3 ms·

How does just-in-time prepare for a fire at the the steering wheel supplier? How is a 30 day supply not sensible?

zdragnar 5y ago
I admit to not having a solid grasp of the numbers in play, but if my limited understanding is correct, a 30 day outage is significantly cheaper than the problems pre-JIT supply chains faced.
jcims 5y ago
Supply chain risk management. There's no black box from which steering wheels emerge. You send people to their factories, you check their finances, you look into their history with regulatory agencies, you examine their leadership, you look at their supplier relationships. The place I work has a hundred folks that do nothing but risk and cybersecurity audits of our vendors and suppliers. It's a whole thing. It's obviously not perfect and can get caught up in systemic issues like a pandemic, but as mentioned in a sibling comment it's still cheaper than the alternatives.
- tiahura 5y ago
  I get that at the end of the day somebody’s got to make the wheel, and the fire could be anywhere. However, the article suggests that Toyota has to shut down all operations the minute their plastic doohickey supplier goes down - ie they have 0 buffer. That doesn’t seem optimal. Given that there are dozens or hundreds of suppliers, on any given day, isn’t at least one of them having issues?
  - stubish 5y ago
    The theory, and the reason Toyota get studied, is that having no buffers means no waste and overall better production. The plastic doohickey supplier is producing plastic doohickeys at an optimal rate, and when supply stops everything stops. The problem is immediately apparent, and all eyes immediately available to fix the problem because all work has stopped. The alternative is a buffer. The plastic doohickey supplier needs to be able to produce plastic doohickeys at a faster rate than Toyota can consume them, to build up that buffer in the first place and to replenish it when problems occur. This extra capacity makes plastic doohickeys more expensive, both because of the excess production capacity and storage logistics. Many believe, including Toyota, that this thinking is the key to their success. It certainly requires good risk management though, identifying points of failure that will just shut down production for a short while vs. points of failure that risk destroying the company (say, a fire in a unique chip manufacturing plant that could take years to rebuild). See Toyota Production System or Lean Manufacturing.
- 1ris 5y ago
  I can see how these people can confirm risks. But how to they falsify risks? And more importantly: How are they managing it? Is the the size of the supply heap derived from the assumed risk and the cost of a default?
  - jcims 5y ago
    It's tricky going from an engineering mindset to risk management. I still struggle with it and have been in the industry for a long time. You can't falsify risks and many of your inputs are going to be guesses based on an amalgam of experience, historical information, trends and available risk/threat models. Today there's all of these metrics available that give it the feel of an exact science, but its not. It's educated guesswork, but there's enough evidence showing it works that its worth substantially increasing the costs and friction of doing business to align with its outputs. Usually. In this case something arguably failed in the process. It's noteworthy but it's not an existential threat to Toyota. They will learn from it, someone's probably going to 'seek opportunities outside the firm' and they'll get back to making great products.
ISL 5y ago
Multiple JIT steering-wheel manufacturers?
letitbeirie 5y ago
If a car factory is making 1000 cars a day, a 30-day supply is all the parts needed to make 30,000 cars. Managing that is a giant, complicated, land-consuming operation unto itself, and it all has to get sorted before you can build car #1.
- stubish 5y ago
  And all the suppliers need to be able to ramp up supply after a problem to replenish the buffer, which means excess capacity normally sitting idle.
Hamuko 5y ago
Toyota actually does stockpile some supplies it deems critical. This is despite the fact that Toyota is basically the pioneer of the JIT method of building cars. https://www.france24.com/en/live-news/20210512-lessons-from-2011-disaster-help-toyota-ride-out-chip-shortage https://www.france24.com/en/live-news/20210512-lessons-from-...
stickfigure 5y ago
Warehousing parts adds "fire at the warehouse" risk. You could have redundant warehouses, sure, but at some point you have to decide how much expense and operational complexity you're willing to tolerate to remove more 9s of risk.
foepys 5y ago
I don't know about other countries but German auto manufacturers require (nearly) all suppliers to submit an FMEA [1] for their products. Some or maybe all US manufacturers do this, too, if I remember correctly. In short, FMEA is a way to calculate risks and prepare for supply chain issues to a certain degree. When done correctly those FMEAs get very large and extensively lay out where risks are high so plans can be made to minimize them. 1: https://en.wikipedia.org/wiki/Failure_mode_and_effects_analysis https://en.wikipedia.org/wiki/Failure_mode_and_effects_analy...

I admit to not having a solid grasp of the numbers in play, but if my limited understanding is correct, a 30 day outage is significantly cheaper than the problems pre-JIT supply chains faced.

jcims 5y ago

Supply chain risk management. There's no black box from which steering wheels emerge. You send people to their factories, you check their finances, you look into their history with regulatory agencies, you examine their leadership, you look at their supplier relationships. The place I work has a hundred folks that do nothing but risk and cybersecurity audits of our vendors and suppliers. It's a whole thing. It's obviously not perfect and can get caught up in systemic issues like a pandemic, but as mentioned in a sibling comment it's still cheaper than the alternatives.

tiahura 5y ago
I get that at the end of the day somebody’s got to make the wheel, and the fire could be anywhere. However, the article suggests that Toyota has to shut down all operations the minute their plastic doohickey supplier goes down - ie they have 0 buffer. That doesn’t seem optimal. Given that there are dozens or hundreds of suppliers, on any given day, isn’t at least one of them having issues?
- stubish 5y ago
  The theory, and the reason Toyota get studied, is that having no buffers means no waste and overall better production. The plastic doohickey supplier is producing plastic doohickeys at an optimal rate, and when supply stops everything stops. The problem is immediately apparent, and all eyes immediately available to fix the problem because all work has stopped. The alternative is a buffer. The plastic doohickey supplier needs to be able to produce plastic doohickeys at a faster rate than Toyota can consume them, to build up that buffer in the first place and to replenish it when problems occur. This extra capacity makes plastic doohickeys more expensive, both because of the excess production capacity and storage logistics. Many believe, including Toyota, that this thinking is the key to their success. It certainly requires good risk management though, identifying points of failure that will just shut down production for a short while vs. points of failure that risk destroying the company (say, a fire in a unique chip manufacturing plant that could take years to rebuild). See Toyota Production System or Lean Manufacturing.
1ris 5y ago
I can see how these people can confirm risks. But how to they falsify risks? And more importantly: How are they managing it? Is the the size of the supply heap derived from the assumed risk and the cost of a default?
- jcims 5y ago
  It's tricky going from an engineering mindset to risk management. I still struggle with it and have been in the industry for a long time. You can't falsify risks and many of your inputs are going to be guesses based on an amalgam of experience, historical information, trends and available risk/threat models. Today there's all of these metrics available that give it the feel of an exact science, but its not. It's educated guesswork, but there's enough evidence showing it works that its worth substantially increasing the costs and friction of doing business to align with its outputs. Usually. In this case something arguably failed in the process. It's noteworthy but it's not an existential threat to Toyota. They will learn from it, someone's probably going to 'seek opportunities outside the firm' and they'll get back to making great products.

ISL 5y ago

Multiple JIT steering-wheel manufacturers?

letitbeirie 5y ago

If a car factory is making 1000 cars a day, a 30-day supply is all the parts needed to make 30,000 cars. Managing that is a giant, complicated, land-consuming operation unto itself, and it all has to get sorted before you can build car #1.

stubish 5y ago
And all the suppliers need to be able to ramp up supply after a problem to replenish the buffer, which means excess capacity normally sitting idle.

Hamuko 5y ago

Toyota actually does stockpile some supplies it deems critical. This is despite the fact that Toyota is basically the pioneer of the JIT method of building cars. https://www.france24.com/en/live-news/20210512-lessons-from-2011-disaster-help-toyota-ride-out-chip-shortage https://www.france24.com/en/live-news/20210512-lessons-from-...

stickfigure 5y ago

Warehousing parts adds "fire at the warehouse" risk. You could have redundant warehouses, sure, but at some point you have to decide how much expense and operational complexity you're willing to tolerate to remove more 9s of risk.

foepys 5y ago

I don't know about other countries but German auto manufacturers require (nearly) all suppliers to submit an FMEA [1] for their products. Some or maybe all US manufacturers do this, too, if I remember correctly. In short, FMEA is a way to calculate risks and prepare for supply chain issues to a certain degree. When done correctly those FMEAs get very large and extensively lay out where risks are high so plans can be made to minimize them. 1: https://en.wikipedia.org/wiki/Failure_mode_and_effects_analysis https://en.wikipedia.org/wiki/Failure_mode_and_effects_analy...