CAE Deploy: Saving Lives by Better Ambulance Deployment
In a crisis, you don't always have time to stop and think carefully
You may not think about it until it happens to you: how fast will an ambulance reach you in a crisis? In the world of para-medicine and emergency response, minutes can mean the difference between life and death. Because of growing concerns about delays, many jurisdictions have mandated a minimum EMS (Emergency Medical Service) response time. At the same time, however, many EMS organizations have difficulty meeting those mandated times.
Minimizing response time while maintaining operational efficiency is challenging. Common approaches that have been tried—such as increasing the number of ambulances, improving dispatcher training and working conditions, and reducing ER delays—all require substantial funding, and don’t work in all circumstances. An alternative approach is to apply more sophistication to the way in which ambulances are deployed.
In 2006, Actenum and CAE Inc. took the initiative of assembling a consortium whose members possess key expertise to address this issue. The consortium, partially funded by Precarn, complete its work in 2008, having achieved excellent results.
Traditional deployment tactics
The key concept underlying ambulance deployment is that, in order to minimize response time, the vehicles should be distributed across a set of “posts”, according to an expected demand distribution. Keeping ambulances in a single location will provide poor response times, so most jurisdictions use a static geographical resource distribution map in conjunction with a deployment plan that is sometimes called the “system status management plan”. The deployment plan for a given area describes a set of geographically-distributed posts, and the priorities of these posts; it’s usually modeled on historical incident data, and takes the mandated emergency response times into account. The priority of a post may vary by time of day, and may also be different on weekends versus weekdays. However, the plan is static: it does not take the real traffic or demand load into account, for example.
Experienced dispatchers, of course, try to anticipate incidents. If a number of ambulances are on call, resulting in a void in coverage for a specific geographic area, dispatchers redirect an available ambulance to a post in that area. This decision is based solely on the dispatchers’ experience. Each dispatcher accumulates a body of tacit knowledge, and makes his or her decisions based on an understanding of the current situation, which includes consideration of the time of day, weather and traffic conditions, past call history, and population density.
Deployment decisions are not always good, or consistent. Complexity puts an enormous burden on emergency command and control centres: not surprisingly, a recent report released by Ontario’s Auditor General stated that, in 2004, 64% of municipalities failed to achieve their legislated response times.
Enhanced ambulance deployment
The consortium looked at ways to improve the gathering and extracting of situational information, how to enhance situational awareness, reason with information obtained, and propose decisions. The key focus involved looking at how ambulances should be dynamically deployed to minimize response time. The result is a decision tool—based on Actenum’s decision and optimization technologies—that allows an operator to quickly make plans, easily visualize the planning results, and continue making changes as needed.
Actenum provides the core technology in this decision tool, which allows a knowledgeable user to effectively handle the challenges of paramedic command and control centres, including how to redeploy assets when required. This includes proposing which geographic reallocations should occur to maintain the best possible response, taking into account the spatial distribution of the likelihood of various events.
Since decision timing is critical, the tool has been designed to provide superb solutions under time pressure. It is able to handle changes on the fly, when a situation is being resolved, or when the decision-maker’s goals change during the resolution process. More importantly, the tool manages a large number of constraints and preferences that are important in the long run, such as work load on the different units, meal breaks, and so on. If these issues are not taken into consideration, turnaround among paramedics is expected to rise. Also, the tool is a true ‘shared initiative‘ system; it is interactive and collaborative in nature, providing guidance in making decision to skilled operators.
The decision tool is expected to significantly reduce the cognitive load on dispatchers, and thereby allowing them to do a better job. Better response times should result, since ambulances will be redeployed sooner and more efficiently. It should also provide a better working environment for the paramedics, because the tool considers factors such as meal periods and workload.