Dynamic System Status Management

System Status Management (SSM) is the fluid deployment of ambulances based on the hour-of-the-day and day-of-the-week in order to match supply, defined as Unit Hours of Utilization (UHU), with expected demand, expressed as calls for service, in the attempt to provide faster response by locating ambulances at “posts” nearer their next calls.  While the practice is still not unanimously embraced by all services, it has a sound foundation both in the research literature dating back to the 1980’s as well as in practice today.  Experience has shown that ambulance response times can be dramatically decreased using this type of dynamic deployment, but it is also recognized that it is possible to reduce performance when these techniques are not applied properly.  The direction of the results of a system implementation are typically influenced by the system design, competence of the managers creating the plan, and commitment of the workforce in implementing it.  Therefore the best practice is a simple and straightforward implementation that will show positive results quickly.  This methodology ensures a positive return on investment along with garnering the necessary buy-in from staff to make the project a success.

In his article, “System Status Management – The Fact is, It’s Everywhere“,  published in the Journal of EMS (JEMS) magazine back in 1989, Jack Stout explained the concept of SSM and tried to dispel certain myths.  Based on foreseen Geographic Information System (GIS) technology and even general computing capabilities of that time, it was quite logical to assume in his Myth #2 that “no matter how thoroughly the response zone concept is fine-tuned in practice, it cannot be made to cope effectively with the dynamic realties of the EMS environment.”  But systems implemented today around the US are capable of calculating dynamic response zones in a small fraction of a second while even being based on time-aware historic driving patterns making a truly dynamic system status management process a reality.  A practical and proven example of a dynamically functioning system status management application is the Mobile Area Vehicle Routing and Location Information System, or simply MARVLIS.

The following Slideshare presentation does an excellent job of telling the story of why and how the system works:

Advertisements

6 Comments

Filed under ems, Technology

6 responses to “Dynamic System Status Management

  1. Pingback: Measured Response to Response Measures | High Performance EMS

  2. Pingback: GIS for EMS | High Performance EMS

  3. For more information about MARVLIS, here is a video interview with Tony Bradshaw of BCS: http://www.youtube.com/watch?feature=player_profilepage&v=KAW3Oa_e79M

  4. Pingback: Index of Suspicion Includes Me | High Performance EMS

  5. Antonia Singleton

    Bryan E. Bledsoe, DO, FACEP, EMT-P wrote in EMS WORLD:
    The Scientific Evidence
    When I began my literature search into the science of SSM, I was surprised that there was no scientific evidence to support the practice. All of the writings pertaining to SSM were in EMS trade magazines or were written as though the process was based on science.1 Most were written by people who had a proprietary interest in implementing the practice.2–5 The only numbers published relative to SSM were from the city of Tulsa, OK. Following implementation of SSM, response time dropped from 6 minutes, 46 seconds to 6 minutes, 9 seconds—a saving of 37 seconds. However, this savings is clinically insignificant, and furthermore, ambulance maintenance costs were increased by 46% after implementation of SSM because the ambulance fleet was constantly on the road.6
    In reality, it is impossible to predict where and when calls will occur with any degree of certainty. Historical data from a 20-week interval, or even a one-year interval, are statistically insufficient to make any reasonable prediction of call location or timing.

    • I always find it interesting when people challenge the practice of SSM by claiming it is unscientific and then state as fact that it is “impossible to predict where and when calls will occur” when MARVLIS has shown statistically that it correctly forecasts 80-85% of calls within 10-15% of the geography. Computing has changed a lot since Stout doubted the likelihood of being able to process a sufficient amount of data.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s