Interactive multicriteria decision support: combining rule-based and numerical approaches

Fedra, K., Zhao, C., & Winkelbauer, L. (1993). Interactive multicriteria decision support: combining rule-based and numerical approaches. In: User-Oriented Methodology and Techniques of Decision Analysis and Support. Eds. Wessels, J. & Wierzbicki, A.P., pp. 48-64 Germany: Springer Berlin/Heidelberg. ISBN 978-3-662-22587-5 10.1007/978-3-662-22587-5_4.

Full text not available from this repository.

Abstract

The main objective of the project is to develop and test, in a practical case study, new methods for multicriteria decision support that seek to combine approaches and methods from both traditional OR and new and emerging methods of AI and expert systems in particular. As a consequence of this experimental approach, the actual computer implementation of the concepts developed is a necessary condition for their practical testing in the field.

Currently existing DSSs are only providing support for an analysis that requires a numerical representation of objectives, and thus more or less directly measurable criteria dimensions. To allow the decision maker to base the analysis and selection from a set of alternatives on numeric and symbolic criteria the hybrid multicriteria optimization tool HYDAS (HYbrid Discrete Analysis System) has been adapted and extended. HYDAS combines numerical, symbolic, graphical, and statistical methods to support the decision maker when exploring the solution space and enables the user to arrive at a well-informed decision.

The case study for testing the methodological research is based on an environmental decision making problem, i.e., investment in air pollution control at the city level, but also at the national level, considering the distribution of investment over a large number of cities in China. The different scenarios which can be created by the decision maker using the air pollution control model form his set of alternatives in the multi-stage decision-making process.

The basic simulation model for alternative strategies for development and the decision support tool HYDAS have been integrated within one framework system, called XDSS (extended Decision Support System). XDSS also incorporates a geographic information system module, a number of dedicated editors that allow the easy manipulation and modification of data sets, as well as auxiliary software for file handling, model control, automatic example generation using Monte Carlo techniques, etc.

All components of XDSS use a common graphics user interface which is completely menu driven and makes extensive use of visualization techniques to provide for easy control and coordination of the modules of XDSS. The system is programmed in C, using the X Windows libraries for graphics display.

Item Type: Book Section
Additional Information: Proceedings of the International IIASA Workshop Held in Serock, Poland, September 9–13, 1991
Research Programs: Advanced Computer Applications (ACA)
Depositing User: Romeo Molina
Date Deposited: 04 May 2016 07:28
Last Modified: 27 Aug 2021 17:41
URI: https://pure.iiasa.ac.at/13091

Actions (login required)

View Item View Item