Paul J. Densham

A MORE EFFICIENT HEURISTIC FOR SOLVING LARGE P-MEDIAN PROBLEMS

The Teitz and Bart (1968) vertex substitution heuristic is more robust than competing algorithms and yields solutions with properties that are necessary, but not sufficient, for a global optimum solution. All documented implementations of this algorithm, however, use a naive spatial search procedure, whereas a more informed spatial search procedure, requiring considerably less computation to solve any given problem, is possible. An algorithm incorporating this new search procedure, called the global/regional interchange algorithm, is described. As problem size increases, proportionally larger reductions in processing costs occur.

Database organization strategies for spatial decision support systems

Abstract The selection of a data model is an important step in designing a spatial decision support system (SDSS) because the database is the foundation of the system, and other system components must draw upon the database to perform analysis and display functions. The diversity of representations that must be maintained to support these functions often results in a complex database. One way to simplify the implementation of a complex database is to exploit features of a data model embedded in database management software. A methodology for SDSS conceptual database design was developed using the Entity-Category-Relationship approach. To accommodate this conceptual structure in database management software, we reviewed available data models and concluded that the hybrid extended network model is appropriate for use in an SDSS context. To illustrate this methodology, we designed a logical database structure for a locational analysis SDSS.

Decision support for regionalization: A spatial decision support system for regionalizing service delivery systems

The problem of finding an optimal number of regions, and service locations within each region, to serve a dispersed geographical pattern of demand is treated as an interactive location-allocation modeling problem. Distinctive elements of the problem involve controlling the location of region boundaries, solving for multiple objectives that involve minimizing average and maximum distances to clients from central facilities, ensuring that a minimum number of clients are served in each region, and, under user-defined circumstances, using existing facilities in selected locations. A prototype microcomputer-based spatial decision-support system is described. We also discuss its use by members of an Iowa state government Workgroup charged with developing recommendations for the geographical reorganization of educational services provided to Iowa school districts from central facilities.

Domain decomposition for parallel processing of spatial problems

Abstract Spatial models often are not used to their fullest potential because they have massive computational requirements. Existing workstations and microcomputers often must solve these models in batch mode and, consequently, decision makers are unable to explore and resolve complex spatial problems in an interactive and graphical environment similar to that provided by general purpose business software. Parallel processing can solve spatial models at high speed, however, greatly decreasing turnaround times and enabling decision makers quickly to see the results of revising parameters and criteria. To reap these benefits in a parallel processing environment, researchers must recast modelling procedures from their existing sequentially-oriented form to one in which parallelism can be exploited. This process, referred to as domain decomposition, is a fundamental enterprise in parallel spatial modelling. Domain decomposition for spatial problems can be structured by a set of general principles which are described and illustrated using an example from location-allocation modelling.

A Knowledge-Based Approach for Supporting Locational Decisionmaking

A system for providing decision support to people who make locational decisions is described in which the domain-specific knowledge of users is combined with the general problem-solving strategies, techniques, and mathematical models of location analysts. The system elicits and stores separately environmental, procedural, and structural knowledge so that experts in particular problem domains can access, examine, and modify this knowledge. A metaplanner interacts with users to generate scenarios which describe the general problem-solving strategy to be pursued. These scenarios are organised into a series of tractable problems which are solved in a subproblem-solver module consisting of location-allocation and other analytical models. The system enables decisionmakers to examine systematically the results of a series of analyses leading to a desired solution. The approach is suitable for location-selection problems in complex geographical decisionmaking environments.

Stages in the Adoption of a Spatial Decision Support System for Reorganizing Service Delivery Regions

A prototype spatial decision support system (SDSS) was used by a task force appointed by the State of Iowa to redraw the boundary lines of education service delivery regions so that “the total number … is no fewer than four and no greater than twelve”. Details of this regionalization problem are described and key stages are identified in the task forces adoption and use of the SDSS to arrive at recommended solutions. The approach separates the policy task of setting regionalization objectives and criteria from the more technical task of locational analysis which involves searching for alternatives that meet the requirements of the policy criteria. It is concluded that the SDSS contributed significantly to meeting the basic objectives of the Iowa Legislature.