go back

Decision Making

Cooperative Creativity

The generation of alternatives that is required in any decision making process can be regarded as a creative step if the alternative is sufficiently novel or conceptually different to all previously known alternatives. For example in civil engineering, a discrete number of conceptual types of bridges can be defined with an arbitrary number of different architectural realizations.

Brainstorming is a well-known example that creativity can be increased through group techniques. In cooperative creativity, we envision that at least one group member is an artificial system. First examples in cooperative creativity are based on measures of novelty and interestingness in design space exploration. They can be used in interaction with the engineer to highlight design regions that have been largely unexplored.


Decision Making

Decision making is the process of identifying and choosing between alternatives based on predefined values or preferences. Decision making can be rational by weighing up or logically combining different preferences based on explicit or tacit knowledge. Including emotional or aesthetic aspects in non-rational decision making is important for understanding, modelling and interacting with the human decision making process, as e.g. in behavioural economics. Furthermore, high-dimensional and dynamic decision spaces and the modelling and anticipation of intentions are important research challenges.

Interactive engineering decision making:
In the case of complex engineering design, the number of preferences is very high and the full range of decision consequences unclear. We study decision support systems that help the engineer in making informed decisions.

Delayed active decision making:
Cognitive systems have to take complex decisions autonomously based on incomplete information. In this case, a delayed decision is beneficial, which can be resolved by incorporating additional information. The research focus is on decision processes based on autonomous and active information search.

For more information

E. Reehuis, M. Olhofer, M. Emmerich, B. Sendhoff and T. Bäck, “Novelty and interestingness measures for design-space exploration”,
Proc. 15th Genet. Evol. Comput. Conf. (GECCO), 2013, pp. 1541-1548.