mishap investigation

The weather has been too crummy to fly for ages. I flew regularly through the first half of December and have been grounded by rain and poor visibility ever since. Didn't get into the air at all in January and only made four flights in Feb. In the meantime, I've passed the written test and been reading up for the oral portion of the checkride. Came across this in an introduction in the book Controlled Flight Into Terrain:

...adapted from a United States Navy mishap investigation presentation:

Simply writing off aviation mishaps to "aircrew error" is a simplistic, if not naive, approach to mishap causation. After all, it is well established that mishaps cannot be attributed to a single cause, or in most instances, even a single individual. Rather, accidents are the end result of a myriad of latent and active failures, only the last of which are the unsafe acts of the aircrew. As described by (James) Reason, active failures are the actions or inactions of operators that are believed to cause the accident. Traditionally referred to as "pilot error," they are the last "unsafe acts" committed by aircrew, often with immediate and tragic consequences. For example, forgetting to lower the landing gear before touch down or hot-dogging through a box canyon will yield relatively immediate, and potentially grave, consequences.

In contrast, latent failures are errors committed by individuals within the supervisory chain of command that effect the tragic sequence of events characteristic of an accident. For example, it is not difficult to understand how tasking aviators at the expense of quality crew rest can lead to fatigue and ultimately errors (active failures) in the cockpit. Viewed from this perspective then, the unsafe acts of aircrew are the end result of a long chain of causes whose roots originate in other parts (often the upper echelons) of the organization. The problem is that these latent failures may lie dormant or undetected for hours, days, weeks, or longer until one day they bite the unsuspecting aircrew...

What makes [Reason's] "Swiss Cheese" model particularly useful in any investigation of pilot error is that it forces investigators to address latent failures within the causal sequence of events as well. For instance latent failures such as fatigue, complacency, illness, and the loss of situational awareness all effect performance but can be overlooked by investigators with even the best of intentions. These particular latent failures are described within the context of the "Swiss Cheese" mode as preconditions for unsafe acts. Likewise, unsafe supervisory practices can promote unsafe conditions within operators and ultimately unsafe acts will occur. Regardless, whenever a mishap does occur, the crew naturally bears a great deal of the responsibility and must be held accountable. However, in many instances, the latent failures at the supervisory level were equally, if not more, responsible for the mishap. In a sense, the crew was set up for failure…

But the "Swiss Cheese" model doesn't stop at the supervisory levels either; the organization itself can impact performance at all levels. For instance, in times of fiscal austerity funding is often cut, and as a result, training and flight time are curtailed. Supervisors are therefore left with tasking non-proficient aviators with sometimes-complex missions. Not surprisingly, causal factors such as task saturation and the loss of situational awareness will begin to appear and consequently performance in the cockpit will suffer. As such, causal factors at all levels must be addressed if any mishap investigation and prevention system is going to work.

I was thinking about this in terms of aviation, and then realized that it's applicable to pretty much any type of "pilot error" in a larger situational context. Construction accidents. Employee failures. Conflicts arising from misunderstandings. Catastrophic failure of levees in New Orleans and subsequent failures in emergency management response.

Mostly, though, I'm really happy to find the concept so clearly defined. One thing you can count on the armed forces for: clear delineation of tasks and goals, evaluation of failure for the purposes of education, and implementation of lessons-learned in education.

Turbidity Trouble

I was looking at the news the other night and there was footage of yet another mudslide and another flooding river. There's been a lot of rain in the area this month; more than usual. Watching footage of the muddy brown water I couldn't help but think of the turbidity issues we're struggling with on a couple projects in Issaquah.

Turbidity is the measure of particulates suspended in water. It's measured using a turbidimeter (say that three times fast). A turbidimeter works by sending light through a sample contained in a glass vial and measuring how much actually gets through and how much is deflected by suspended particles. The turbidity of the sample is then calculated in NTU's (Nephelometric Turbidity Units) based on the amount of light deflected. The more suspended particles there are in the sample, the higher the turbidity.

For kicks, here are trubidity measurements for a couple of things around the office
Tap water = 0.5 NTU
Filtered water from the fridge = 0.15 NTU
Regular coffee = 34 NTU
Decaf = 25 NTU

Even though the coffee is dark, it has a relatively low turbidity reading because the coloring is caused by something dissolved (like a dye) and not suspended in the water. Now if your coffee filter is old, or if you actually like your coffee with the grounds you can expect a much higher reading.

The City of Issaquah has a turbidity limit for construction sites of 100 NTU in the wet weather season, which out here runs from October 1 through April 31. This is no different from the County turbidity limit which also calls for a maximum of 100 NTU. But in Issaquah they up the ante: the penalty for exceeding that limit is a red-tag and a $5k fine. All work on site comes to a screeching halt until the NTU's clear, the city is paid off and a new work plan is submitted along with a formal apology for being a bad contractor.

It's not cheap to keep your NTU's below 100 when you're working outside and it's raining. Our sites are fairly large and they're in all stages of construction from building foundations to final finish and landscaping. In order to meet the standard we've got at each site: a full time employee to sweep all roads clear of dirt and take twice daily turbidity readings keeping logs of findings, large holding tanks and sand filters to contain runoff and filter out sediment, crushed concrete and straw bales over bare flat ground and plastic on all unplanted slopes to keep the soil from washing away.

Between the cost of the tanks, sand filters, new employees, ground protection, not to mention the fines... but that's the price we pay for a healthy ecosystem right?

But then I was watching the news with all that muddy water on the screen, and rusted out cars under floodwaters in backyards, and a couch floating by... and I just wondered.

BuilderGrrl & the Flying Fish