Decision Support Technology for Public Safety Resource Allocation: Location of Fire Stations in a Fiscally Constrained Environment

The City of Pittsburgh is facing a severe financial crisis and seeks strategies to reduce expenditures while maintaining an acceptable quality of services. Currently the Bureau of Fire accounts for nearly twenty percent of the City’s expenses, and evidence from cities of similar size to Pittsburgh suggests that fire service expenditures may be one source of fiscal economies. This report represents the culmination of efforts to design and implement an information system to aid City decision makers in designing policy alternatives for fire services design. This decision support methodology generates service characteristics for existing and proposed station configurations of Bureau of Fire services in the City of Pittsburgh. Additionally, this methodology develops alternative station configurations that optimize stated goals of the decision makers. This system methodology was developed using best practices from the disciplines of management science, information systems and policy analysis. It is not intended to provide a single recommendation regarding public safety expenditures. Instead, it provides information regarding a collection of public safety strategies that offer, in some way, an improvement over current practice. The goal of this system is to assist decision makers at the City level to identify specific service alternatives and gain deeper understanding of trade-offs between the cost, service and equity implications of these recommendations. The results of this analysis demonstrate significant opportunities for resource savings that may not degrade service quality substantially. For example, we find that reductions of 27% in the number of engine companies and 45% in the number of truck companies in the city of Pittsburgh preserve standard measures of service quality. Results of our models are robust to alternative measures of demand for fire services and service coverage. This study is based on quality data and reasonable modeling assumptions. However, these results should be interpreted as preliminary since this study has benefited from limited feedback from academic researchers but has not been the subject of journal-quality peer review. In addition, there are a number of modeling extensions that may generate even more realistic and accurate results. Nevertheless, we believe that our work provides a basis for informed discussions between various stakeholder groups regarding public safety expenditures.