1. Congratulations on the publication of the book, Decision Science for Housing and Community Development: Localized and Evidence-Based Responses to Distressed Housing and Blighted Communities which is a multidisciplinary approach to problem-solving in community-based organizations using decision models and operations research applications. How did the writing process begin?

The writing process for this book started around 2012. During one of the monthly team meetings that my co-authors and I were having for the National Science Foundation-funded project which forms the basis for this book, Jeff Keisler suggested that the many threads we were developing – namely, qualitative analysis of values and objectives, quantitative analysis of property acquisition criteria and decision models for property bid design – all folded into a case study-like approach to our topic. Not only that but all also seemed to lend themselves to a book. At the time, his suggestion seemed a bit fanciful – we were focussing on gathering data, engaging clients and presenting our research results at conferences and in journal manuscripts. But the notion stuck with all of us, and I increasingly saw our research in terms of potential chapters for a book. Also, I realized how fortunate the team and I were to receive this federal funding, and how important we felt our community-engaged decision modelling could be for local non-profits. So that made me determined that we would produce a research product that might appeal to more than just the usual academic journal audiences.

2. What were your main objectives during the writing process?

The main objectives I had during the writing process was to tell a correct, compelling story about a new kind of response to neighbourhood blight and housing market distress, one that I hoped (dreamed) would engender the sort of response I had to strong stories, well-told, on topics related to housing and community developed that I had used in my own courses, such as Moving to Opportunity: The Story of an American Experiment to Fight Ghetto Poverty by Briggs, Popkin and Goering (Oxford University Press, 2014), Investing in What Works for America’s Communities: Essays on People, Place and Action by Andrews and Ericson (Eds.) (Federal Reserve Bank of San Francisco, 2012) and A Primer for Policy Analysis by Stokey and Zeckhauser (W.W. Norton, 1978).

The first is a book-length treatment of a landmark government-directed study on housing mobility, the second is an edited volume on cutting-edge methods for high impact community-based policy and programming, and the third is an introductory textbook that I’ve found very useful in a number of my courses. If I could write a book that the thematic unity and policy relevance of the first book, the orientation towards action in a policy and planning context, oriented towards practitioners from the second, and the utility in terms of analytic methods of the third book, then I would feel that I was successful.

3. The book addresses critical problems in urban housing and community development through a diverse set of decision models and applications. Please could you give us an example of a problem that is tackled within the book?

The core problem that the book addresses is that of neighbourhood-level housing acquisition and redevelopment in the wake of widespread residential foreclosures in many neighbourhoods (in the U.S., the focus of the book, but certainly across many developed countries from 2008 – 2012). The Federal government knew of the high level of foreclosures, especially in newly-built middle-class or working-class and poor communities, but the scope of the problem – acquiring, rehabilitating and re-marketing these abandoned homes – was beyond the scope of its resources. The responses had to be local, and in the U.S. that means directed by community development corporations. So, these CDCs were faced with a problem: which houses, known to be at various stages of foreclosure, should they attempt to purchase, renovate and re-sell, or rent themselves, knowing that the purchase-acquisition-rehab-resale cycle is time-consuming, technically challenging and quite financially resource-intensive?

In addition, CDCs had to act quickly and nimbly, knowing that they could be easily outbid by private developers who could pay in cash for housing that they would likely ‘flip’ for quick profits rather than neighbourhood improvement. This is a difficult problem in community development practice, and my co-authors and I realized that it spanned the scope of the traditional operations research/management science problem solution cycle (identify problem, model, solve, implement), involved qualitative and quantitative methods, and required close engagement with community partners. We thought it could be a model for many other problems in housing and community development that we felt had not received the attention they should have from the OR/MS/analytics discipline.

4. Throughout the book, you emphasize analytic approaches that increase the ability of local practitioners to act quickly, thoughtfully, and effectively. By doing so, practitioners can design and implement responses that will in turn help build sustainable communities. Please could you tell us more about how data analytics plays a part in this book?

Data analytics is central to the solution strategies we describe in this book. First, data analytics helped us identify community partners that we thought could represent the diversity of local practitioners across the Boston metropolitan area, in terms of organization size and technical and financial capacity, as well as neighbourhood characteristics. Second, data analytics allowed us to decide what sorts of interventions or responses might be most suitable, on the basis of the intensity of housing market distress in the areas served by the various CDCs. Third, data analytics enabled us to quantify notions of impacts associated with foreclosed housing acquisition and redevelopment, such as ‘strategic value’ or ‘property value’ that had never been developed before, at least in the context of decision modelling-based housing response. Finally, and most importantly from the perspective of OR/MS/analytics as conventionally taught, data analytics enabled us to formulate and solve decision models that would give CDCs specific, detailed guidance on a range of strategies for actually targeting, bidding on and redeveloping particular housing candidates to balance multiple goals related to sustainable and equitable community development.

5. If there is one piece of information or advice that you would want your readers to take away and remember after reading your book, what would that be?

I would like my readers to realize that important problems in the public sector, that may have traditionally been seen as the province of those trained in areas such as public policy or community development, can benefit greatly from the contributions of those whose training has been in areas traditionally seen as not central to the community development mission: operations research, management science, data analytics and information technology. These methods, we have demonstrated, have the potential to generate specific responses to neighbourhood blight and housing distress that represent improvements over what community development practitioners can do on their own, and are presented in a way that community development practitioners can adapt into their daily practice in a straightforward way.

Can we demonstrate that an OR/MS-based intervention actually resulted in the beneficial impacts we claim? This gets into areas of social science-based research design that are not typically taught to engineers, yet I think should be. I dream of social interventions based on data analytics and decision science that use the same sorts of rigorous experimental designs as those used for social programs like job training or law enforcement that allow us to understand how, why and to what extent data analytics and decision science can generate substantial and beneficial changes in individual and community outcomes.

6. Who should read the book and why?

This book is written for three audiences: researchers, students and practitioners, in three core areas: operations research/management science/analytics, public policy and public affairs, and urban and community planning. Our analytic methods represent adaptations of existing tools and are interesting to researchers because they have not been deployed in this combination for this particular problem. Our problem discussion and policy responses are of interest to practitioners because we make substantial efforts to speak their language and demonstrate a deep understanding of the problem context. Our work is of interest to students because they represent a way to take certain fundamental notions of data analytics, location models, stochastic models and math optimization that they have learned in classes and provide interesting, real-world applications that are more complex than what they would typically see in a textbook, yet organized and at a level of detail that would normally be unavailable to them except after a year or more in collaboration with community partners.

Since four of the six co-authors are trained in operations research/management science, we understood that the book would be of greatest interest to those who identify as OR/MS or analytics professionals. However, the problem of housing and community development is one that has not received the same level of interest in our field as, say, production and logistics, and we realized that we should make the book relevant to those trained in public policy, public affairs and planning, since those are the domains in which the substantive problem is situated. More than that, three of the co-authors teach and do research in policy and planning-related fields, and we wanted our work to be part of the academic and practitioner conversation in those areas as well.

7. Why is this book of particular interest now?

Though the foreclosure problem has receded in intensity in the past few years, the larger issue of how to design evidence-based responses to challenging problems in housing and community development, such as gentrification, vacancy and abandonment, or sustainable development, remains an area of intense inquiry and action in the field. There is still a great need for decision scientists to engage in problems at the community level with the potential for substantial individual and neighbourhood impact, with an eye on social justice and community change.

8. Were there areas of the book that you found more challenging to write, and if so, why?

I found chapter 4, “Analytic approaches to foreclosure decision modelling”, the most difficult. It was intended to bridge the mostly descriptive portion of the book, describing the scale and scope of the foreclosed housing problem, and government and community reactions to the problem, with a set of diverse decision modelling responses for select community partners. I had to make the case for data analytics and decision modelling as potentially important to this problem, but not over-promise what these fields could do for practitioners. I wanted the chapter to speak to those trained in OR/MS as well as those in policy and planning, and I wanted all the readers to understand that analytics can help us understand what problem to solve, where to solve it, and how to solve it, in a way that is not a simple-minded list of steps, nor a high-level argument about ‘what works’, but a thorough, empirical development that makes the case for a new kind of response to neighbourhood and housing challenges. I feel I met that challenge, and provided a bridge to the more decision modelling-oriented material in chapters 5 – 10.

9. What is it about the area of causality that fascinates you?

Operations research, my core training, is about using analytic methods to solve practical problems in operations and management. But what does a good solution mean? Can we demonstrate that an OR/MS-based intervention actually resulted in the beneficial impacts we claim? This gets into areas of social science-based research design that are not typically taught to engineers, yet I think should be. I dream of social interventions based on data analytics and decision science that use the same sorts of rigorous experimental designs as those used for social programs like job training or law enforcement that allow us to understand how, why and to what extent data analytics and decision science can generate substantial and beneficial changes in individual and community outcomes.

10. What will be your next book-length undertaking?

I’m working on a manuscript related to shrinking cities and vacancy management. In many places in the U.S. and other parts of the developed world, deindustrialization and demographic changes (and maybe climate change) too has resulted in formerly highly-populated and economically vibrant cities shrink in size and influence. How can planners design empirical, evidence-based strategies that support creative responses to shrinkage and abandonment at the neighbourhood level? The answer, I believe, can leverage the substantial promise of ‘big data’, ‘smart cities’ and data analytics and decision science, embedded in a commitment to community participation and equity, that has the potential to provide great value to residents of distressed and declining communities who often do not have a voice in what happens to the places in which they live.

11. You are currently an Associate Professor, Department of Public Policy and Public Affairs, McCormack Graduate School. Please could you tell us more about your educational background and what was led you to focus your career on operations research?

I am African-American, and I attended a historically-black college, Morehouse College, along with a top-ranked technical university, Georgia Tech, to receive my bachelor’s and master’s degrees in 1987 in a ‘dual-degree’ program. This experience helped me combine a tradition of service to my community, and to all people, coming from Morehouse, and a commitment to technically-sophisticated and high-impact inquiry from Georgia Tech. As a math major at Morehouse, and an operations research minor (electrical engineering major) at Georgia Tech, I was well-positioned to pursue doctoral study in OR/MS. This field seemed like the perfect combination of analytic methods and human organizations and systems, one that could enable me to help others make better decisions for social change (though I hadn’t really worked through how decision models and social change could really be connected).

However, it took a few years of travel, professional consulting work and another master’s in OR before I was able to complete my dream of a doctorate from Northwestern University in 1997. There, I was able to write a dissertation on subsidized housing policy using the tools of operations research, and present my work to both engineering and policy audiences. I realized that I might have an opportunity to do public-focused OR/MS as a career, and have been fortunate enough to do so for the past 18 years.