Greedy Search-artificial intelligence:

If we have a heuristic function for states, as defined above, then we may simply measure each state with respect to this measure and order the plan items in terms of the score of the state in the item. So, at every stage, the agent determines which state scores lowest and puts agenda items on the top of the plan which contain operators acting on that state. Like this, the most promising nodes in a search space are expanded before the less promising ones. This is a type of best first search known particularly as a greedy search.

In some situations, a greedy search may lead to a solution very rapidly However, a greedy search can frequently go down blind alleys that look promising to start with, but ultimately don't lead to a solution. Often the best states at the beginning of a search are actually really quite poor in comparison to those further in the search space. One way to counteract this blind-alley effect is to turn off the heuristic until a amount of the search space has been enclosed, so that the truly high scoring states can be identified. Another problem with a greedy search is that the agent will have to keep a record of which states have been explored in order to avoid repetitions (and ultimately end up in a cycle), so greedy search might keep all the agenda items it has undertaken in its memory. Also, this search method is not optimal, because the optimal solution can have nodes on the path which score faultily for the heuristic function, and thus a non-optimal solution will be found before an optimal one. (Remember that the heuristic function only estimates the path cost from a node to a solution).