Robust Decision Making: The Cognitive and Computational Modeling of Team Problem Solving for Decision Making under Complex and Dynamic Conditions
Abstract
The focus of this work is to understand impactful aspects of team functioning as they solve complex problems, and propose the means to improve the performance of teams, under changing or adversarial conditions. By analyzing the characteristic differences between high performing and low performing teams, we identified behavior that characterized and differentiated the problem solving approach of the teams especially when the problem itself changes during solving. We then derived a simulated annealing-based algorithm that mimicked human team problem solving under these conditions; aspects of the algorithm also performed well as an optimization algorithm when solving a variety of hard numerical problems. Identification of the impact of team structure on problem solving behavior under changing conditions indicates that heterogeneous-grouped teams outperform homogeneous or non-grouped teams. Yet solving multiple problems sequentially is not as effective as breaking problems into smaller intermixed chunks, breaking fixation. Finally, a meta-analysis on the design literature where the impact of examples on problem solving was studied within the design process identified under what conditions examples will induce fixation or alternatively, inspiration.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 14, 2015
- Accession Number
- ADA625533
Entities
People
- Jonathan Cagan
- Kenneth Kotovsky
Organizations
- Carnegie Mellon University