Airbus "Violated Key Tenet of Lean"?
Airbus Aims to Pull Back Without Stalling – WSJ.com:
This recent article about Airbus talks about facing a challenge of declining airplane orders.
Airbus said Friday that it booked orders for just 16 planes in March, compared with 54 orders in March 2008 and 37 orders the previous year. The company has said it may capture only between 300 and 400 new orders this year, down from 777 orders minus cancellations last year.
The article mentions some of the challenges:
- How to cut off orders without sticking suppliers with tons of inventory, which could bankrupt them?
- How to scale back on labor without having a “brain drain” when orders pick up again?
On the second question, I'll give Airbus some credit for not treating production workers as a disposable, interchange direct unit labor cost:
And laying off skilled workers could cause a brain drain that slows an eventual recovery. “It takes a long time for us to train our folks who design and assemble planes, so we've got to be careful,” said Mr. Williams, Airbus's executive vice president for programs, in an interview at the company's headquarters here.
Hooray for being careful.
Now back to the first question on inventory… this comes back to a question of lead time. If you have long lead times (or wildly fluctuating customer orders), you need inventory as a buffer. A goal should be to reduce lead times (as the supplier) or level out production (as the customer) to help reduce total supply chain inventories. Or, the customer can partner with the supplier to work on reducing lead time through Lean improvements. My MIT masters thesis covers some of these relationships and makes the case that you can't put the cart before the horse. You must reduce the root causes of holding inventory before just blindly cutting inventory.
As an aside, this is a concept that Dell never understood well when I was there. This is just one reason why I don't consider Dell a true Toyota “Lean” company.
While they wanted “just-in-time” delivery of parts to factories, this was partly a ruse and an accounting trick. Large amounts of inventory were staged in supplier warehouses just off site and were delivered every two hours. That part of the equation was relatively easy – figure out what parts you need for the next two hours of production (taking planned orders and running that through the BOM to transmit needs to the warehouse).
One thing that made the Dell scenario complicated is that you often didn't have ALL of the components required to build a given PC or given order. So the computer system had to run through and see what was “Available to Build” meaning all the parts were available. If an order was missing one part, you didn't want to pull the other 20 parts needed for something that couldn't be built.
And why was Dell often missing a key component?
- Long lead times from suppliers
- Wildly variable demand
Dell kept pushing suppliers to Asia (as part of the industry trend). Even mechanical items, like computer cases, were being made in Asia instead of down the road in Texas. As supply chain lead times get longer, you need more inventory. Dell was INCREASING total supply chain inventory, even though they still only had hours on hand on their own books. Or, you had more stockouts, which meant you couldn't build all of your customer orders.
Dell also didn't do much to practice “heijunka” or level loading of sales orders. For all of the typical reasons, Dell's sales incentives drove a huge “hockey stick” effect where production and orders were huge in the last few weeks (and days) of any quarter, in order to hit quarterly production numbers. A friend of mine wrote a great MIT thesis about why Dell should take steps to level out sales, as the non-level production was getting increasingly expensive in many ways. But, I think this thesis was persona non grata with Dell and I suspect that's why it's not on the MIT website.
So back to Airbus… if they want to avoid sticking suppliers with inventory, they should have worked with them earlier to reduce lead times. Or, maybe they can use this slow period to work with them to prepare for the next ramp up (being able to respond with production, not just warehoused inventory) or prepare for the next slow down. Maybe they should all play the MIT Beer Game together as Mike from the Got Boondoggle? blog did with his company?
The airplane industry has learned to avoid big swings in production. Boeing has taken care to not increase production too quickly (but that leads to really long waiting times as the sales backlog increases). This article calls their backlog a “blessing in disguise” because their leveled production means there's work (and jobs) today even with fewer orders. You can now get a plane in 2010 instead of having to wait until 2015.
Boeing's experience shows that sudden shifts in production can be crippling. A decade ago, the plane maker tried to boost output in a short period and quickly faced shortages of parts and qualified staff. Dozens of unfinished jetliners sat outside factories under tents as workers scrambled to finish them. Resolving production problems pushed Boeing deep into losses even as it delivered a record number of planes. Since then, both Boeing and Airbus have tried to avoid big swings in production volumes.
I don't know about you, but that doesn't give me a warm fuzzy. I don't want to fly in a plane that was built under those circumstances. It's one thing to build SUVs without brakes to rework them after a parking lot stint (as Ford did). It's another thing to have poorly trained people off the street building a PC (as some companies might have done)… so that's why Boeing has erred on the side of level production… focusing on quality and safety, I'll give them that much, that they have learned. This article talks about the too-fast ramp-up for 737s and how that “yo-yo-ed” suppliers.
Airbus has been working with suppliers, this article says. I guess that's Toyota-like, to partner and be at the supplier's “gemba.” I'm not sure if having suppliers design and build 80% of the value-added content is Toyota-like….
Now, Mr. Williams said Airbus procurement staff are “walking the shop floor” at suppliers' factories to spot signs of weakness, such as thin staff or insufficient inventories.
OK, so +1 point for Airbus.
But long lead times make it harder to forecast what you need. That was the case at Dell. As suppliers were further away, the lead times increased. It's fundamental in operations that it's harder to forecast out over a longer time horizon. This is true for Dell, Airbus, or Toyota. It's like a law of physics.
“Of course we'd like to have more robust information, but it's very difficult for Airbus themselves to have an accurate forecast,” Mr. Bolette said.
It was harder for Dell, because Dell was true “build to order” at that time. They had to order parts from suppliers based on forecasts. With Boeing, with their huge order backlog (waiting years to get a plane), if their suppliers are FAST enough, they should be able to order from suppliers based on TRUE volume, the way Dell orders parts from the off-site “it's not ours” warehouse, right?
From the day Airbus decides to boost or cut output, its supply chain needs around a year to react through steps such as hiring staff, buying machine tools and sourcing raw materials. To shorten that period, Mr. Williams' team has violated a key tenet of lean manufacturing — keeping parts inventories to a minimum — and squirreled away extra supplies of components that take particularly long to prepare, such as the metal forgings inside landing gear.
Again, a WSJ writer doesn't understand Lean. This is quite a trend. “Low inventory” itself is NOT a “key tenet” of Lean (unless the 1980's book “Zero Inventories” is the only book you've ever read about Lean and you only read the cover). That's putting the cart before the horse.
The key tenet that Airbus (and other companies) might be violating is doing everything you can to reduce lead times. To reduce total supply chain inventory (-1 point to Dell). To first off, keep the line running (meet customer needs in a timely way, -1 point to Boeing) and then keep inventory as low as possible.
Airbus HAS to “squirrel away” parts precisely because of the long lead times. Either Airbus doesn't get it, or the WSJ writer doesn't get it. Could be most likely the latter, but you never know…
Subscribe via RSS | Lean Blog Main Page | Podcast | Twitter @markgraban
What do you think? Please scroll down (or click) to post a comment. Or please share the post with your thoughts on LinkedIn.
Don't want to miss a post or podcast? Subscribe to get notified about posts via email daily or weekly.
- How to Create a Continuous Improvement Culture by Closing the GAPS — with Katie Anderson - March 30, 2023
- Interview with Mit Vyas: Insights on Learning from Toyota, Entrepreneurial Success, and Mindfulness Practices - March 29, 2023
- Recorded Webinar on Building a Culture of Continuous Improvement through Organizational Habits - March 22, 2023
A topic that’s near and dear to me professionally.
Boeing’s approach, while failing to meet customer needs in a timely way, I believe, actually does the customer a service. For decades, the airlines have bought in batches in good times when they could afford it, and suffered the costs of aging fleets in bad times. By level loading production from the good times into the bad, Boeing is doing what airline management has failed to do, giving their customer a fleet more balanced in age. Airbus, I would add, also works primarily to a backlog. The industry rarely builds whole aircraft to forecast. So called “white tails” (without airline markings) are actually verboten at Boeing, being most common among the smaller regional jet producers.
In terms of lead times, you are correct that Boeing can largely order from suppliers through the first one or two tiers based on actual orders rather than forecasts. Two things complicate this: one, the lead times on boutique alloy forgings (turbine disks, landing gear, structural components) are extremely long. Billets have to be purchased based on forecast. In theory, they can be traded between applications if the market mix changes, but a general downturn leaves you hanging.
Second, a big chunk of the aerospace market, much like automotive, is spares and not new make. Spares consumption is cyclical and often unpredictable. 9-11, the SARS outbreak in Asia, and the current economy all served to devastate the airlines, and subsequently, wreak havoc on the best laid plans of the commercial aircraft supply chain.
High inventory turns for machining and assembly of aircraft and major components are one thing, but raw materials and forgings remain largely batch and queue, high cycle time operations.
“We’re helping you” said the supplier to the customer. Trust us, you don’t really want these planes.
That sounds like an arrogant supplier to me.