Markov decision process (MDP) is used to describe dynamic systems with randomness and decision elements. It provides a mathematical framework model for decision makers to make decisions in a random environment, and provides effective mathematical tools for optimization problems in dynamic programming and reinforcement learning. MDP is useful for studying optimization problems solved by dynamic programming. It has been known since at least the 1950s and is used in many fields, including robotics, automation, economics, and manufacturing. 

Markov decision processes are an extension of Markov chains, with the addition of actions (allowing choices) and rewards (giving motivation).