Medical Device Link.

engineering insight

New medical devices that offer patients a host of advanced features are constantly being developed. Manufacturing these products in large numbers is a complex job in and of itself, but just getting to the point where you are ready to manufacture is a challenge all its own. This article offers a number of suggestions on how to achieve that goal while meeting time and budget commitments.

Starting Out

Once the investment decision has been made, the first step often involves documenting (or receiving) production requirements as well as establishing time and budget commitments. This type of formal process may lead to the creation of a “charter” for the project team. Major milestones such as design freeze, mould tool qualification, clinical trials, and product launch need to be defined within the overall timeframe. It is also useful to reflect on the appropriate production strategy for your product and project. What are your priorities: Time to market? The flexibility to make changes or add product variants? Overall cost? It helps to understand these goals and to accept that they will affect important decisions as the project moves forward.

Engineering

Even if the design of your new device has been defined, there is still considerable engineering work that needs to be done. The steps outlined below should be followed in the order indicated.

• Design for Manufacturing (DFM) involves evaluating each component to ensure it has been optimised in terms of tolerances (too tight or too loose) and tolerance stack up, material choices, draft angles for mould tools and so forth.

• Design for Feeding (DFF) evaluates the suitability of each component for automated feeding and handling. Questions to consider include: Will the components lock together during vibratory bowl feeding, making them impossible to handle? Are there critical features that might break off? Are there surfaces where small scratches are unacceptable? Where do I have to use a palletiser and where can bulk handling do the job? Feeding trials are often the best way to reduce risk and eliminate surprises down the road.

• Design for Assembly (DFA) is a step-by-step evaluation of the assembly process. This involves looking at component tolerances and tolerance stack up, reference and locating surfaces, chamfers and lead-ins and other factors to ensure that the design is assembly friendly.

• Failure Mode and Effect Analysis (FMEA) is a technique whereby each possible way that things might go wrong during the assembly process are identified and analysed in terms of the risk of occurrence and the impact it would have if it were to happen. This technique also identifies the steps in the manufacturing or assembly process with the greatest amount of risk. A follow-on question is whether or not an error can be detected by the assembly system. A well-designed product allows in-process mistakes to be identified and eliminated. Risks and impacts are weighted, and a numerical analysis is produced. Once all of the possible errors have been tabulated, the next step is to take action to reduce high-level risks.

Each of these engineering-driven tasks provides feedback into the design of the device, thus helping to ensure reliability.

Process – The Heart of the Matter

A process can be considered as any step that physically transforms a component. Examples might be UV curing, ultrasonic welding, heat staking and printing. Each process contains variables (or parameters) that need to be defined such as temperature, force, frequency or time. Process understanding is often the key to high volume manufacturing. For example, a heat stake head that’s too cold doesn’t make a good product, but one that’s too hot will damage the plastic. A critical process might even be handled in closed-loop mode with the assembly system automatically monitoring key parameters to ensure that they remain within an acceptable range.

Key processes and their parameters need to be documented early in a project, and this is usually done through process validation--a systematic evaluation of the key process parameters on the device itself.

For example, take the process of bonding two plastic parts together by means of ultrasonic welding. A bond will have a required minimum strength, but key parameters such as the ultrasonic weld frequency, pressure and time influence the bond strength that is obtained. Measuring the impact of changes in these parameters on product quality can require making thousands of devices under controlled conditions, varying only one parameter at a time. Only a certain range of settings will create an acceptable weld all of the time, while others will result in unacceptable bonds on part or all of the devices. Documenting and analysing these results produces the definition of a process window--the range of settings that allows high quality bonds on the product--and provides important data in terms of the sensitivity of the process to changes in key parameters.

Proof of Principle Tooling and Trials

Certain assembly operations–process-related or not–might benefit from a Proof of Principle (PoP) trial. Typically these trials take a manufacturing step that has been flagged as particularly risky (probably in the FMEA), and create a working model of that operation. The goal is to try to simulate the results that will occur in the manufacturing environment. If you can’t get it done reliably using PoP tooling, then you probably have a manufacturing nightmare in the making.

The key to good PoP work is to simulate the conditions of high-volume manufacturing. Ideally, you will run your trials with components that are representative of the tolerance ranges that will be seen under high-volume conditions. Try to use the same technology for the trials as you would on high-volume equipment: cams, servo or stepper motors, automated feeding systems. Trials cost money, but when they are done correctly, the investment can be partially or fully recovered when you make the leap to pilot or high-speed production.

Pilot Production

Pilot production is a critical intermediate step to high-speed production. Typically you will be making many thousands of parts, but not yet millions. Pilot production should allow you to automate your most critical processes. Doing so in an intelligent manner allows you to fully re-use expensive tooling in the future. Validating a new device usually requires clinical trials of the product. Pilot production is the way to get there, but you will want to ensure that your key production processes remain exactly the same as you transition to the high-volume line. Changing a key process means re-running clinical trials, which takes time and money.

Configuring your pilot-scale production equipment with the technologies that you will require for the high volume line increases the up-front investment, but it reduces costs and saves time on the programme as a whole. Using pilot-scale equipment to validate key processes is also a great way to remove risk from the programme.

A Need for Speed

High-speed production is the ultimate target. The first step usually involves outlining the user requirement specifications, incorporating all the device requirements and its in-process inspection. You will need to document how components are to be handled and how your device will be off-loaded from the assembly system.

Giving consideration to the production environment is key: Does it need to be a cleanroom or will a grey room do? What skill sets will your operators need? Do you have the right people on hand?

Another key consideration that is often neglected is data management. Producers need to evaluate their needs in terms of the information they want the line to gather and manage, and how that information needs to be transferred and stored. Overspecifying data requirements increases both the initial and operating costs of a line. Know what you must have and what is nice to have but not necessary.

Conclusion

Getting from a great device concept to manufacturing millions of devices successfully is an engineering-intensive activity that is filled with challenges and dangers. Developing a strategy and an overall game plan that integrates risk analysis and well-designed pilot-scale equipment can help ensure the path taken is both fast and cost efficient.

About the Author

Erik Poulsen is Sales and Marketing Director for Komax Systems LCF SA. A mechanical engineer, he has spent more than 20 years in the automatic assembly industry occupying positions from designer to project manager. Komax provides a full range of automated production solutions to the medical device industry. For more information, contact Poulsen by e-mail: erik.poulsen@komaxgroup.com.