Bridging the Gap from Concept to Production, a Consultant’s Perspective

April 12th, 2017:

There are many articles and books discussing product development; however, an area that I personally find to be a challenge within a medical device company is navigating the design transfer process from concept development (R&D) to production (sustaining engineering).  Design transfer, or the introduction of the design to production, includes a number of stakeholders with varying priorities. Generally speaking, the business wants to move fast and release a product for as little money as possible; R&D wants to move fast and release a product that meets requirements; and  Quality/Regulatory wants to ensure the product conforms to the quality management system that is typically expressed under a 13485 certificate and CGMP, or 21 CFR 820.  Please understand that in this context, conforming to a process equates to ensuring the product is safe for human use, meets requirements, and conforms to FDA standards.

Why is design transfer so difficult? 

In part, it has a lot to do with ensuring the product specifications, design documentation, and testing have all been thoroughly assessed.  This assessment is an important stage gate to commercial release and gets a lot of attention in an organization.  Failure to perform an effective design transfer can lead to delays in launch, field failures, etc.  All of which can negatively affect the company. 

Preparing for design transfer also points out gaps in the product specifications and manufacturing documentation.  Typically, any units that have been built prior to this point have been built by an engineer or someone who is very familiar with the product and has learned the hard way how to build units that don’t fail.  Particular attention isn’t paid to XFMEA’s, suppliers, and other documentation, which is required for design transfer and likely hasn’t been completed yet.  Unfortunately, the product teams spend their time trying to build something that works, and not enough time on the necessary documentation surrounding the product.

So how do we make the design transfer more efficient?

The answer is really quite simple: Throw more bodies at the problem.  Obviously, I’m being facetious as this isn’t really an option for most companies.  If anything, you are likely going to slow the process down due to multiple hand offs and “over the wall” development activities.  To make a design transfer more efficient, you need to make the product team more efficient with their time.

First off, the company needs a quality system that allows for a level of control appropriate for the activity, at the time of the activity.  Part of the reason why things aren’t done when they should be is due to the amount of time and effort to get them completed.  Typically, it’s not the authoring of documents, it’s the review and approval that takes a lot of time, leading to delays.  Keep the review and approval to those with the technical knowledge.  Put others on distribution once the document is released.  If someone has an issue with the document, they can review their concern with the highest level approver of the document.  If the concern is valid, modify the document and release it again.

Second, be clear as to the requirements of a design transfer.  While it may be difficult to have a single checklist that would cover all transfers within an organization, it should still be understood what is required to be reviewed.  You should not have someone at the review withholding approval due to a missing document if it wasn’t understood that the document was required for the review.

Last, and for me the most important, leverage the engineering build process to create your design transfer documentation. 

What is an engineering build?

An engineering build is a low volume product run that is used for multiple purposes by the product development team.  Primarily, an engineering build is for verification purposes but it can also be used for validation, and other documents (work instructions as an example).  Typically, the product team will want to build units as quickly as possible and have them be fail-fast, all for as little money as possible.  The verification of the design inputs is of utmost priority to the product team.  To that end, product teams will focus all of their efforts on getting components, building assemblies, and testing product until they are confident they can meet product requirements.  To get components quickly, the team will leverage any manufacturing process to produce product that could be claimed as “production equivalent” as quickly as possible.  Unfortunately, by focusing on the verification of the product, other documentation required for design transfer will lag until the product team has more time, the design is more stable, or other resources free up to help out.  To make the design transfer efficient, the product team must use the engineering build process to complete the requirements for design transfer during the engineering build, not after.

We’ve already identified that the company must look at the document review and approval process as well as being clear on the design transfer requirements.  So where else can we look to put more time into the product team so they can focus on other activities?  The answer is Supply Chain.

A major source of time and money is the manufacture or purchasing of items the product teams need for the engineering build.  A typical quality management system will have controls over vender selection and order placement.  To streamline the vender selection, and the sourcing of products, the quality system should leverage risk-based assessments for the requirements of the item being built or purchased in order to determine an appropriate level of control.   If you’re buying a stainless steel 1/4-20 screw, you should be able to get it from any supplier with a credit card order as long as your risk assessment says it’s appropriate and the process is controlled.

A fast, flexible purchasing system will allow the product team to quickly and efficiently procure the components necessary to verify the design via verification and validation activity.  For most components in a medical device, unless there’s some proprietary method of manufacture, or the risk assessment dictates otherwise, allowing the product teams to operate quickly and efficiently within the boundaries of the quality management system will certainly help to eliminate frustration and delays within the product team.

Now that you’re capable of purchasing what you need from suppliers who can get you your product quickly, you need to agree on what you can use the product for.  Ideally, you’d want to use the parts for everything necessary for the design transfer.  To do that, you must have alignment between the product team and quality team on the term “production units.”  Typically, a quality system may equate production units to the highest level of control, thereby negating any time advantage the product team had during product development.  Generally, people consider production units the equivalent of what would be available for commercial sale.  This does not have to be the case.  If you have a family of products, you may test your worst case design (verification or validation), yet never make it available for commercial sale.  Unfortunately, quality teams tend to be very conservative, and have strict beliefs in what can and cannot be used for verification and validation.  From the FDA, verification does not require the use of production units.  Per 21 CFR 820.30 (f), verification activity stipulates that “identification of the design, method(s), the date, and the individual(s) performing the verification, shall be documented in the DHF.”  Again, per 21 CFR 820.30 (g), “Design validation shall be performed under defined operating conditions on initial production units, lots, or batches, or their equivalents.  The results of the design validation, including identification of the design, method(s), the date, and the individual(s) performing the validation, shall be documented in the DHF.”

Some companies will use a two-stage development process where you utilize a numerical revision to denote “R&D” and alpha revision for “production.”  Before the design can be transferred, the product team needs to go through a release process to change the revision from numeric to alpha.  While useful, I’d recommend that the switch to alpha occurs as part of the transfer record, which is to say that this should happen after the meeting.  Creating documentation to equate a numeric revision used for testing to an alpha revision is a huge waste of time.  As long as process controls are in place, and the product teams deliver as expected for design transfer, spending time converting revisions isn’t necessary prior to the design transfer.

Conclusion:

The proper assessment of the design during design transfer does not need to be a stressful, long, and drawn out process full of aggravation and frustration.  By building a Quality Management System that provides (1) an efficient review and approval process; (2) clear requirements for design transfer; (3) a flexible, risk-based approach to vendor selection and product purchasing; and (4) agreement on how to utilize a complaint supply chain process for the manufacture and purchase of product for verification and validation, an organization will have reduced a number of limitations from the product development team, allowing them to focus on completing all of their work for an efficient design transfer.

Thank you for reading.

Disclaimer: The opinions expressed in this article are mine, and mine alone.  They have been developed over years of observing and working with many medical device companies.  As they are my opinions, I welcome others thoughts on how to improve the development of medical products.  If anything, I’ve learned that there are many ways to implement a compliant system while developing medical products.  Some just happen to be better than others.

-Charlie Sears, Principal Mechanical Engineer

Posted in: Design Transfer