| Description |
Out-of-hospital cardiac arrest (OOHCA) is prevalent in the United States. Each year several hundred thousand people die due to cardiac arrest. The automated external defibrillator (AED) has greatly enhanced survival rates for OOHCA. However, one of the most important factors in successful resuscitation is emergency medical services (EMS) response time. Unmanned aerial vehicles, or drones as they are more commonly called, have routinely been used for remote sensing but there are new opportunities to use drones for medical emergencies due to their high speeds and ease of navigation. While a drone with an on-board AED could potentially reduce response times to OOHCAs, it remains unclear how effective it is compared to ground EMS. It also remains uncertain how a network of AED-enabled drones should be implemented so that it can best serve cardiac arrest patients. This study examines historical out-of-hospital cardiac arrests and develops a new location model, referred to as the backup coverage location problem with complementary coverage (BCLP-CC), to aid in the deployment of a network of AED-enabled medical drones. By explicitly considering overlapping and partial coverage, the BCLP-CC optimally places drones and the corresponding launch sites while significantly improving backup coverage. Results show that 90.4 percent of historical out-of-hospital cardiac arrests in Salt Lake County can be responded to within one minute by using seventy-one drones and sixty-eight launch sites. In addition, 58.9 percent of incidents are covered two or more times, a significant improvement over existing models. The BCLP-CC was then extended to the backup coverage location problem with complementary coverage and capital improvement (BCLP-CCCI) to minimize implementation costs. Analyses results of the BCLP-CCCI show that by upgrading forty-four existing EMS facilities, by building twenty-six new launch sites, and by using seventy-six drones, 90 percent of the historical incidents could be reached by at least one AED-enabled drone within one minute, 65 percent of the demand could be reached by a secondary drone within one minute, and implementation costs could be reduced by 17 percent as compared to the results of the BCLP-CC. Although there are many concerns and limitations associated with medical drones, this study shows that an optimized network of drones has the potential to significantly reduce live-saving equipment travel times. |
