Decisions which do not yield a positive benefit should clearly not be chosen. Optimally the
decision yielding the largest utility is selected but as outlined in the foregoing there could be
constraints on the decision alternatives which are not explicitly included in the formulation of
the utility function. In this case not all feasible decision may be acceptable.
4.3 System Modelling
Decision making can be seen as being equivalent to participate in a game where the decisions
(moves) by the decision maker aim to optimize the utility in correspondence with the
preferences the decision maker is representing. The main opponent in the game is nature but
also the individuals of society which by lack of knowledge, by accident or by malevolence
may impose damage to the society must be accounted for. Figure 4.2 illustrates risk based
decision making in a societal context from an intergenerational perspective. Within each
generation decisions have to be made which will not only affect the concerned generation but
all subsequent generations. It should be emphasized that the definition of the system in
principle must include a full inventory of all potentially occurring consequences as well as all
possible scenarios of events which could lead to the consequences.
At an intra generational level the constituents of the game consist of the knowledge about the
system and the surrounding world, the available decision alternatives and criteria (preferences)
for assessing the utility associated with the different decision alternatives.
Figure 4.2: Main constituents in risk based intra-/intergenerational decision analysis.
Knowing the rules (constituents) of the game, i.e. the system, the boundaries of the system,
the possible consequences for the system and how all these factors interrelate with the world
outside the system and into the future is essential for winning the game. For this reason a very
significant part of risk based decision making in practice is concerned about system
identification/definition as well as the identification of acceptance criteria, possible
consequences and their probabilities of occurrence. Playing the game is done by “buying”
physical changes in the system or “buying” knowledge about the system such that the
outcome of the game may be optimized.