Boeing missing key elements of safety culture: FAA report

The opposite of 1-4 could be described as the "culture of lies, ignorance and fear". Fear is a good strategy for getting people working hard (if not always well) and lies make fear universal. Compartmentalizing information is needed to allow more and more functions to be subcontracted. If the company is extracting maximum value from it's assets this year, it has no incentive to report problems that will only appear in the future - by the time the future rolls around, the share holders have their and the shell of the remaining company can be tossed away. etc.

Also, another HN commentator mentioned how eliminating a culture of lies and retaliation is once it's in place. There's never a guarantee that those revealing a problem won't be punished once regulators turn their backs.

And 5 is only useful once 1-4 are in place. Otherwise, it's a culture of flexibly hiding your shit in different places.

Edit: This article was on HN a while back. https://qz.com/1776080/how-the-mcdonnell-douglas-boeing-merg... Key quote: These decisions, made by Boeing CEO Phil Condit, were made with a close eye on the company’s bottom line ahead of a hotly anticipated commercial-jet boom. An ambitious program of cost-cutting, outsourcing, and digitalization had already begun.

It just confounds me (but explains a lot) that the manufacturer of the aircraft the airlines operate does not share a similar safety culture given that they are in a similar ecosystem (airlines report issues to the manufacturer and the FAA/NTSB all the time)

>> It also noted that employees do not understand how to use the different reporting systems and which reporting system to use and when.

As was noted by the purported insider, re: multiple overlapping systems of record/not-record, Boeing's actual processes themselves are badly in need of overhaul.

This feel like a clear example of where top-down + bottom-up independent read-back verification would have been useful.

I.e. management decides they're going to create Safety Process X using Systems A, B, and C. They do so, then circulate training (top-down). THEN you conduct independent interviews with employees at the bottom, to measure whether the new processes are understood at that level (bottom-up). If results aren't satisfactory, then add additional training or reengineer the processes.

Too often, it seems like this shit gets done at the VP PowerPoint level, and ground reality diverges without anyone noticing.

The map is not the world: interviews with a representative random sampling aren't hard.

Boeing has made numerous missteps in the last 15 years after being the world leader in airliners for around half a century. This only happens when knowledge about how to make a safe product is purposefully discarded and attempts to bring that knowledge back are intentionally ignored. In Boeing's case, it's due to desires for increased profits. They are unwilling to learn these lessons because it costs money that _may_ be there at quarter's end.

No sane organization would ever implement this. If someone repeatedly makes mistakes, they're going to get fired even if the mistakes are unintentional. Anything else is going to cause more safety issues in the long-term as inadequate employees are allowed to proliferate.

Yes-- this represents formal acknowledgement by a regulatory agency. The hope is that agency can now use this formalization to enforce change within Boeing.

Not sure if confirms the cause of those issues or where/when the infection took hold.

Spreading manufacturing all over the US is also more to do with getting kore congressional districts “pregnant” than with national defense. In war you want multiple, as in redundant, supply lines so if one is cut, you can source matériel from somewhere else. What we have is multiple, as in single point of failure, supply lines. Lose one and everything collapses.

Everything Wendover Productions makes is so helpful!

They didn’t do what you asked. They found a way to cheat. And worse, their coworkers and reports know what they did, and see them getting rewarded. The “morally flexible” copy, and the boy scouts leave, or burn out.

In the US we have Cessna and Gulfstream, and in Canada we have Bombardier which designed and sort of made the CSeries/A220 in Alabama in conjunction with Airbus.

The whole Bombardier CSeries fiasco was basically Boeing using the US government to try to kill Bombardier because they had managed to put together a plane that was very competitive with with the 737-MAX in a number of categories. The takeaway though is that it is possible, with significant government support, for a small jet manufacturer to put up a feasible competitor to Airbus/Boeing.

Combine all that with the inherently high costs of running a commercial airline manufacturer, and there just isn't enough demand to support more companies in the space. Changing that would require huge technical breakthroughs, or fundamental changes to how passenger air travel works. Neither of those seem to be likely in the near future.

It would have to be a horizontal play by an existing company with large amounts of capital and relationships, like a Lockheed Martin or something.

Somebody, somewhere will make an electric jet that is good enough. It will be very destructive for the old manufacturers, for old airports, and for many airliners. It won't need the long airways we are used to so we will likely get more point-to-point like travel to/from city centers (multiple sites for bigger cities).

Longer-distance travel will still remain the remit of traditional jets -- but they will have a much smaller market so there won't be much R&D, except through state subsidies and military contracts.

Welcome to Techno-Neo-Feudalism. Your superiors are handling their own compensation quite nicely, thank you. Now get back to work !!

In another thread I read that the EASA and FAA used to send Airbus/EASA engineers to Boeing (and maybe vice versa) who could raise all sorts of hell if mistakes were found. Such a setup seems perhaps harder to "game". I do not know this for a fact, I recall it from reading another debate, so take it as hearsay.

But a few chats with employees and a factory tour isn't the most reliable source to judge this.

I read at least two different sets of problems in this report. But first, some background. In the following paragraphs you can substitute "safety" with "quality" in every instance to get equivalent statements that might be more analogous to your experiences.

There is big letter "Safety Culture". This is what happens when you study emergent behavior that you want to replicate, and try to systematize it as much as possible. For excample - as noted in the report, "Safety Culture" consists of 5 pillars - this categorization is purely the result of research and analysis and post-hoc reasoning. The point of "Safety Culture" is that we noticed some organizations that have (little letters) "safety cultures" or "cultures of safety" which were able to achieve long-term excellence in terms of safety, and decided to study their common elements. A company "implements" a big letter "Safety Culture" in hopes of inoculating and maintaining an actual "safety culture".

A Safety Management System is a tool used to achieve and maintain the Safety Culture. For those not sure of what "X management system" means - it's basically a stack of documentation that defines a meta-process and processes that all of your other processes need to conform to, and by doing so, your employees will be forced into "doing the right thing", and aligning their actions and outputs with the goals of Safety Culture, and therefore eventually getting you an actual culture of safety.

In the worst case when you fail at actually sustaining a real safety culture, an SMS then becomes a tool to enforce a minimal standard of safety, from even the most apathetic employee. This comes at enourmous cost of course. Anyone who has had to wait for 3 different authorizations to get a replacement computer at work has witnessed an analogous situation.

Another point that's relevant is that the "Safety Culture" model that Boeing (and ICAO) is referencing is acutally quite young compared to Boeing's overall age. The Safety Culture references in the report are from 1997. The first edition of the ICAO Safety Management manual is from 2006. Boeing has been building safe plans for decades before these "new fangled" capital letter things have even existed. It's absolutely possible for an organization to build safe product without formalized adherence to the formalized "Safety Culture".

Back to problems identified in the report:

The first is that Boeing rolled out a new Safety Management System (SMS) in the last 5-8 years, along with adopting "Safety Culture" policies. But they seem to have blotched the roll out. The report notes that Boeing has its legacy policies and processes for dealing with safety, and those continue in parallel to the new policies and procedures defined in their SMS. They also noted that employees were skeptical of the sustainability of the SMS - ie, they were not sure if this was just some management fad. Many of the findings about "lack of knowledge" read exactly as I'd expect from someone who apathetically clicked through an online training module because they assumed it was useless fluff, because all the real work they've ever seen was handled through legacy processes. Note that a blotched roll out is not the predestined result, even in an environment which was previously lacking a real safety culture, or even middling management.

This is a problem, but could maybe be tolerable (from the perspective of short-term safety), except for the fact that it seemed that that legacy backbone has been rotting away in terms of its effectiveness. The dual system surely isn't helping with its effectiveness.

In other words, while this report focuses on Boeing's failure to achieve "big letter" Safety Culture, reading between the lines also implies a general lack of actual safety culture, and a lack of competent change management.

That being said, I think the purchase of several commercial aircraft liners (and the pilot/crew training to go with it) is more than anything a true long term commitment. Safety is one thing, reliability is another. Boeings shtick always was to go for old and tested technology. That had some appeal. But nowadays you can't help but feel that Airbus went with the times and evolved what planes are while Boeing forgot how to do them.

The Boing might actually be too big to fail but their failure, IMHO, looks like what we have today: An inability to make high quality cutting edge aircraft. For the USA, the disaster would be to be reliant on EU/Russa/Brazil/Canada/China for conducting its transportation operations in this massive country.

What happens if people start freaking out when their planes are not Airbus? Would increase in government contracts keeping the stocks and profits the same mitigate the problems?

So maybe Boing has failed already, its just that its still institutionally solvent for one reason or another.

Given that they are largely not responsible in delivering the value that will ensure continued success of the company, signal that risking the lives of people is not a good business strategy, and may act as a wake up call for others in leadership positions that they should be leading towards what is best for customers and the business and not what is going to give them the biggest short term payday.

Grow to dominate so much of the market and of stock and pension portfolios at all costs, that you'll have to be bailed out no matter your incompetence.

So as long as this behavior only gets rewarded and never punished, why would you expect different results?

That's the harm that monopolies do to society and yet somehow they have been even incentivized in recent times.

For a company that purports to be an energy storage and generation business (with cars as an initial application), Tesla remains hugely dependent on their own suppliers. Panasonic occupies a major chunk of Tesla's own Gigafactory and has repeatedly delayed the production of new cells [0], [1].

[0] https://electrek.co/2024/01/15/panasonic-to-soon-make-new-ba... [1] https://www.reuters.com/technology/panasonic-delays-producti...

But to continue the metaphor, a great company will have enough forward momentum that at any time they can pull back on the stick, relying on inertia in their supply chain, designs, customer name recognition, and existing capital assets to briefly zip almost straight upwards really, really fast for a short amount of time. If you want an upwards trajectory for 100 years, it won't work, you'll soon stall, but if you want an upwards trajectory for the next quarter it works phenomenally well!

The root of the problem, I think, is that it's really hard to measure long-term assets like culture and trust, but really easy to game short-term metrics by dumping long-term assets.

And hell, Elon almost ruined Tesla with his drive to over-automate manufacturing.

Getting to work 100s of suppliers, not all of which are directly managed by the OEM, together is happening all day, every day in all industries building hardware.

Seriously, HNs ignorance when it comes to real engineering and manufacturing is really frustrating.

Apple outsources manufacturing and it’s doing just fine. Has not hallowed out their engineering yet.

Great engineering culture doesn't mean a lot if can't sell your product.

I was reading a story about the decisions made by miners in UP Michigan in 1912. You know what they did when shit sucked? They walked and did something else for 2 years. They let the equipment rot with intent going as far as to coerce the machine maintainers to cease and desist with machine hibernation procedures.

That said I think it’s a bit suspicious when so much of the ownership is institutional investors who seem to just own each other, and appear to work against the very interest of maintaining and growing the value of the investment, which is what one would think being a shareholder is all about.

The original sin is paying executives with stocks and especially stock options. It creates catastrophic and corrupting incentive structure.

Their incentive is to raise the stock price long enough to sell some amount of stock options. The company be damned.

There are many examples, but the classic one is Dick Fuld, the CEO of Lehman Brothers. He drove Lehman Brothers into Bankruptcy all while becoming dynastically rich.