Why Do BOM Errors Keep Coming Back?

Most manufacturers are familiar with the frustration of fixing a problem only to discover that exactly the same thing has happened again.

A component doesn’t fit. A drawing contains an error. An operator discovers that the manufacturing instructions are wrong. Someone with experience spots the problem, makes an adjustment and gets the job moving again.

The immediate problem is solved, the product is finished and everyone moves on.

But a few weeks later, another job encounters exactly the same issue.

The question isn’t necessarily why the original mistake happened. It’s why the business allowed it to happen twice.

Fixing the job isn’t the same as fixing the problem

Manufacturing teams are often very good at finding practical solutions to immediate problems. When a customer is waiting and production is under pressure, there’s a strong incentive to do whatever is necessary to get the job completed.

That might mean modifying a component, adjusting a machine setting or asking an experienced colleague to explain how something should be assembled.

There’s nothing inherently wrong with finding a workaround. Sometimes it’s the only sensible way to keep production moving.

The problem comes when the workaround becomes the solution.

If the underlying drawing, bill of materials (BOM) or manufacturing instruction remains unchanged, the correction may exist only in somebody’s memory. The next person follows the original information and the problem returns.

And if that error isn’t discovered until finishing, quality control or dispatch, the consequences can be considerable. Materials have been used, production time has been committed and the business may have to repeat work it can only charge the customer for once.

As Stephen Rendall, Operations Manager at Redhill, puts it in The Manufacturing Playbook:

“Mistakes happen. Fix them permanently. Fix them now.”

Give operators a way to correct the information

One of the approaches Stephen describes is using engineering change requests, or ECRs, to capture problems and ensure they are reviewed.

At Redhill, employees can use the ERP system to raise issues with drawings or manufacturing instructions and propose corrections. Management can then review the request and, where appropriate, update the relevant information.

That might involve changing a drawing, correcting a BOM, revising a manufacturing instruction or updating an associated machine pattern.

The important thing is that the correction becomes part of the information used for the next job, rather than something an individual employee has to remember.

Not every request will result in an immediate change. Sometimes a problem reveals a limitation in the available equipment or tooling, which may require further investigation or investment.

But even then, the issue has been recorded and understood rather than simply worked around.

The feedback loop matters as much as the software

An engineering change-request system is only useful if people use it and somebody acts on the information.

If employees repeatedly report problems but never hear what happened, they may eventually stop bothering. That’s why closing the feedback loop matters.

If a change is accepted, tell the person who raised it. If it isn’t, explain why. If further investigation is needed, make that clear too.

Good software makes it possible to report problems. Good management makes people believe it’s worth reporting them.

It’s also important to look beyond the number of change requests being submitted. More requests don’t necessarily mean more problems, just as fewer requests don’t necessarily mean everything is working well.

The real measure is whether the corrections are improving subsequent jobs. Are the same errors still appearing? Is rework decreasing? Are jobs being completed without the same interruptions?

Start with a job that went wrong

You don’t need a major improvement programme to begin.

Choose a recent job that required rework, was delayed or involved somebody having to correct the manufacturing information. Work backwards to understand where the problem originated.

Was the BOM inaccurate? Was a drawing out of date? Did somebody make an assumption? Was the manufacturing sequence wrong?

Then establish whether the relevant information was corrected and communicated to everybody who needed it.

Finally, ask a simple question:

If we manufactured exactly the same product tomorrow, would the same mistake happen again?

If the answer is yes, the underlying problem hasn’t been fixed.

Mistakes will always happen in manufacturing. The objective isn’t to create a business where nobody ever gets anything wrong, but one where problems are identified early, corrected properly and less likely to be repeated.

Because the real cost of a BOM error isn’t always the first mistake. Sometimes it’s making the same mistake over and over again.