At the Seattle INFORMS conference, we considered the potential for O.R. at the community level in a tutorial entitled "Community-Based Operations Research" [1]. Community-based operations research (CBOR) is an emerging subfield of public sector O.R. that focuses on community problems, such as poverty, "food security" [2] and homelessness. Former INFORMS President Richard Larson defines public sector O.R. as "quantitative methods and an analytical mind-set applied to problems in which the client is a government entity or non-profit organization [which] is best used to empower individuals to make better decisions, to identify unconventional solutions to difficult problems, and to ensure equality of opportunity for all" [3].

Like other areas of public sector O.R., the outcome measures in CBOR are not directly related to profit maximization, but rather focus on welfare maximization more broadly. In addition, CBOR often identifies equity and sustainability as important complementary objectives, and attempts to account for limited, and often unstable, resources that constrain decision alternatives and limit the scope of solutions that may be implemented on the ground.

Community-based problems place significant emphasis on the needs of the specific populations in well-defined neighborhoods. Such populations are often underrepresented, isolated and/or vulnerable. The local nature of the problems exerts a strong influence over analytic models and policy or operational prescriptions. Examples of CBOR can be found in human services (humanitarian logistics, libraries and literacy, education and family supportive services), community development (housing, community/urban planning, transportation), public health and safety (health care, criminal justice, emergency services, hazardous/undesirable facility location, food insecurity) and nonprofit management.

There is a pressing need to address problems of poverty, food security and housing at a local level: 36.5 million Americans lived in poverty in 2006 [4]; 12.6 million were food insecure in 2005 [5]; and 34 million households had at least moderate housing burdens in 2004 [6]. Research in community-oriented domains has often been descriptive, and concentrated within the social sciences, urban planning, and related disciplines. Less emphasis has been placed on methods within operations research/management science whose models and methods tend to be more prescriptive. We believe that O.R. models and analytic methods, especially those associated with community-based operations research, can significantly improve the quality of life of families and communities facing a wide variety of challenging and localized social problems.

Research Potential and Societal Impact of CBOR


While preparing the Tutorial, we sought to assess the potential of O.R. to address these serious problems and to understand why these problems have been understudied within the O.R. community. To do so, we asked the following questions: (i) research potential: Are CBOR problems rich enough for academic study? and (ii) societal impact: Are solutions to CBOR problems applicable to real communities?

We found that community-focused applications can yield technically challenging problems that are not straightforward adaptations of their private-sector counterparts. Papers have appeared in top journals in the field, including INFORMS journals Operation Research and Transportation Science, including work on library services [7] and public education [8]. The library services work, which began as an effort to improve interlibrary loan operations at a time of stagnant funding and increased demand, led to the development of a new variation of the Period Vehicle Routing Problem. Jan Hayes, assistant director of the North Suburban Library System (NSLS), noted that the project allowed NSLS to "expand our knowledge and understanding of principles and practices which has helped us in the long term with making decisions" [9]. Hayes added that "NSLS was able to implement some changes almost immediately; however, other changers are still in process due to political issues related to service levels of the member libraries."

Our Tutorial highlighted a recent study of food distribution among donors (e.g., restaurants, grocery stores) and agencies (e.g., shelters, soup kitchens) for the Greater Chicago Food Depository [10]. Much research has been conducted on related distribution problems in commercial settings where the goal is either to maximize profit or to minimize cost. However, in non-profit applications, the objectives are often more difficult to quantify since issues such as equity of distribution and sustainability of agencies must be considered. The use of alternative objective functions has important implications on the solution methods used to determine routing and inventory allocation policies. Many of the useful mathematical properties of commercial distribution problems with profit maximizing or cost minimizing objectives (e.g., as in the multi-period newsvendor problem) do not hold with service-based objectives related to equitable food allocation. As a result, new solution methods must be developed. The food distribution work is one of a number of community-based applications that have led to significant analytical and methodological advances.

Community-focused applications can also yield problems in which it is difficult to link model-based prescriptions to improvements in individual and community outcomes. Also, controversy associated with high-stakes social issues may make it very difficult to apply model-based prescriptions to the real world. Successful implementations can be found in journals such as Management Science [11] and Interfaces [12].

Two more recent examples that illustrate these challenges are the design of senior services networks and the location of community corrections centers. Senior centers provide site-based recreation, learning and social supports and are often viewed protectively by clients, who may not wish to patronize centers in other neighborhoods. In addition, seniors are increasingly transit-dependent as they age. However, funding limitations and scale economies in service provision may dictate fewer, but larger centers. These competing needs may be balanced using a multi-echelon facility location model that maximizes net social benefit, a function of utilization and distance traveled [13]. A key assumption in this model is that a rationalized senior network can provide services whose quality is no worse than the status quo, which will result in increased satisfaction with services provided and improved life outcomes. While the first causal assertion seems reasonable, there is little evidence in support of the second assertion. This is not uncommon for services of the kind addressed by CBOR.

Consider, for example, community correction centers (CCCs). CCCs provide alternatives to traditional incarceration and help those who have been incarcerated to transition to civil society. Siting community corrections centers is much more controversial than siting senior centers. A multiobjective model for location of CCCs has been shown to generate siting policies that appear more socially efficient, while accommodating potential local opposition, as compared to classical undesirable facility location models [14]. But will better-sited CCCs result in lower recidivism or reduced community opposition? Though there is some evidence that CCCs may modestly reduce recidivism for some types of clients [15], there is no evidence known to us that CCCs sited according to CBOR will lead to better social outcomes than those sited according to traditional methods.

Importantly, there is an inherent conflict between the "generalize-ability" of problems required to increase the research potential of CBOR and uniqueness inherent in problems that, by definition, are quite localized yet have great potential to impact communities. Uniqueness makes CBOR relevant to populations that might benefit, but generalize-ability elevates CBOR from case studies and pro bono work to scholarship. For example, a model for provision of affordable housing in a particular region may require significant data analysis to generate plausible estimates of social impacts, as well as complex objective functions and constraints that reflect special local social and political characteristics. Developing an affordable housing model that may be applied to many different regions may require more stylized models to generate nontrivial analytical insights, but which may not fully address the needs of any particular local affordable housing provider.

Integrating Scholarship, Service and Action


The questions raised by CBOR research as described above motivate the need for further work in this area, and a flexible mindset that embraces counter-intuitive insights and competing viewpoints. Larson, a long-time innovator within public-sector O.R., notes that "perhaps the best way to achieve the outcomes CBOR wants is to consider very long-term modeling approaches, in which a neighborhood focus is de-emphasized in favor of a systems view." For example, in the long run, family level social interventions, combined with O.R.-enabled policy design, may provide the greatest potential to reduce public subsidies and improve family and neighborhood outcomes. These concerns must be addressed through a combination of academic scholarship to develop principles governing CBOR, collaboration with community-based stakeholders to ensure that analysts are addressing significant problems build community support for participation in policy initiatives, and action research that establishes the efficacy of social interventions based on prescriptive models.

Future of CBOR


Community-based O.R., as we have defined it, has a fairly modest profile within the larger field of operations research, as measured by university courses, journal articles and funded research. Indeed, Ed Kaplan has noted that "operations research applications in the community have been rare relative to other areas of endeavor" [16]. However, the potential for positive social contributions by CBOR is quite large. In the same paper, Kaplan describes multiple applications of community-oriented operations research that have had significant real-world impacts, such as a budget allocation model for prevention of HIV that, despite resistance to its use in the United States, has been applied by the World Bank in Central America and Asia. INFORMS President-elect Cynthia Barnhart maintains that one way the profile of O.R. may be raised is by emphasizing the local and social impacts of O.R.-based solutions, such as the food bank scheduling problem mentioned previously [17].

We have a number of ideas to increase the influence of CBOR (and public-sector O.R. more generally). First, an increase in undergraduate and graduate course and workshop offerings in this area will serve to increase the pool of researchers who are familiar with problems, methods and theory related to CBOR. Academics across universities could create a series of public-sector case studies, particularly in the area of CBOR. Universities could promote field studies and capstone design projects with community-based clients.

Second, an increase in the visibility of assessments of the public policy and equity implications of many different applications, including those not designed with public impacts in mind, may serve to sensitize the O.R. community to aspects of CBOR. Third, collaborations across disciplinary lines, especially social sciences, education and human services, will enable O.R. researchers to reach multiple audiences and find many new problems that can be addressed using CBOR. Finally, many international applications of CBOR are possible, especially since government has a larger role in policy and daily life in most countries outside of the United States.

There are encouraging signs of change with respect to CBOR. INFORMS member Jim Cochran of Louisiana Tech is establishing a new society called Community of Operations Researchers for Public Sector Efforts Forum whose focus overlaps CBOR to a substantial degree. The IIE Conference of Fellows recently identified eight "grand challenges" that organizers say the field of industrial engineering should address. At least three of these are salient to CBOR, including sustainable society and food and water security. Cynthia Barnhart has advocated field study for O.R. students, and Wayne State University professor Ken Chelst has secured $3 million in funding from the National Science Foundation to develop O.R. curricula for high school students; both initiatives have the potential to highlight local, community-based applications of O.R. Finally, the INFORMS 2008 fall National Conference, which will take place in Washington, D.C., is inspired by the historical and current impact of O.R. on important social issues. This represents a great opportunity to highlight innovative applications of community-based operations research.

Acknowledgment: This work was supported in part by the following NSF grants: CAREER: Public-Sector Decision Modeling for Facility Location and Service Delivery; CAREER: Strategies to Improve Goods Movement: Operational Choice in Routing; and NSF Grant 0654398: Design and Control Principles of Non-Profit Supply Chains.

Karen Smilowitz (ksmilowitz@northwestern.edu) is an associate professor in the Department of Industrial Engineering and Management Sciences at Northwestern University and holds a joint appointment with the Transportation Center. She received her Ph.D. in Civil and Environmental Engineering from the University of California, Berkeley, and her BSE in Civil Engineering and Operations Research from Princeton University.

References

1. Johnson, M.P. and K. Smilowitz, 2007, "Community-Based Operations Research," in (T. Klastorin, ed.), "Tutorials In Operations Research 2007," Hanover, Md.: Institute for Operations Research and the Management Sciences, pp. 102-123.
2. "Food security" and "food insecurity" are government terms describing populations that have access to sufficient or insufficient food, respectively, according to government thresholds.
3. Interview, Jan. 2, 2008.
4. Current Population Survey, 2007 Annual Social and Economic Supplement.
5. Economic Research Service, 2007.
6. Joint Center for Housing Studies, Harvard University 2006.
7. Francis, P., Smilowitz, K., and M. Tzur, 2006, "The Period Vehicle Routing Problem with Service Choice," Transportation Science, Vol. 40, No.4, pp. 439-454.
8. Eisenstein, D. and Iyer, A., 1996, "Separating logistics flows in the Chicago Public School system," Operations Research, Vol. 44, No. 2.
9. Interview, Jan. 14, 2008.
10. R. Lien, S. Iravani, and K. Smilowitz, 2008, "Sequential allocation problems for nonprofit agencies," working paper, Northwestern University, Evanston, Ill.
11. M. P. Johnson and A. P. Hurter, 2000, "Decision support for a housing relocation program using a multi-objective optimization model," Management Science, Vol. 46, No. 12, pp.1569-1584.
12. Bartholdi, J. J., Collins, R. L., Platzman, L., and W.H. Warden, 1983, "A Minimal Technology Routing System for Meals on Wheels," Interfaces, Vol. 13, No. 3, pp. 1-8.
13. Ozaltin, O and M.P. Johnson, 2007, "Redesign of Elderly Service Network," presented at INFORMS National Conference, Seattle, Sept. 5, 2007.
14. Johnson, M.P., 2006, "Decision Models for Location of Community Corrections Centers," Environment and Planning B: Planning and Design, Vol. 33, No. 3, pp. 393-412.
15. Lowenkamp, C.T. and E.J. Latessa, 2002, "Evaluation of Ohio's Community-Based Correctional Facilities and Halfway House Programs: Final Report," Cincinnati, Ohio: University of Cincinnati, Division of Criminal Justice.
16. Kaplan, E.H., 2008, "Adventures in Policy Modeling! Operations Research in the Community and Beyond," Omega, Vol. 36, No. 1, p. 1.
17. Horner, P., 2007, "Doing Good with Good O.R.," OR/MS Today, December 2007, pp. 44-49.

Table of Contents

OR/MS Today Home Page


OR/MS Today copyright © 2008 by the Institute for Operations Research and the Management Sciences. All rights reserved.


Lionheart Publishing, Inc.
506 Roswell Rd., Suite 220, Marietta, GA 30060 USA
Phone: 770-431-0867 | Fax: 770-432-6969
E-mail: lpi@lionhrtpub.com
URL: http://www.lionhrtpub.com


Web Site © Copyright 2008 by Lionheart Publishing, Inc. All rights reserved.