BOOK EXCERPT: Optimization models, once the domain of military logistics and business planning, are now also widely used for biological conservation. Margules and Pressey (2000) and Pressey and Cowling (2001) include reserve selection and design models (or “algorithms”) as part of a multistage process for systematic conservation planning. Optimization models and algorithms may serve as stand-alone tools for reserve site selection and design, or be integrated within decision support systems, based on geographic information systems (GIS), that perform site selection along with data analysis and mapping. This chapter provides an overview of what optimization models are, and attempts to clarify their role in relation to: (1) the way reserve site selection problems are described and formulated, (2) methods and algorithms that are used to selection reserve sites, and (3) decision support systems that contain optimization models or algorithms as one of several components. In addition, to illustrate how simple optimization models can be used for reserve site selection and design in the real world, this chapter incudes a case study of North America’s Inland Temperate Rainforest. This chapter is intended for conservation practitioners who have a modest (or no) background in mathematical optimization, operations research, and computer science...The need to generate, evaluate, and display alternative reserve plans rapidly has led to the development of computer-based Decision support systems (DSS)s. A well-designed DSS is an interactive software system that will help the user compile raw data, documents, personal knowledge, and/or models to identify and solve problems and make decisions. .. Optimization models are primarily tools for decision support as opposed to decision making. Their purpose is not to dictate a course of action but to provide guidance, information, and insight for planning and decision making. To this end, optimization models provide several benefits for reserve site selection and design, in addition to finding mathematically optimal solutions. First, a model formulation provides a concise, exact statement of the problem. The model may not match perfectly what is needed in the real world, but it is transparent and is therefore open to criticism and possible improvement. Although the model’s underlying mathematics may be inaccessible to nonspecialists, the purpose and function of each mathematical component (objective and constraints) can usually be explained in plain language. Second, optimization models are highly general in that they can be transferred from one setting to another, from one dataset to another, and potentially from one spatial scale to another. In addition, results are replicable; different users employing the same model and the same data will obtain the same result (although alternate optima may exist). Third, although the mathematically optimal solution may not be the best solution, it can establish absolute benchmarks or baselines for the criteria being modeled that can help put other (perhaps more practicable) alternatives in perspective.
AUTHOR INFORMATION :
Justin C. Williams
Department of Geography and Environmental Engineering,
Johns Hopkins University, Baltimore, MD 21218, USA
Microsoft Scholar Page