Analysis of tree based search techniques for solving 8-puzzle problem

The 8-puzzle is a sliding puzzle that consists of a frame of numbered square tiles in random order with one tile missing. The more general n-puzzle is a classical problem which can be solved using graph search techniques. The problem of finding the optimal solution is NP-hard. In this paper, we analyze the performance of various informed and uninformed search techniques for finding the optimal solution for a solvable goal state of the 8 puzzle. For using the uninformed search techniques, the problem is converted into a game tree and all of the states are examined till we find the goal. In case of informed search techniques, we do not examine all of the states. Essentially, a search tree which is a component of the game tree is built. In a search tree, we make use of an appropriate heuristic which guides us to take a decision from a current state to the next state in an intelligent manner and hence the state space is significantly reduced resulting in memory and time optimization.