Decision models for the location of community corrections centers

Community corrections centers (CCCs, or 'halfway houses') represent a community-based justice strategy intended to reintegrate offenders into civil society. Site selection for these facilities is complicated by negative perceptions of CCC impacts on the part of potential host communities. I solve the CCC-location problem through the use of a framework based on value-focused thinking and quantitative decision models. The first of two mathematical models for CCC location is a novel integer programming formulation that incorporates neighborhood characteristics and equity considerations; the second is a straightforward application of the analytic hierarchy process. These models are applied to a case study in the city of Pittsburgh, PA. In contrast to traditional dispersion models, the mathematical programming model identifies sites in high-amenity neighborhoods which are argued to be associated with beneficial CCC client outcomes. Also, I find that sites selected using multicriteria decision models are relatively easy to generate and are competitive with mathematical programming models in terms of objective function measures.