1. Conclusion:
-Influence diagrams represent the relationships
between variables. These relationships are important
because they reflect the analyst's, or the decision
maker’s, knowledge about a problem.
-The construction of such a model often involves
collaboration between an analyst and the decision
maker.
-This collaboration represents an exercise in
knowledge acquisition — the analyst
attempts to construct a model that reflects the
decision maker’s understanding of
the problem domain.
2. THE USAGE OF INFLUENCE DIAGRAM
• building a common understanding of “how things work”;
• facilitating communication among technical experts, decision makers and
stakeholders;
• integrating knowledge from different sources in decision making (e.g.,
science, TEK, etc.);
• encouraging disciplined thinking about cause and effect relationships;
• being explicit about uncertainty, in particular, emphasizing the existence of
competing hypotheses and facilitating informed debate about them;
• defining evaluation criteria;
• determining modeling and information needs directly related to the
evaluation criteria;
• structuring subsequent quantitative modeling (especially when constructed
under more formalized rules to describe inter-related conditional
probabilities);
• documenting the basis for and improving the transparency of expert
judgments
3. Influence diagrams are particularly helpful
• when problems have a high degree of conditional
independence,
• when compact representation of extremely large
models is needed,
• when communication of the probabilistic
relationships is important, or
• when the analysis requires extensive Bayesian
updating