#4: Sometimes, a system is not the safety answer

The FAA announced two weeks ago that every Part 135 operator in the country (plus some air tour operators and manufacturers) would be required to have a Safety Management System (SMS) just like Part 121 air carriers have been required since 2015. In case you are wondering what SMS means, here’s a word cloud I created from the federal registry entry about the proposed rule:

I realize this is not exactly helpful, but you can see how words like “information”, “data”, “system” and “systems” jump out. Short answer on SMS - it’s a one-size fits all attempt at stopping accidents.

The FAA is very excited about it. Here’s a bit of their press release:

The NTSB is even more excited (and apparently has more management folks willing to get descriptive):

First, this is, by definition, incredibly boring. The reason I’m writing about it is because I have an actual Part 135 SMS in front of me and I know what an SMS collection of binders looks like and, well, I remain (as I have ever since the FAA and NTSB first started talking about this for Part 135) incredibly unimpressed.

If you want to do it, more power to you. But do I think this is the silver bullet answer to aviation accidents? No. I don’t think the silver bullet exists and I’m very tired of seeing the folks at the FAA and NTSB pursue it. (I’m also damn tired of the “what” answer to crashes instead of the “why” answer. The “why answer requires looking at each company individually and accepting that sometimes people do foolish things for no good reason and all of that analysis is very time-consuming. The aviation community remains stuck in “what”.)

But I digress.

SMS is nothing grand. It is binders full of documents that discuss procedures and plans and controls and assurances and forms. There are lots of forms. If you have ever worked for any company that does any kind of work then you have likely seen binders like these on the shelf. (I don’t care if your company made widgets; something like SMS was there.)

Keep in mind that in commercial aviation you already have a General Operations Manual and Training Manual and General Maintenance Manual and Hazardous Materials Manual and…well, you get the idea. SMS on top of what you already have, which is already about safety and always has been, is just a new mandated way of saying you need a safety policy that includes your safety objectives and a commitment to those objectives and components for promotion of safety and compliance with regulatory standards and monitoring of operational processes and a means for communicating safety information to employees that explains why safety procedures are introduced or changed and…I could go on and on.

As someone who has read thousands of commercial accident reports, what I see in SMS is an attempt to treat aviation like a monolith. It’s a well meaning attempt, but that doesn’t mean it’s one that will get the job done. The good people (and I understand their frustration) at the FAA and NTSB want to prevent accidents not by simply investigating and studying existing accidents but by piling so many systems in place for operators (redundancy is not a buzzword here; it’s a true description of SMS), that an accident will never occur.

Except, even companies with SMS still have accidents.

All the major airlines (and Boeing) have had SMS for several years. When the 737 Max crashes happened, Boeing had a SMS. The recent door plug issue with Alaska Airlines? Both companies have SMS. United Airlnes with its recent problems? Yep, it has a SMS. So did Rhodes Aviation dba Transair, a Part 121 cargo operator that crashed in July 2021 off the coast of Oahu. The subsequent FAA investigation of the company found it wasn’t adhering at all to basic safety standards. (As that linked FAA press release notes, TransAir flew an aircraft 33 times that was not airworthy.)

There was also PenAir, which crashed in Dutch Harbor, AK in 2019. Here’s how the NTSB felt, in its final report, about that company’s SMS:

Atlas Air which crashed in Trinity Bay, TX in February 2019, killing the flight crew and a jumpseater, had a SMS at the time as well.

I know some Part 135 operators voluntarily obtained SMS in the past few years. They include Air Methods, primarily a medevac company, which has had one since July 2021. Since then, it crashed in January 2022, July 2022 (twice), September 2022, April 2023 and September 2023. Hop-A-Jet, which crashed on I-75 in Naples, FL earlier this year, also has one.

There’s also NetJets, a fractional ownership company (which means you buy a portion of the company so you aren’t officially chartering - you are an owner), which voluntarily has a SMS. Earlier this year, its pilots union took out ads in the Wall Street Journal decrying the fact that the company was threatening to sue to the pilots “over how often they document aircraft maintenance issues.”

NetJets claimed the pilots were writing up more maintenance issues as a pressure tactic in negotiations. Safety seems to be more of a bargaining chip then company standard in this situation.

I would like to know all of the Part 135 operators that already have an FAA-approved SMS and when they got that approval so it could be compared against their individual accident records. (The FAA does not provide this information.) But more than that, we should look at the studies the FAA and NTSB relied upon in their rulemaking to prove that SMS is worthwhile. In the federal register listing, the FAA refers to an Australian study as supportive of SMS effectiveness and writes:

“Although the authors of a 2012 study by the Australian Transport Safety Board acknowledged the prevalence of earlier studies that were inconclusive, they ultimately concluded that ‘recent studies have demonstrated that well-implemented SMS, especially those where the organisation invests effort into the SMS, are associated with enhanced safety performance.’”

I looked at the Australian study, which was based on analyzing 2,009 previously published safety studies. Here’s how many were applicable to aviation:

Three. Three total studies of 2,009 that had anything to say about aviation. I’m a little confused as to why the FAA thought this was a paper worth referencing.

Moving on, the FAA cites a Harvard Business Review report on “…near accidents in dozens of companies across industries and in laboratory simulations.” I looked at that article entitled “How to Avoid Catastrophe” and found that it included only one aviation reference, JetBlue, and it plays hard and fast with the definition of “catastrophe”. The study authors state that JetBlue took “an aggressive approach to bad weather, canceling proportionately fewer flights than other airlines and directing its pilots to pull away from gates as soon as possible in severe weather so as to be near the front of the line when runways were cleared for takeoff…”

An important note - JetBlue is bound by the same regulations as other major air carriers (Part 121) in bad weather. I would be curious to see the weather conditions in which, as the study claims, it flew and other carriers canceled. Regardless, the “catastrophe” in this study is February 14, 2007, when an ice storm hit the NYC region and JetBlue elected to wait it out lined up on the taxiway rather than at the gates. (So, in theory, they could takeoff first when the weather improved.) The end result as reported in the study:

Distressed passengers on several planes were trapped for up to 11 hours in overheated, foul-smelling cabins with little food or water. The media served up angry first-person accounts of the ordeal, and a chastened David Neeleman, JetBlue’s CEO, acknowledged on CNBC, “We did a horrible job, actually, of getting our customers off those airplanes.” The airline reported canceling more than 250 of its 505 flights that day—a much higher proportion than any other airline. It lost millions of dollars and squandered priceless consumer loyalty.

This is not a safety catastrophe. It has nothing, really, to do with flight safety. This is a public relations catastrophe for sure and while I think it’s horrible, it’s not anything that SMS is designed to prevent. This was all about money and marketing and it’s the only thing in this study that has anything to do with aviation. I fail to see why the FAA thought it was worthy of mention.

The FAA did cite some specific accidents in the federal register where fatalities or injuries could have been mitigated if the companies, or “design and manufacturing” companies involved, had SMS. Right off the bat, two of them were Atlas Air (as cited above) and Southwest Airlines which had an engine failure in one of its Boeing 737s in 2018 resulting in one death. These companies already had SMS.

Getting into those lesser known accidents requires another newsletter, however I’m sure this one is long enough. (I appreciate that you stayed with me through this.) I did want to add one more thing here. The FAA estimated 1,907 Part 135 operators, about 694 air tour operators and about 65 aircraft design and production approval holders would all have to implement the SMS in the next three years. I’m going to go on the record and state there is no way the FAA will be able to accomplish this without urging operators to purchase packaged SMS products. The agency is way too understaffed with inspectors, who are already ridiculously behind on many other duties, to get brand new manuals/software approved. In the meantime, I’ll keep reading commercial accident reporters and continue to ask specific questions starting, as always, with “Why?”.

More on Safety Management Systems in a future newsletter. And as I wrote above, if you want one then get one! That’s fine with me! I will remain skeptical of SMS effectiveness and, because I have seen the FAA come out hard for such “systems” in the past without studying them adequately, I am not done here. (Yes Alaskans, that’s a Medallion Foundation reference.)

I am also eagerly awaiting some documents from Princeton University so hopefully a near future newsletter will be all about 1930s diaries! (Not flight safety but still cool. I promise.)