Training systems and artificial intelligence

A few days ago, I got a message from a man named Momchil Penev, asking what I think about training systems and their vulnerabilities? It was an interesting question, and I had to tell him that I haven't thought about the subject in depth. Then he had a few more questions, and we started a conversation that I'll explain in some detail below. But in the meantime I started to wonder ... why haven't I spent more time thinking about training systems?

How important is training, really?

We all know that training is important to Quality: if people don't know how to do a job, they can't be expected to do it well. It even shows up in ISO 9001—clause 7.2 (c) says explicitly that if your people don't already know how to do the work, you have to make sure they learn. But as an auditor, I've never found that checking training records contributes a lot to the overall outcome of the audit, other than by checking the box for clause 7.2. If I'm going to find problems in the organization, it seems like they are always somewhere else.

But why? I think there are three reasons:

• Training records usually consist of a list of classes (for office staff) or a list of operations (for manufacturing and logistics personnel), each one signed off by a supervisor or instructor. Often there is no indication what the signature actually means, so all I can check is whether the paperwork is complete. (Remember this point, because we'll come back to it.)
• Most of the organizations I've audited are fairly small, or else they are small units inside large companies. It's hard for incompetence to hide in small companies, because people largely know what their neighbors are working on. I've been more likely to find that an employee's training record shows no signature for a skill that he knows very well, than to find a true gap in some mission-critical competency.
• On the whole, when people fail to follow established procedures, I haven't found that ignorance is the root cause. There's always something else going on. At least in my experience, the employees who take shortcuts that end in disaster know what they are supposed to do, but they choose not to do it.   

Sometimes training is not the problem!

This last point is important. Remember Alaska Airlines flight 1282, that lost a door plug shortly after takeoff on January 5, 2024? The best explanation I've ever read of the root causes behind this accident comes from an anonymous account posted to the Internet by a current Boeing employee. You can find it here. 

• This account makes the point that the door plug blew out because four bolts to hold it in place were missing.
• The bolts were missing because they had been removed in order to fix a different problem with the door plug.
• Then after that problem was fixed the bolts were never replaced because the workers forgot about them. 
• But if they had followed the established repair procedures they could not have forgotten. The only way they could possibly forget was by taking shortcuts.
• Why did they take a shortcut? Not because they were ignorant of the correct procedure, but because the procedure was too cumbersome and time-consuming. And they were already under pressure from management to hurry up.

Or consider a story that I tell in this post from a few years ago. In a factory that I represented, an external auditor wrote up one of our manufacturing personnel because he was running a plating bath and the liquid was at the wrong temperature. The procedure told him that if anything was wrong he should shut down the line and call the responsible manufacturing engineer (who was on vacation at the time). But in fact this employee had decades of experience with plating. He knew that he could compensate for the wrong temperature by making minor changes to the chemical mix in the bath. The end product was indistinguishable from one made according to the instructions, and this way we could ship it to the customer on time rather than delaying the order two weeks until the engineer returned. We got an audit finding, but from the customer's perspective his solution was perfect.* In this case, the reason the employee failed to follow the written procedure was that he was confronting an unforeseen situation, and had the deep knowledge to improvise a solution.

This is why I say that in my experience, ignorance of a procedure hasn't generally been the root cause of failure to comply. By the same token, when I am coaching a team through an 8D report to resolve some failure, I don't accept "Retrain employees" as a corrective action. There has got to be something more substantive behind the problem.  

But training systems can still be more robust

All that having been said, training systems are still necessary. And like anything else, they can always be improved. This brings me back to my conversation with Momchil Penev, who wants to use artificial intelligence technology to improve them.

Penev is the founder of skillia.AI, and his idea is an interesting one. It's a bit of a right-angle turn in our discussion, but let me take a minute to explain what he is doing.

As I say, it's an interesting idea. In general I'm something of an AI-skeptic; and as noted above, training by itself won't solve all Quality problems. But it still has to be done, and the records still have to be up to date! If Penev can find a problem and then solve it, that's how Progress happens. I wish him well. 

POSTSCRIPT: During our conversation, Penev told me that he is eager to get feedback from people in the Quality business about the performance and limitations of his tool. This means you! So while I have no financial stake in Penev's success or failure, I told him I would post his Contact page. You can reach him at https://skillia.ai/contact. Tell him what you think of his ideas, or ask how you can test the tool for free. 

__________

* For further discussion of this finding, see the referenced post.     

 

      

What do you really need?

I always appreciate hearing from my readers, because I learn what is on their minds. Without feedback, it is easy to get stuck in my own little bubble, and to assume that whatever I happen to think must be obvious to the world. Spoiler alert: it isn't. 😃 So it is genuinely helpful to me to hear what other people are thinking about Quality. Their thoughts are invariably based on where they are today in their Quality journey.* Not where I am. And not where some textbook says they could be. But where they are in real life. And addressing real-life circumstances is where the value comes from. 

Last month, for example, I posted about a contract-manufacturer my company used to work with, whose documentation procedures were exemplary. One reader replied as follows:

I liked how the story so well illustrated three separate points: (1) that companies should do the kinds of things that ISO 9001 requires even if they don't care about certification, because it results in Quality; (2) how important documentation IS, and then (3) that documentation doesn't always have to have WORDS (or be wordy) to be most effective. (This was the most useful to me, because I hadn't fully thought this one through.)  

Of course it was delightful to hear that my reader liked the post. At the same time, I could scarcely resist the urge to reply "Well almost." Because for each of these points, it all depends on what you really need.

Should companies do the kinds of things that ISO 9001 requires, even if they don't care about certification? Sure, on the whole I think it's a good idea. I argue as much in this post from four years ago. But there is nothing magical about ISO 9001. My real advice would be: If something is going to mess up your work or ruin your day, find a way to avoid it. The only thing special about ISO 9001 is that it happens to cover many of the ways that most companies can go wrong. So if you implement the steps it recommends—with or without external certification—you stand a better than average chance of catching the things that routinely mess up your work or ruin your day. But there's no guarantee. Companies with ISO 9001 certification can still fail. And if you've got unusual risks, you might need to take unusual measures, regardless whether ISO 9001 mentions anything of the kind. If there's a serious infestation of vampires in your area, invest in garlic.

How about documentation? Again, on the whole it's a good idea. But make sure it's useful. The key here is that one of the things which can regularly "mess up your work or ruin your day" is simple human forgetfulness. How do you protect against forgetfulness? Write it down. If you want something done the exact same way every time, write a procedure. That's why documentation was so important for the contract-manufacturer I wrote about before. Their operation depended on an assembly line, to make the custom components that they sold. And assembly-line manufacturing depends critically on doing the same step the same way, every single time. So they worked from written instructions.

Even simpler operations can benefit from written procedures, as long as they don't get out of hand. Kyle Chambers told me about a client of his, a rancher who offered basic services to other ranchers. For one reason and another he decided that ISO 9001 could benefit him, which is how he started working with Kyle. He didn't have a lot of procedures, because most of his work fell under the umbrella of "general rancher competence": he hired only people who already knew how to do these things. But he found that he did have to specify a few details. One of these was how to clean out a trailer after it had been used to transport animals. So he wrote up a single page of bullet points—what not to forget—and added it to his management system as an official "shit-kicking procedure." Focus on what you need.    

For the third point, that documentation doesn't always have to be in words, ... well there I have to agree. Use whatever it takes. If words get the message across, that's great. But we all know that a picture is worth a thousand of them. 

There are multiple examples to illustrate the limitations of verbal communication, but one of my favorites comes from the movies and not the workplace.

Did you ever see the 1971 movie "Walkabout"? It's set in Australia. A couple of children from the city are abandoned in the Outback by their suicidal father. The girl is in her teens, and the boy is younger. They get lost and dehydrated, but are found by a teenaged Aboriginal boy on his yearlong walkabout. He speaks no English. The girl pleads with him for water, and he stares at her impassively. 

She keeps saying, "We need water. Water! I can't make it any plainer for you." 

Nothing.

Finally the little boy mimes drinking, and says "Glug glug."

The Aboriginal boy laughs, and pounds a stick in the ground. Up comes water.

I looked for the scene on Youtube and didn't find it per se or entire. But there is a glimpse of it in this trailer.

__________

* To be clear, "where someone is in their Quality journey" is almost never a reflection on the person. It's always about the place or business where that person is working, and what level of sophistication the business is ready for when you account for all its context. Remember a couple of years ago, when I wrote about a colleague in ASQ who described her company by saying, "We just can't afford Quality right now."  

      

When doctors go bad

What do you do when doctors go bad?

Of course it happens. Doctors are human beings, and they have the same faults as other human beings. They can be rude, short-tempered, and hard to work with. They can make enemies in the workplace. They can make mistakes with patient care, and sometimes those mistakes are serious. Some doctors commit sexual harrassment. Some doctors are addicted to alcohol or other drugs. All these things happen. But what can anyone—a clinic, a hospital, or the profession at large—do about it?

Is it a Quality problem?

In a sense this sounds like a Quality problem. Certainly when a doctor goes bad, his patients (and colleagues) aren't "getting what they want." But in the same breath we have to admit that it is a very unusual kind of Quality problem!

• As noted, doctors are human beings; and we don't know nearly as much about human behavior as we do about machines. So while a machine may require no more than simple adjustment to correct its behavior, humans are more complicated.
• In fact, humans have free will, and therefore cannot be simply adjusted the way a machine can. So if a doctor is behaving badly, we have to get his consent before we can expect him to behave any better.
• Why do we need his consent?
• Partly to respect his innate human dignity, that he is free and not a slave.
• And partly because if he does not consent, he or his friends may take steps to frustrate the changes we want him to make.¹ 
• There may also be barriers of an organizational kind.² 
• Complainants might fear legal or organizational retribution.
• The offending doctor might be sufficiently valuable to the organization that management takes his side in a complaint.
• The delay built into processing a complaint might give the offending doctor time to build alliances, and might contribute to a sense in the workplace that the offending behavior is considered "normal."

What has been done so far?

"Disruptive behavior" by doctors is not new. I found one article that surveyed the relevant literature, all the way back to a news item in the New York Times from June 20, 1875.³ But it seems that systemic responses are a comparatively recent development. Back in the late 1970's Dr. Kent Neff, a psychiatrist working in Portland, Oregon, chaired a committee created by the Oregon Medical Association to address concerns about alcoholism among doctors working in hospitals. In 1994, he used the lessons learned from this committee to take charge of the Professional Assessment Program at Abbott Northwestern Hospital in Minneapolis, a program to help hospitals and medical groups with troubled doctors.⁴ Neff has continued to champion this topic in the years since then, as have others of course. In 2012, he prepared a presentation⁵ for The Foundation for Medical Excellence, summarizing the research and practical steps that had been taken at that time. And over the last two decades (at least in the United States), the medical community has been making progress on this front.

Why is it happening only now? I don't know. Studies suggest that disruptive physician behavior (DPB) is not uncommon,⁶ and it is obviously detrimental both to the workplace and to patient outcomes. And yet it seems that even during a century when medicine became ever more professionalized, this topic was overlooked or brushed to the side. According to one study, part of the problem is that there are at least 207 different terms used to describe DPB, so that systematic literature searches are prohibitively difficult. It is possible that this question belongs in a history of medicine, and not in a Quality blog. In any event, the topic of how to respond to DPB is alive today.  

Can we approach this as a Quality problem?

Actually, yes! That's more or less the approach that has been taken. After what I said above about how unusual a Quality problem this is, you might be skeptical. On the other hand, remember that I have characterized the 8D process as "a way to apply the scientific method to solving your organization's problems." And in fact, it turns out to be general enough that we can apply it here too.

So, for any medical organization (hospital, clinic, or other medical provider) to innoculate itself against DPB, the steps look something like this:

Ground rule: Don't blame individuals

• Neff's research is absolutely clear that disruptive physicians are often good doctors (at least potentially) who are caught in a maladaptive system. He insists that the process must be respectful, confidential, and based on objective data. All of these provisions are critical parts of getting the consent of the affected physicians to the new culture that you plan to implement.

D1: Name a team

• Because the solution involves a system-level approach, there will have to be a team to implement it.
• More critically, the team has to include top management, because the work needs their support. They are the only ones who can break down the organizational barriers that might stand in the way. Therefore the administration must clearly understand the cost—in dollars!—that DPB drains from the business every year in operational friction, on top of the risk of huge damages in case of patient harm. And remember—when I reviewed the Quality culture of the Mayo Clinic two years ago, one of my findings was that their "whole Quality initiative had unflinching management support over the long haul."    

D2: State the problem

• Spell out clearly what exactly constitutes unacceptable behavior, and also what constitutes acceptable behavior. It sounds extreme, but there are some physicians who have never had these things articulated for them.
• As Peisah et al. state: "How can those who lack internal loci of control or appropriate social and behavioural skills (for whatever reason), behave appropriately if nobody has articulated what behaviour is expected of them, or the consequences of such behaviour?"

D3: Contain the problem

• Every organization will do what it must to keep the doors open while working this topic.

D4: Find the root cause

• A lot of research has already been done on this question, and mostly the root causes aren't different from one organization to the next. Among the common characteristics of doctors who exhibit DPB are:
• High intelligence but poor emotional or social skills.
• A history of suffering emotional neglect or abuse. 
• A maladaptive personality.⁷ (Naturally some of these features interrelate.)
• Genuine psychiatric disorders (e.g., depression).
• "Life in the fast lane."
• Medical illness.
• That having been said, it's still early days. Peisah et al. lament that "much of the abundant literature pertaining to DPB is theoretical, with little empirical investigation."
• This is a long list of causes. There does not appear to be a single root cause that can be proven responsible because turning it on and off turns the effect on and off like a light switch. But Peisah et al. argue, "We make this complicated because it is, and because ironically, the more contributors to the problem, the more there are potential targets for remediation, again justifying our position of hope."

D5: Brainstorm possible corrective actions

• Mostly not needed, because some well-defined corrective actions already exist.

D6: Implement corrective actions

• Staff education and training. Identify champions within the staff for the new culture.
• Write policies and procedures to define a consistent reporting and review process supported by audits of adherence.
• Access to support for physical and mental health, and to interventions when necessary.

D7: Assess risks and learn lessons

• Reflect on how the system attracts and fuels bad behavior.
• Institutionalize the new culture through constant repetition. Again, remember our earlier consideration of the Mayo Clinic. In fact, the literature on DPB highlights the Mayo Clinic in particular as an organization where this kind of cultural change has been successful.⁸  

D8: Close the 8D and thank the team

• It's hard to go wrong with pizza.

Of course there are subtleties that are specific to medical care. And clearly there is a whole literature addressing this topic with no explicit connection to the Quality discipline.

But I find it reassuring that the basic techniques of Quality can be made to fit this kind of problem too, unusual though it be. It's another indication that these tools really are as broadly applicable as we have always said they are. So that's nice to know. 

__________

¹ There is an extensive discussion of this risk in Markian Hawryluk, "Doctors under the influence," The Bend Bulletin, Thursday, December 9, 2010. A successful Oregon program for alcoholic doctors, one which guaranteed their confidentiality and emphasized recovery, was replaced by a program with a more punitive focus. The intent was to prioritize justice for patients who had been harmed. The unintended consequence was that doctors stopped volunteering for the program, and also stopped referring their colleagues to it. 

² Citations to professional medical literature will generally be made with a URL. (And when possible, I will direct the URL to the specific relevant section of the article.) But the first time I reference an article, I will give the formal citation as a footnote. For this point, see specifically: Peisah C, Williams B, Hockey P, Lees P, Wright D, Rosenstein A. Pragmatic Systemic Solutions to the Wicked and Persistent Problem of the Unprofessional Disruptive Physician in the Health System. Healthcare (Basel). 2023 Sep 1;11(17):2455. doi: 10.3390/healthcare11172455. PMID: 37685490; PMCID: PMC10487014. I will refer to the authors of this article in the text as "Peisah et al." As for the article's title, Wikipedia defines a "wicked problem" as "a problem that is difficult or impossible to solve because of incomplete, contradictory, and changing requirements that are often difficult to recognize." 

³ Yes, more than a hundred and fifty years ago. That date is not a misprint. The article surveying the literature is: Petrovic MA, Scholl AT. Why We Need a Single Definition of Disruptive Behavior. Cureus. 2018 Mar 18;10(3):e2339. doi: 10.7759/cureus.2339. PMID: 29796352; PMCID: PMC5959733. 

⁴ Atul Gawande, "When Good Doctors Go Bad," New Yorker, July 31, 2000. (Published in the August 7, 2000, print edition.) 

⁵ Kent Neff, "Understanding & Managing Physicians With Disruptive Behavior," The Foundation for Medical Excellence, Organizational Professionalism, October 19, 2012.

⁶ "The data presented show that disruptive behaviors are frequently observed in the daily life of health professionals, and compromise the quality of care, the safety of the patient, and can lead to adverse effects." See: Moreno-Leal P, Leal-Costa C, Díaz-Agea JL, Jiménez-Ruiz I, Ramos-Morcillo AJ, Ruzafa-Martínez M, De Souza Oliveira AC. Disruptive Behavior at Hospitals and Factors Associated to Safer Care: A Systematic Review. Healthcare (Basel). 2021 Dec 23;10(1):19. doi: 10.3390/healthcare10010019. PMID: 35052183; PMCID: PMC8775368. 

⁷ Compare, for example, this article: Bucknall V, Burwaiss S, MacDonald D, Charles K, Clement R. Mirror mirror on the ward, who's the most narcissistic of them all? Pathologic personality traits in health care. CMAJ. 2015 Dec 8;187(18):1359-1363. doi: 10.1503/cmaj.151135. PMID: 26644545; PMCID: PMC4674404. 

⁸ Mueller PS. Incorporating professionalism into medical education: the Mayo Clinic experience. Keio J Med. 2009 Sep;58(3):133-43. doi: 10.2302/kjm.58.133. PMID: 19826207.      

      

The risk is always there

Last Wednesday—a week ago yesterday—I read two stories that had nothing to do with each other, except they somehow breathed the same air. In a funny but uncanny way they echoed each other.

Amazon lays off 16,000 employees

One story was widely reported. On that day, Amazon laid off 16,000 workers ... and notified them by accident. You can find the basic story multiple places around the Internet—for example, here. Discussion on LinkedIn (for example here and here) fleshed out some of the details. In principle it is no surprise these days when a tech giant announces layoffs. What was surprising in this case was precisely that Amazon failed to announce them! The layoffs just kind of happened all by themselves, or so it seemed, and then were referenced off-handedly in other, subsequent emails. Finally there was an official announcement (you can read it here), but only after the damage had been done.  Amazon employees shared their experience on Reddit. [Quotes are mostly as-is, and not edited for grammar or spelling.]

I work in SLU. I went to check on my baby at 2am and got and email at 230 that I've been let go. What a great day.... 

i got a txt saying check your personal email before coming to work; work email was disabled. so yeah, break up over txt. 

I didn't know until I got to Blackfoot at 0530 this morning, and my badge didn't work. Tried to slack my boss, and no slack access. Went to email, no email access. After being escorted out of the lobby, got the email in my personal account that my position was eliminated.

Dam I’m at Blackfoot and had to pack up my coworkers stuff.

Again, what's remarkable about this story is not that thousands of people were laid off, but that the communications were handled so ineptly.

Dorx Communications shuts off one account

The second story was a lot smaller. A blogger I follow was moving from Rhode Island to Maryland, and tried to cancel the Internet service at his old lodgings. John Michael Greer tells the story as follows:

In the process of getting all my Rhode Island utilities shut down and all my Maryland ones switched on, I contacted my former [Internet] provider―to keep the lawyers at bay, let’s call the firm Dorx Communi­cations, shall we? You can get almost any other imaginable service from the Dorx website with a few clicks, but shutting off service is a different matter entirely. You have to go to chat, put up with the maunderings of their clueless robot, and get trans­ferred to a supposedly live agent, who then does everything possible to keep you from doing what you’ve come there to do. I slogged patiently through the process, and finally got the allegedly live person to agree that my service would be shut off on January 30.

The moment the chat ended, my phone and internet access shut down. When I walked three blocks to the local public library to get online and contacted Dorx to inform them politely that they’d made a mistake, I found that my shutoff order had been redated to that day, January 18. What’s more, once I was in my account, I was unable to access the chat function at all. It was a pretty obvious middle finger from the dorks at Dorx.

What do these stories have in common?

Both stories involve companies doing something unpleasant. In each case, the company behaves in ways that look to be at best careless and callous; and in each case it is easy to slide from that (admittedly generous) perspective towards one that evaluates the company's actions as "a pretty obvious middle finger." But I have to disagree with Greer. When he reads Dorx's behavior as hostile, he is forgetting an important principle that the Internet calls Hanlon's Razor.

"Never attribute to malice that which is adequately explained by stupidity."

Stupidity is always a risk. It is especially a risk in the actions of corporations, and even more especially in the actions of large corporations. 

But why?

At the most basic level, stupidity is always a risk because we are all human beings, and human beings make mistakes. That's why there is such a discipline as Quality in the first place—to protect against the inevitability of human failure.

Stupidity is a risk in the actions of corporations because many of the "actions" that corporations take aren't based on the decision of any single individual. Multiple people have to work together, often through business processes, and nobody sees the whole picture. Outcomes are the result of the processes themselves rather than of a concrete decision. And if a process is not very well-designed, it can misfire. This is why John Seddon says that his first step working with a new client is to send top management to the front office to watch a single order come in, and then to track that order through its life-cycle. Invariably they learn that the company's systems actually work very differently from how they are supposed to work! (I discuss the point at greater length in this post here.)

And stupidity is especially a risk in the actions of large corporations because those organizations have disporportionately many employees on the inside, where their only contact is with other employees and they are insulated from customers. This means that they are (overall) more likely to trust their business processes than the employees of small companies, because they have fewer opportunities for disconfirmation or falsification.* If unhappy customers can walk into your store and disrupt your day, you are more likely** to take care to keep them happy regardless what the process tells you to do.

Constant vigilance

Of course, at some level everyone knows all this. As noted, the only reason we have a Quality discipline in the first place is that everyone knows people are fallible. But it is important to understand that the most any Quality system can ever do is to help. No system can ever eliminate the risk of human stupidity.

And even the best system can grow flabby and ineffective over time. In earlier posts, I have cited Amazon as a company that strove to avoid bureaucratization. Back in November 2019, Franklin Foer wrote in The Atlantic that "In contrast to the dysfunction and cynicism that define the times, Amazon is the embodiment of competence, the rare institution that routinely works."*** And yet here was Amazon, just one week ago, laying off 16,000 employees without remembering to let them know! 

"The embodiment of competence"? Well, times change. (For other examples, search this blog site for "Boeing.")

Ultimately the only way to avoid stupid mistakes is through constant vigilance. Quality systems can help. Caring about your work makes a huge difference. Paying attention, following through ... all these things are pretty basic. But they are also frighteningly easy to lose track of.

Constant vigilance.   

_____

* In the sense made famous by Sir Karl Popper. See also, for example, this discussion of bureaucracy, which highlights the connection to size.  

** Of course this is no guarantee. Small companies fail too. 

*** Franklin Foer, "Jeff Bezos's Master Plan," The Atlantic, November 2019, p. 58.