Markov-renewal programming under incomplete information
Scientific decision making involves model building for operational systems and solving them to determine the optimal or an acceptable decision. Generally, as the mathematical model gets closer to the true representation of the system, its degree of complexity increases, rendering solution of the model very difficult. The solution of most practical problems is therefore the outgrowth of several prior developments involving simplified problems. This has been the case in the development of the theory of sequential decision making also. Since the development of dynamic programming, it has grown from simple models with limited practical utility to more complex models with correspondingly wider applicability.