospitals’ financial incentives with quality of service. However, the design of these policies is subject to a variety of practical and institutional constraints, such as the use of “small” performance-based incentives. We develop a framework based on a stylized principal-agent model to characterize the optimal P4P mechanism within any set of feasible mechanisms in the regime of small incentives. Importantly, our feasible set can be flexibly modified to include institutional constraints. We apply our results to examine debated design choices in existing Medicare P4P programs, and offer several insights and policy recommendations. In particular, we find that these mechanisms may benefit by incorporating bonuses for top-performers, and using a single performance cutoff to uniformly assess performance-based payments. We also examine a number of comparative statics that shed light on when P4P mechanisms are effective.
Hamsa Bastani is a Herman Goldstine postdoctoral fellow at IBM Research. She will be starting as an Assistant Professor in the Operations, Information and Decisions group at the Wharton School in the Fall of 2018. Her research seeks to leverage data-driven techniques to improve quality of care and reduce costs in healthcare. She has developed efficient machine learning algorithms for personalized medicine, and designed pay-for-performance contracts that better align provider incentives with patient needs. She focuses on novel methods that cater to the many practical challenges in healthcare systems, including the high-dimensionality of electronic medical records, the ethical concerns of randomized decision-making on patients, and the unobservable biases that pervade healthcare claims data. Dr. Bastani completed her undergraduate and master’s degrees at Harvard University, and her PhD at Stanford university.


Marta C. Gonzalez

UC Berkeley, Associate Professor of City and Regional Planning

Modeling and planning urban systems with novel data sources
I present a review on research related to the application of big data and information technologies to urban systems. Data sources of interest include but are not limited to: Probe/GPS data, Credit Card Transactions, Traffic and Mobile phone data. Key uses of interest are modeling, adoption of new technologies and traffic performance measurements. In a second part a present multi-city study, we unravel traffic conditions under various conditions of demand and translate it to the travel time of the individual drivers. First, we start with the current conditions, showing that there is a characteristic time that takes to a representative group of commuters to arrive to their destinations once their maximum density has reached. While this time differs from city to city, it can be explained by the ratio of the vehicle miles traveled to their available street capacity. Moreover, we systematically characterize the macroscopic dynamic of the system by increasing volume of cars in the network, keeping the road capacity and the empirical spatial dynamics from origins to destinations unchanged. We identify three states of urban traffic, separated by two distinctive transitions. The first describing the appearance of the first bottle necks, and the second the transition to a complete collapse of the system. The transition to the second state measures the resilience of the various cities and is characterized by a non-equilibrium phase transition.

Marta C. Gonzalez is Associate Professor of City and Regional Planning at the University of California, Berkeley, and a Physics Research faculty in the Energy Technology Area (ETA) at the Lawrence Berkeley National Laboratory (Berkeley Lab). With the support of several companies, cities and foundations, her research team develops computer models to analyze digital traces of information mediated by devices. They process this information  to manage the demand in urban infrastructures in relation to energy and mobility. Her recent research uses billions of mobile phone records to understand the appearance of traffic jams and the integration of electric vehicles into the grid, smart meter data records to compare the policy of solar energy adoption and card transactions. Credit to identify habits in spending behavior. Prior to joining Berkeley, Marta worked as an Associate Professor of Civil and Environmental Engineering at MIT, a member of the Operations Research Center and the Center for Advanced Urbanism. She is a member of the scientific council of technology companies such as Gran Data, PTV and the Pecan Street Project consortium.


A PDF of the schedule can be found here

Event Time: 

2018 - 09:00