HAEP: Hospital Assignment for Emergency Patients in a Big City

In the largely populated city of a developing country, the ambulance service usually sends an emergent patient to the available hospital with shortest pre-consultation delay. The problem is, a life-critical patient may encounter the lack of treatment resource, such as sickbed, in desired hospitals, and the delay to a next appropriate hospital would cause his death, because non-critical patients already occupied the resources. In the worst case, the service encountering a catastrophe may hold hundreds of people on their way to the hospital and require sickbeds be reserved in advance. In this paper, we propose a resource allocation to balance delay in sending patients to hospitals. We extend the scheme to consider sickbed reservation along the time scale by estimating from the past records in history. As a result, the occupancy is balanced in order to reduce the risk of life-critical patients being delayed. Then we develop an in-hospital waiting queue to keep serious patients waiting locally, when it costs more to reach another available hospital. Simulation results show the substantial improvement of our approach in average delay and number of failure-of-assignment.