For a model of Markov Decision Process, Policy creation via two methods : Value Iteration and Linear Programming