AutoHS: The Intelligent Hospital Search

One of the most important factors in reducing casualty figures during natural or man-made disasters is timely transport of victims to hospitals. Often, crucial time is lost due to the inability to find the hospital that is most appropriate for a victim. While it is not difficult with today's technology to find out the nearest hospital, the problem lies in the fact one does not know in advance whether the hospital has the resources that are required by the victim. Further, even if such information is made available, it is often difficult to find out the current waiting times for these resources. In this paper, we present AutoHS, a cloud-based crowdsourced system that helps disaster victims locate nearby hospitals that meet their medical requirements using data provided by a distributed network of volunteers. Given the multitude of hospitals in an urban setting, it is not enough to just make the hospital distance, resource and the current resource utilization data available to the ambulance. We use a system of ranking of hospitals for each victim based on the above parameters that allows the victim to make an intelligent and informed choice from a manageable list of options. Finally, we compare the performance of our ranking scheme against a naive find-nearest-hospital approach using discrete event simulation.