Boeing has not been far from controversy in recent years, with highly publicised 737 Max air crashes about 5 years ago and, more recently, a near-disaster on a brand new Alaska Airlines Max which suffered a mid-air door plug blowout putting all passengers and crew at risk. Worse still, the problems keep on coming for Boeing who are now being described as ’embattled’ by the press.
We examine the bad manufacturing habits that seem to have caused Boeing’s recent troubles based on information we have read in the news reports, such as accepting parts from suppliers that had quality issues, leaving obvious problems for later and relying on inspections and rework to handle them when planes were complete, and a business culture which placed pressure on workers to deliver planes as quickly as possible to keep orders rolling in.
With these bad manufacturing habits in mind, we also go through best practices that good manufacturers follow so you get an introduction into how to avoid similar quality and reliability risks.
If you prefer listening to reading…
Introduction: Travelled work is a bad habit Boeing have fallen into
‘Travelled work’ is where some issues are seen and not immediately fixed, with the plan being to fix them later…this, perhaps unsurprisingly, sometimes leads to them being missed. This is what led to the Air Alaska 737 Max near-disaster.
Boeing has been a financially driven company for some time, arguably leading to financial results being placed above quality and safety by management. They knew their fuselage manufacturer, Spirit Aerosystems, was delivering parts with problems for some time but weren’t keen to slow down plane manufacturing to deal with them, as lower output would equal lower profits.
Fuselages were coming in with quality issues, such as the door plug with missing bolts that made it into the new Alaska Airlines plane, but despite being noted it became travelled work with the next team perhaps being expected to deal with it as it made its way on the production line. For whatever reason, it was missed, and we know the rest of the story. (00:00)
What do good factories do to avoid problems like travelled work?
The key here is to get things right the first time.
Factories need to put in place the right processes and the correct people for the positions, then empower them to take ownership and responsibility over their own work and the coming upstream work, too. Here are some examples:
- Cross-checking, where the operator next to the person working on a safety-critical part checks it when it comes to them, is a simple best practice we use to reduce quality risks. However, it seems that Boeing may not have implemented practices like this on their assembly lines.
- Risk analysis is a complex topic, but basically, it helps manufacturers to identify and then fix problems before they’re able to become defects, improving product quality and safety.
It may be that Boeing employees were under pressure to keep the aircraft moving down the line and out for delivery as fast as possible, leading to station after station overlooking the problem because they felt it was not ‘their responsibility’ and thinking that ‘the next guy will catch it instead.’ This is the opposite of best practices on the factory floor.
The key metric for this would be first-pass-yield, so in the factory, you examine the process (sub-assemblies and components come in and then end as a finished product) – all along the line what is the percentage of finished products that are good the first time with no rework, etc? For complex products, this could be as low as 70%, maybe less, at first. But if you follow a solid NPI process, this should get into the 90s. (06:15)
Investing time in training suppliers in the upstream supply chain to reduce quality risks
Boeing’s problem was that they had defective sub-assemblies and parts coming into the factory, but should a manufacturer who properly manages the supply chain just accept this? No.
Good manufacturers need to have a good relationship with upstream suppliers, from raw materials to complex components, and understand their processes. Analyze your upstream supply chain’s quality systems and processes, and if they lack what you require, you can invest time to train them and improve both you’re relationship and also the quality coming from them, ensuring that incoming parts are improved and your first-pass-yield is better. Incoming QC will always be important, but if you invest time in the suppliers, hopefully, it is less critical.
Boeing’s relationship with Spirit Aerosystems seems to have been unbalanced. Even though they sometimes issued financial penalties Spirit seemed to be providing them with sub-assemblies (fuselages) that didn’t reach their standards, and Boeing kept accepting them, presumably deciding to fix the problems on the line. It was only very recently that they started rejecting fuselages with issues and sending them back to Spirit to fix, but this is perhaps what should have been happening long ago. Due to the nature of the products being ordered Boeing should probably also have insisted on placing engineers in Spirit with the authority to stop production and help their staff fix problems at this early stage (and charging this back to them as well), thereby helping them to improve the quality of their output.
Was corporate financial pressure from top management a reason why this oversight of Spirit was being overlooked, with quality and production staff rushing to build new planes and not paying so much attention to the quality? It may well have been. In most industries, even for consumer products like apparel or electronics, if you see a problem in the factory you deal with it. You don’t put it aside and forget about it. That’s a huge failure of Boeing given that planes have a very low tolerance for quality issues. (10:08)
What can we learn about dealing with in-process issues from the automotive industry?
In the auto industry operators don’t see an issue in a new vehicle and let it carry on down the line, maybe waiting to fix it after the final check or when it is in the parking lot outside. They also don’t have to stop the entire production line to fix it, rather operators will be able to hit a button or pull a cord to alert their management that an issue has been found.
A team leader, for example, will run over and assess the problem. If it can be fixed within one or two minutes (different companies have different standards for this) they will do it, otherwise, they escalate and more senior engineers will come and check and work on the car while it is still on the line. Suppose there is a problem with the seat. In that case, they can still keep working on or replace the seat even if the car has moved on from that area (maybe someone needs to run with the right parts or maybe they need several people to lift in the seat manually because they are away from the specialized jig they usually use). But it’s highly unlikely that the entire production line will be stopped for, say, 20 minutes because of one vehicle, as they have that system to assess and deal with problems during the process and they leave a buffer of several cars anyway. However, at Boeing, it seems like they did not have a system like this on their assembly lines. (18:24)
Why a focus on inspection and rework later increases risks.
Boeing exhibits 2 big issues:
- They don’t find, contain, and fix problems straight away while feeding back to the source to reduce future instances.
- They wait until the plane is complete before fixing problems, instead focusing on inspection and rework later rather than in-process fixes.
Rework makes things easier for production line leaders because it moves the correction of problems later without production needing to be halted which would be a big headache for them. However, it is more costly than getting it right the first time and it comes with serious quality & reliability risks as by disassembling a complete product to rework something or repairing a part, you potentially introduce new failure modes that can occur later in the field as reliability failures (and that’s very unsafe in airplanes). (27:14)
Summary: Good manufacturers’ best practices.
Finding and fixing one problem, let’s say a surface crack in metal, by welding it and sanding it down, doesn’t solve the cause of it. Maybe that is because the cast metal is too porous and prone to defects, but just fixing one crack won’t find this root cause. A best practice for good manufacturers is to find the root causes of any issue you come across to stop them from occurring in the first place.
Aside from that, to reiterate, the best practices to be aware of are:
- Understand the supply chain and that the suppliers can produce the parts, materials, and sub-assemblies at the right quality standards.
- Help the suppliers to improve their own processes.
- Do IQC on incoming materials and parts.
- Empower your workers to stop production when needed without fear of ridicule or punishment.
- Train staff to handle issues that come down the production line and have the right mentality of wanting to deal with them.
- Prioritize issues by order of seriousness and tackle them by implementing a corrective action plan with the people and process that led to the issue, finding the root cause/s. (33:32)