The framework presented so far characterizes statistical decision theory, which covers a broad range of applications and research issues. Virtually any context in which a decision must be made automatically, by a machine or a person following specified rules, is a candidate for using these concepts. In Chapters 10 through 12, this decision problem will be repeatedly embedded into complicated planning problems. Planning will be viewed as a sequential decision-making process that iteratively modifies states in a state space. Most often, each decision step will be simpler than what usually arises in common applications of decision theory. This is because planning problems are complicated by many other factors. If the decision step in a particular application is already too hard to solve, then an extension to planning appears hopeless.
It is nevertheless important to recognize the challenges in general that arise when modeling and solving decision problems under the framework of this section. Some examples are presented here to help illustrate its enormous power and scope.