Piotr Jankowski
Department of Geography
University of Idaho
Moscow, ID 83843

Position Paper for the Initiative 17: Collaborative Spatial Decision Making.

Many decisions concerning development, planning, and the management of environment are complex issues requiring the cooperation of various involved parties. Examples of such decision situations include habitat restoration and economic redevelopment problems. These types of problems are characterized by a slow decision making process evolving through a series of meetings, many of them public involving multiple stake holders that represent conflicting interests and agendas. The goal of decision making process in these situations is to seek a solution that can provide a compromise acceptable to the majority of stake holders. The information describing various aspects of the problem is the key to finding consensus solution. The effectiveness of information used by group members may directly effect the outcome of decision making process.

The relationship between the tools and techniques used for structuring and presenting information about the decision problem and their effect on problem solving and decision making performance of teams was noted in many studies on Group Support Systems (Galegher et al., 1990, Bowers and Benford, 1991, Jessup and Valacich, 1993). Their authors tried to discover if the use of information technologies had any positive impact on the effectiveness of group decision making measured by such indicators as: decision outcome satisfaction and the time it takes a group to converge on the consensus solution. Mixed results were found in regard to the benefits of information technology supporting group decision making. Much of the mixed results were most certainly due to differences in the character of the group, different research methods used, and differences in the information technology being tested. These studies also established two important findings:

1) the larger the group the more effective computer decision support systems are, and

2) due to learning effects, the effectiveness of computer-aided group support grows over time.

Many conceptual designs and empirical research questions developed in the field of Group Support Systems can be applied to Spatial Decision Support Systems for Groups (SDSS-G). This is because collaborative non-spatial decision making has similar conceptual characteristics to collaborative spatial decision making. In both cases it is an activity involving a group of people who are jointly responsible for generating possible solutions, evaluating potential solutions, or formulating strategies for implementing solutions (DeSanctis and Gallupe, 1988). The research issues that are more specific to spatial domain are related to the effects of combining tools for structuring and presenting spatial information (maps) with tools for presenting non-spatial information (decision models). The following specific questions/issues may become a part of research agenda in collaborative spatial decision making:

what is a fundamental difference (is there any ?) between non-spatial and spatial decision problems tackled by groups?
what constitutes the set of SDSS-G tools? The common elements of GSS include brainstorming, meeting organization, and consensus building tools. Are there any tools not represented in GSS that should be included in SDSS-G?
Is there a need for multiple criteria decision methods for groups (MCDM-G) or are MCDM too complex to be used in group meeting situations? There is a large number of both non-compensatory and compensatory decision models that can be used for discrete decision problems (small number of alternatives) by an individual decision maker. Researchers have been integrating compensatory models (allowing to trade-off a poor performance of alternative on one criterion for better performance on another criterion) with GIS software to create spatial decision support tools. The compensatory models, however, become burdensome for the user if the number of criteria is large. These models could be replaced by new compensatory MCDM techniques that will be more user friendly, especially in eliciting the criterion trade-offs and decision maker preferences.
New cartographic symbols and map types need to be developed to support collaborative decision making. Some work on the development of maps for collaborative facility location analysis has already been done (Armstrong and Densham, 1995). More work needs to be done on developing symbols and maps that can present group solutions and help in elicitation of trade-offs between decision criteria. The examples of new maps include a pairwise comparison map that could be used to support the elicitation of trade-offs between criteria pairs, and a vote map which would present the ranking position of each voted alternative. The group support maps may be designed as interactive views combining hypertext, images, sound, and animation.
Integration of interactive maps and MCDM-G. The SDSS-G can include interactive geographic visualization tools integrated with multiple criteria decision models for groups. One such system has been already developed at the Universities of Idaho and Washington. It is comprised of a group decision support module and ArcView-2. The group decision support module provides tools and methods for multiple criteria evaluation of alternatives in private and public modes and it is linked through the dynamic data exchange with the customized application in ArcView-2. Thanks to the dynamic link, the solutions generated in the group decision support module can be visualized in the real time on maps and images developed in ArcView-2. The new solutions update, transparently to the group members, maps such that spatial aspects of different solutions can be compared. Maps in ArcView-2 also facilitate the elicitation of evaluation criterion priorities.

The development of such integrated tools presents not only the design and implementation challenges, but more even so an opportunity to explore the basic research question about the dynamics of collaborative spatial decision making supported by the geographic groupware software.

Local vs. distributed (in space and time) collaborative spatial decision making. The collaborative decision making that takes place in a meeting room (local) is currently the most common form of group decision process. Other forms, however, are also possible such as collaborative decision making distributed in space or distributed in space and in time using the existing technologies of videoconferencing and computer networks. The tools for space and time distributed SDSS-G need to be developed, followed by the exploration of the dynamics of distributed collaborative spatial decision making.

References

Armstrong, M.P., P.J. Densham. 1995. Cartographic Support for Collaborative Spatial Decision-Making. Proceedings of Auto-Carto 12. In press.

Bowers, J.M., S.D. Benford (Eds.) 1991. Studies in Computer Supported Cooperative Work Theory, Practice, and Design. New York: North-Holland.

DeSanctis, G., R.B. Gallupe. 1988. A Foundation for the Study of Group Decision Support Systems. Management Science. Vol.33, No.5, pp.589-609.

Galegher, J., R.E. Kraut, C. Egido. (Eds.) 1990. Intellectual Teamwork: Social and Technological Foundations of Cooperative Work. Hillsdale, N.J.: Lawrence Erlbaum Associates Publishers.

Jessup, L.M., J.S. Valacich. 1993. Group Support Systems: New Perspectives. New York: Macmillan Publishing Company.

Biographical Note

Piotr Jankowski is an Associate Professor in the Department of Geography at University of Idaho. He received his Ph.D. in Geography from University of Washington, in 1989, and M.S. degree in Econometrics and Operations Research from the School of Economics in Poznan, Poland. His current professional activities include teaching and externally funded research in the area of GIS, computer mapping, and collaborative spatial decision support systems. Dr. Jankowski has nine years of experience in the areas of computer mapping and GIS applications, mathematical programming, systems simulation modeling, and multiple criteria decision making methods.