Dear Gemba Coach,
What happened to factory physics? It used to be all the rage some years ago, but we hardly ever hear about it these days. Was it wrong?
Thank you for an intriguing question. It led me to read … Factory Physics, which I had somehow missed. And it made me feel guilty. You are absolutely right. My first job ever was to program dynamic simulation of stocks, flows, and influences to model systemic behavior. So, yes, it seems obvious that Throughput = Work in process/ Cycle Time – the formula is actually in The Gold Mine.
The gap between expected results from the formula and real-life is obviously variability. Both necessary variability because no two products or two operations have the exact same cycle time – some activities are high work content, others low — and unwanted variability because of Murphy’s Law: whatever can happen will happen, and sometimes does.
For old timers like myself, it seems quite obvious that the instruction-stock-flow (or should I say, instruction- flow in-stock-instruction-flow out) explains pretty much what should be going on, and that what really goes on is explained by the type of variation that gets into that process.
But, as I realize when on the gemba, many of the younger people haven’t been taught any of this. It’s in the computer somewhere. I was recently walking the wards of a hospital and asked the CEO to think of it as an engine. How many people go home every day as a result of:
- Instructions to go home (have they got somewhere to go to)
- The number of available beds and staff in the ward
- The treatment cycle time (ideally, how long does it take to administer the full treatment)
- The demand for treatment (how many people are sent there by their doctor, or sent from emergency ward)
The CEO looked back at me and said, “But I thought lean was all about developing people – you’re treating it in a purely mechanistic way.” Hmmm.
Well, any quantic entity is both a particle and a wave.
Yes, lean is a system to continuously develop people – but what does that mean? It means that we teach people in any system to:
- Keep the overall purpose in mind – This is surprisingly hard to do, particularly when senior executives muddle the issue with functional purposes rather than company purposes. In a hospital, return people to their normal lives as fast as they can and treat them as kindly as possible while they’re in our hands (and do so as effectively as possible overall) sounds like a reasonable purpose.
- Understand what they do – They must fully understand their part of the job, in theory, practice, and experience to deal with a large variety of cases.
- Understand the impact they have on the rest of the process – Know how decisions one makes in one’s professional field often impact other decisions in unpredictable ways and require a lot of discussion to figure it out. Medical, nursing, and organizational specialties are many, but every patient is a single individual.
- Take responsibility when things go awry, figure out what happened, and do what they can to fulfill the purpose – This is usually where we separate the wheat from the chaff, people who’re willing to go above and beyond their jobs, and those who hide beneath bureaucratic rules and boundaries to avoid getting involved. This is also where motivation matters most, as an atmosphere of collective commitment to quality, a spirit of camaraderie and a sense of personal interest in one’s job make all the difference to taking responsibility in practice.
- Take responsibility for their own development and continuous learning – You can force people to do this or that, but not to think and learn, and with adults, learning mostly happens while reviewing experience, good or bad, which requires both self-confidence (to view mistakes with remorse but without guilt) and confidence in colleagues and bosses (to admit to mistakes and discuss them rationally).
With that in mind, an understanding of how processes work people-free is key to have mental models of how people’s actions impact the way work is designed.
And it has a dramatic impact on people-centric development. For instance, the hospital CEO conceded he had never looked into how or why people left the hospital (after a long argument about the other pressing impossible problems he was facing). Without mastering the demand for returning home, nothing makes much sense.
Coaching Gone Overboard
Months later, he confirmed that, sadly, several patients did not have a clear next step destination. Returning home might need some home care. Or going to a further institution, which in some cases was an administrative headache. This failure of demand meant some patients occupied ward beds longer than necessary, and completely screwed up capacity calculations in the ward, creating blowbacks all the way to the admissions line in the emergency ward. Without some understanding of the physics of the process, he would never have seen it.
Truly understanding something is 1/having a predictive model of how things should work out and 2/ the experience to see where something happened differently when they don’t. In Kahneman’s terms, to understand anything you need a baseline (statistically, this is what happens), seeing the difference of the actual to the baseline, and having some way to explain this difference.
Your question makes a lot of sense. Yes, in the 1990s, we were so focused on fixing process people-free by calculating capacity and reducing variability that we did not grasp that processes are what people do and that each person’s understanding of their job makes all the difference in practice. But, fast forward to the 2020s, we’ve probably gone so far overboard with coaching and mentoring and so on that we’ve forgotten to teach the basic mental models people need to understand what goes on around them – how the system behaves “mechanically.”
Creative thinking really occurs when we move from one frame to the other. Helicopter thinking, as we described in The Lean Strategy, moves your mind from the smallest detail to the largest big picture. Here’s another thinking strategy: moving from completely people-centric explanations of expertise, character, who likes who, who hates who, who shows spirit and who does not, to mechanistic thinking of stocks, flows, instructions, and variability. Time to revisit “factory physics,” I guess!