Engineering Changes (ECs) are fulcrums: They’re powerful, and a small mistake one day can have incredible consequences for weeks, or even months, afterwards. When companies mature, they need systems to limit the impact of potential unintended consequences, but you’ll always hear engineers grumbling about the red tape required to make this “one tiny, obviously logical change”. Where’s the balance: How do you de-risk an ECO (engineering change order), but still ensure efficient changes when necessary?
Why ECO Processes Are Necessary
Design engineers focus on the end product. Manufacturing engineers focus on ease of manufacture and assembly. The sourcing department looks for ways to streamline the supply chain. Each of these perspectives is great, but changes made to a design with an emphasis on any of these can create issues with the others—having an ECO process where all the involved departments have a chance to offer input is critical for avoiding expensive mistakes.
But ECO processes don’t just filter bad ideas, they also ensure that designs evolve in the right direction. For instance, imagine you notice two bolted parts have a tolerance that’s too loose to allow them to consistently mate. You decide to tighten the tolerance of the hole locations and want to make the change immediately. Is that the right choice?
It could be, but there are also other options: Instead of tightening the location tolerance, you could actually open up the size of the holes; or use smaller bolts; or choose a different solution altogether. Which change would be most reliable and least costly? It takes a team to answer that question.
Identifying Stakeholders
Changes are rarely simple. Even for a minor tweak, like tightening dimensional tolerances, a design engineer needs to perform a tolerance analysis, manufacturing engineers need to verify that the machines are capable of the tighter tolerance, and the supply chain team needs to look at current stock to see if it will cost too much in scrap to make the switch. This can seem overwhelming, so how might you simplify the process?
The key is to limit decision-making to the smallest possible pool. Early in the product lifecycle, while designs are still in the proof-of-concept stage, you’ll be able to make engineering changes with only the design engineering team involved. When plans go to a manufacturer, you’ll need their sign-off and that of the manufacturing engineers. Once you near full-scale manufacturing, a supply-chain manager will need to look at existing stock. The pool widens with product development, and changes to mature products require a signature from each department.
If the change isn’t your idea, though, how do you evaluate the engineering change request (ECR) and decide whether or not to sign off?
Evaluating ECO Costs
Like all business, ECRs turning into ECOs is a matter of costs and benefits: How much does the change cost, and is the planned benefit greater?
As design engineers, it’s easy to be myopic: A change costs the number of hours to update the drawing. But that’s only in our department. You may be facing tooling changes, process changes (like assembly steps and quality control checks, and their documentation), and scrap costs.
With tooling changes, the process is occasionally easy (for instance, in 3D printing, you only need update the model to make a change), but typically significant. For injection molded or cast parts, you’ll need to examine current tools to see if the changes are “mold safe”; that is, do the changes fit with the current tool venting and gating, and can the changes be made by cutting into the current tool (adding material to the cast/molded part)? If not, then you may have to undergo the expensive process of creating an entirely new mold or cast. Even for CNC-milled parts, a small ECO may require extensive reprogramming or create excessive raw material scrap.
In changes to assemblies, look at the assembly process: If the ECR is approved, can the assembly process be updated to match? Will assembly instructions need to be changed? What about listed bills of materials (BOMs) for both assembly and the supply chain? And when the parts are received, what quality control procedures are in place which will need to be updated with the change? At the very least, the first shipment will need to be inspected more closely to ensure that the change has been made correctly. The costs for each of these process updates will need to be included when weighing the balances.
Finally, while some changes are critically urgent, with most changes, it’s best to examine the timing for the change. Maybe you’ve just done a molding run and produced 10,000 parts with the old geometry—if possible, use those parts before changing to the new geometry. If the change can’t wait, look at re-working the parts with some minor machining to save money, instead of scrapping the current stock.
Once you and your team have weighed the cost of an ECR and decided it’s worth the cost, it’s time to make a formal ECO and work on communicating the change to all involved.
Clearly Communicating ECOs
If there’s one takeaway, let it be this: Make sure that your engineering changes are crystal clear. Even worse than a poorly thought out change is a poorly communicated one—at least in the first instance, someone thought the change was a good idea.
The first step in this communication usually involves the ECR and a “redline”—a drawing or document that clearly shows the change to be made, by marking up the current revision with literal redlines and red text on the PDF. The old design should still be visible, along with the change to be made, and in the revision block, it should note that this is “rev X – redline 01”. For every redline adaptation, the last number should be incremented, and only one department at a time should have change rights—avoid parallel changes.
As the ECR is being approved by different departments, all the documentation for parts and processes that will require updating should be noted, as well as all the parties who will need to be notified of the change. When the ECO is implemented, be sure to update all the drawings and part files with a new revision letter or number and archive all the old files. Send the new files to all the involved parties, along with a schedule of rollout for implementation, the signed-by-all-parties ECR, and a summary of the changes made in an attached Engineering Change Note (ECN). When in doubt, err on the side of over-explaining.
Streamlining the ECO process
With modern files being stored electronically, the change management (CM) process is often also handled in the digital world. Redline drawings are handled through marked-up PDFs, and to ensure that revision changes and archival of older versions is handled smoothly, most larger companies use PLM (product lifecycle management) software, such as Matrix or Oracle, which automatically store ECRs, ECOs, and revisions in one central database; and also electronically pass around files for digital signatures.
Smaller companies and startups may opt out of CM software but should still have an EC process as part of their PLM plan, typically the responsibility of the reliability engineering or sustaining engineering team. The process should specify which departments are required to sign off on changes and schedule monthly review meetings for all non-urgent changes. Customize the process to your company’s needs, but keep the general framework intact.
Implementing an EC Process in Your Company
Maybe the change you want to make is going to save the company millions and make you a hero, or maybe it will cost thousands in scrap, but in either case, it’s great to have a team of level-headed people taking a look at potential consequences. If you’re hoping to create a new change management process for your company, try implementing an EC process to get your team started. And don’t forget to sign up here for future articles on hardware to find out what changes are best for your product!