The Opportunity For Failure

It is in our nature – experiment and try things out. We are, actually, really good at it! Amateur Radio Operators makes cool things work. And often it is the seed for a professional system to follow and be perfected. Commonly we can take our cool things to the public (public service event) and wow some people. Most of the time it actually works rather well, but then there is the opportunity for failure when it really counts.

In Safety Management Systems (SMS) we rate the likelihood of a failure and then the  severity of the outcome. As a result we get a risk factor. Can you trust the service when it really matters?

We, the ham community, have about perfected voice communications. And since we now, to a great extend, use commercial equipment for repeaters and not so much home brew equipment, even “new” modes like DMR is rock solid. I doubt you ever did an event on D-Star, but if  you haven’t yet, you will do an event on DMR. Just a matter of time.

As a pilot, I read maps like most people read books. The maps are fascinating to me and the value of maps and APRS is invaluable – but only to a person who can read the map! The key is – can you trust the information?

So we take a RF Tracker and slap a mag mount on the SAG car, we ask all the hams to download XYZ app for their cell-phone and voila – we have an event painted on a map. Kind of when generalities are acceptable.

What is the opportunity for failure, and how serious is the outcome if decisions are made based in this potentially bad information? Do we send a paramedic unit to a trauma-crash at the last (unverified) APRS location? What happens when the patient receives late care after discovering that the target had actually moved, but the maps didn’t update? Maybe the droid device went to sleep and stopped updating?

When designing the APRS Map System, do we look at the “Opportunity For Failure”? We should, because we are convincing our event coordinator to trust us and our technology. No major events wants amateurs in key roles! Maybe this is why we slowly but surly are being replaced by commercial equipment and pro’s?

Causes for Concern may be:

• RF Based APRS Coverage – Let’s face it, in most areas APRS coverage is dismal and certainly not down to “every inch” of the event. I do apply portable i-gates to identified weak-spots, but for a large scale event there will be voids.
• Cell Based Apps going to sleep – I tracked my ham-licensed son from his cellphone all the time when he comes visiting. Sooner or later his phone goes to sleep or the battery goes low and progress stops. Worked great for a while!
• View of the sky for GPS coverage – Living in the Berkshires (or a city) we know all about this. Unless the GPS is on the roof of the car (mine is the back window for protection), between terrain/buildings and the relative direction of the car, GPS coverage goes from excellent to non-existent. Or, at best, the accuracy varies.
• Using Random Equipment – Have you ever asked a group of hams to QSY to a random frequency and PL tone on the fly? Interesting exercise! ’nuff said! Do i trust (s)he can operate their just downloaded app and keep it alive for a 12 hour event – or to turn it on when a panic situation is right next to them?

Here comes Risk Analysis into the picture:

The basic question is – What is the opportunity for BAD INFORMATION being given to the event organizers?

RISK PROBABILITY TABLE

 

SAFETY RISK SEVERITY TABLE

 

SAFETY RISK ASSESSMENT MATRIX

RISK PROBABILITY	RISK SEVERITY
	A CATA-STROPHIC	B HAZARDOUS	C MAJOR	D MINOR	E NEGLIBLE
5 FREQUENT	5A	5B	5C	5D	5E
4 OCCASIONAL	4A	4B	4C	4D	4E
3 REMOTE	3A	3B	3C	3D	3E
2 IMPROBABLE	2A	2B	2C	2D	2E
1 EXTREMELY IMPROBABLE	1A	1B	1C	1D	1E

THE NUMERIC COLOR CODE

The above code (and color) is almost self-explanatory;

• If Injury is not a part of the “failure”, even red can be an acceptable risk because accuracy or failure is only inconvenient.
• If Injury or technical failure is not acceptable, we need to be in the green.

THE SWISS CHEESE EFFECT

The Swiss Cheese model of accident causation is a model used in risk analysis and risk management, including aviation safety, engineering, healthcare, emergency service organizations, and as the principle behind layered security, as used in computer security and defense in depth. It likens human systems to multiple slices of swiss cheese, stacked side by side, in which the risk of a threat becoming a reality is mitigated by the differing layers and types of defenses which are “layered” behind each other. Therefore, in theory, lapses and weaknesses in one defense do not allow a risk to materialize, since other defenses also exist, to prevent a single point of weakness.

The good news  of the Swiss Cheese Effect is that it takes multiple “Failures” for the holes to align reducing the chance drastically.

Now, look at the typical “Ham” Setup of an APRS event:

 

Equipment & Failure	Chance Of Failure
APRS Tracker Technical Failure	Remote
APRS GPS Coverage below Optimum	Occasional
RF Digi / iGate Coverage or Cellphone Coverage	Occasional
Transport Medium Failure	Remote
Display Failure	Remote

When we combine the numbers of single event failures, in this case, of two Occasional we get Frequent.

If we simply use this to track the front and back of an event and the accuracy required is +/- 1/2 mile and 30 seconds we are in good shape. 

Could we use this to time the event to a fraction of a second accuracy? Obviously NOT!

Could we use this system to dispatch a medic to the correct side of the street? By the time we add the Severity + Probability I think we will be up for criticism at best.

Thus, a system design must start with defining accuracy requirements in distance and time. Then design all the components to operate to this defined standard.

APRS is one of my favorite ham-modes, but I also have great respect for the usability of the data and promises to those I support.

The goal is to project us professionally, not as failure or “half-ass”!