Determining probabilistic spatial patterns of lost persons and their detection characteristics in land search & rescue

Search theory allows the optimization of scarce resources to find a missing person in the shortest possible time. This work focuses on the central paradigm in search theory from a land-based search and rescue perspective: A search planner must send a resource where the subject is located, which is u...

Full description

Bibliographic Details
Main Author: Koester, Robert J.
Other Authors: Teeuw, Richard Michael
Published: University of Portsmouth 2018
Online Access:https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.765686
Description
Summary:Search theory allows the optimization of scarce resources to find a missing person in the shortest possible time. This work focuses on the central paradigm in search theory from a land-based search and rescue perspective: A search planner must send a resource where the subject is located, which is unknown. Original research furthered search theoery in probability of area (POA) by adding the dementia subject category, development of new spatial models, and creating the International Search and Rescue Incident Database (145,000 incidents). Research methodology required developing new data standards to integrate multiple databases. Twenty new subject categories were formulated from the research, as well as heuristic rules based upon search theory. In addition, several models were developed to help predict the location of missing aircraft. A program called Map-Score was created which quantizes the effectiveness of the models. This information has been widely adopted by search and rescue planners around the world. The second component is the conditional probability that the searcher/sensor would detect the subject if the subject were in the search area; this is known as probability of detection. Sweep width experiments have been carried out in order to develop the methodology, create tools to simplify the process, determine actual sweep width values for visual search, and to examine various correction factors. The visual methodology was modified in order to carry out the two-way search problem of detecting subject shouts in response to whistle-blasts along with characterizing clues on a trail. Finally, with enough sweep width experiments completed it was possible to determine the correlation between the easily obtained range of detection and the experimentally derived sweep width, providing a quick field estimation tool. Optimal resource allocation can be obtained by maximizing the probability of success rate. This is determined by using previous research to determine POA, sweep width values and additional research that determined searcher velocity by GIS analysis. Ultimately, all search theory is integrated into a tactical decision aid, which for the first time will allow the search and rescue planner to easily use search theory.