TY - JOUR
T1 - Distributed artificial intelligence for group decision support. Integration of problem solving, coordination, and learning
AU - Shaw, Michael J.
AU - Fox, Mark S.
N1 - Funding Information:
Designi s a searchp rocessth ate xploresth e elaboration of portions of a design resultingi n a top-downd ecompositioonf the design problem into smaller, more manageables ubproblems. Subproblemasr e further decomposeudn til solvable problemsa re found. At any particularti me during the design process,t he solving of a subproblem is the goal of the designer,w ith the ultimateg oal being the creationo f an artifacto r descriptiono f an artifact.T o satisfy goals, the design agentsm ake changest o the design.T he designp rogressetsh roughm ultiplel evelsof ab-stractiona nd refinemenutn til all goals are satisfied. The architecturoef the designs ystemw ould supportt hep rocesso f goal formationa nddecom-positiona, ndthe selectiono f subtaskus ponwhich to focus designa ttention. The task assignmenot f the design systemi s executedin an opportunistifca shion,a lwayss e-lectingt he perspectivoef a design agent that is the most relevantin termso f constrainsta tisfac-tion. The task-assignmendte cisioni s supported by the blackboarda rchitecture.
PY - 1993/6
Y1 - 1993/6
N2 - Decision support systems (DSS's) for aiding group problem-solving situations have become increasingly important for supporting and coordinating complex organizations. This paper describes a framework for designing, developing, and formalizing group problem-solving systems based on distributed artificial intelligence (DAI). Among the issues, we find the coordination mechanisms and the learning schemes to be of particular importance in supporting group problem solving. Two implementation examples, one on a network of expert systems, one on a multi-agent concurrent design system, are used to illustrate the distributed artificial intelligence approach to group decision support.
AB - Decision support systems (DSS's) for aiding group problem-solving situations have become increasingly important for supporting and coordinating complex organizations. This paper describes a framework for designing, developing, and formalizing group problem-solving systems based on distributed artificial intelligence (DAI). Among the issues, we find the coordination mechanisms and the learning schemes to be of particular importance in supporting group problem solving. Two implementation examples, one on a network of expert systems, one on a multi-agent concurrent design system, are used to illustrate the distributed artificial intelligence approach to group decision support.
KW - Distributed artificial intelligence
KW - Group decision support
KW - Multi-agent learning
UR - http://www.scopus.com/inward/record.url?scp=0027608408&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0027608408&partnerID=8YFLogxK
U2 - 10.1016/0167-9236(93)90046-6
DO - 10.1016/0167-9236(93)90046-6
M3 - Article
AN - SCOPUS:0027608408
SN - 0167-9236
VL - 9
SP - 349
EP - 367
JO - Decision Support Systems
JF - Decision Support Systems
IS - 4
ER -