This looks interesting and promising.

Searchers could find lost hikers faster with help from a computer program that brings a modern touch to algorithms used to find lost people since World War II.
The new software, under development at Brigham Young University, takes into account the spot where a hiker was last seen, the nature of the landscape and GPS track logs of others who have bushwhacked the same terrain. Those logs can show how terrain steers hikers one way or another.
"A lot of search and rescue has a fairly mathematical component," said BYU professor Michael Goodrich.

The news article is here and the link to the tech model (A Bayesian Approach to Modeling Lost Person Behaviors Based on Terrain Features in Wilderness Search and Rescue) is here. (google docs)