Regulatory Open Forum

 View Only
Expand all | Collapse all

design inputs and traceability matrix

  • 1.  design inputs and traceability matrix

    This message was posted by a user wishing to remain anonymous
    Posted 01-Apr-2019 15:03
    This message was posted by a user wishing to remain anonymous

    Hi everyone, 

    For a system with both hardware and software components, how can the design inputs for both hardware and software be documented in the input traceability matrix? We have always coded software input requirements starting with SRS, but not sure how the hardware inputs can be coded in a coherent way in the design traceability matrix.  Does anyone have suggestions? 

    Thank you.


  • 2.  RE: design inputs and traceability matrix

    Posted 01-Apr-2019 15:57
    ​Remember that design inputs (when properly devised) consist of functional, performance, and interfacial requirements that may spread across more than twenty different categories (i.e., safety, sterility, biocompatibility, mechanical, manufacturing, packaging, labeling, etc.).  This is germane because ultimately, each design input (of which there could be hundreds) might demand both a hardware and a software solution, or just a hardware solution, or just a software solution.  Moreover, each of those must be verified and, as applicable, validated.  Traceability between each input and the resulting outputs (whether hardware or software), verification, and validation must be maintained and should be accomplished using a trace matrix (or multiple trace matrices if needed).

    Consider first organizing and laying out the design inputs in a master trace matrix such that the aforementioned fundamental input categories (whether ultimately leading to software or hardware) are properly distinguished and maintained. This step is critical for assuring that the resulting design outputs (whether software or hardware) will ultimately cover the full scope of the inputs.  Then, for inputs ultimately leading to a software-based solution, consider adding a segue to corresponding SRS elements, thereby transferring their traceability to a separate SRS trace matrix used solely for traceability of software solutions and their corresponding verification and validation.  The same type of segued approach could be used for the hardware solutions, or the hardware design traceability could just remain in the master matrix.

     

    Hope this helps,



    ------------------------------
    Kevin Randall, ASQ CQA, RAC (U.S., Canada, Europe)
    Principal Consultant
    ComplianceAcuity, Inc.
    Golden CO
    United States
    www.complianceacuity.com
    © Copyright 2019 by ComplianceAcuity, Inc. All rights reserved.
    ------------------------------



  • 3.  RE: design inputs and traceability matrix

    Posted 02-Apr-2019 08:24
    One thing that I see missing in this discussion and in the design output discussion response is user needs. The traceability starts with identified user needs which lead to design inputs. The "user" is not only the clinical user but also others such as regulators, services, installers, certified, and any others that affect the design. User needs drive design inputs, which drive design outputs. Design Validation is done back to user needs, so it is important that the user needs be documented as the beginning of the development. Then the Design Validation steps are documented as part of traceability to close the complete design-development of the product, documenting that all user needs are met by the various design inputs and corresponding design outputs that complete the design requirements. 


    ------------------------------
    Edwin Bills MEd, CQA, RAC, BSc, CQE, ASQ
    Principal Consultant
    Overland Park KS
    United States
    elb@edwinbillsconsultant.com
    ------------------------------



  • 4.  RE: design inputs and traceability matrix

    Posted 02-Apr-2019 15:26
    ​I always take note of, and am thankful for, Edwin Bills' commentaries. He is one of our industry's foremost experts and stewards of proper quality system and regulatory solutions.  Case in point:  Consideration of user needs (and other elements) when deriving design inputs is certainly critical to the integrity of design inputs and the overall design control process.  To facilitate cataloging and searchability in the Forum library, I'm wondering if it might be best for us to discuss that important topic in a new thread.  It could be worthy of an entire discussion in its own right and seems to have a different (but related) scope than the original question about using trace matrices for recording design inputs.  Forum thoughts?

    ------------------------------
    Kevin Randall, ASQ CQA, RAC (U.S., Canada, Europe)
    Principal Consultant
    ComplianceAcuity, Inc.
    Golden CO
    United States
    www.complianceacuity.com
    ------------------------------



  • 5.  RE: design inputs and traceability matrix

    Posted 03-Apr-2019 08:30
    Hello Kevin,

    Let me say that I think you've done a masterful job of describing the Trace Matrix concept to several recent discussion threads.  I was also thinking along Ed Bills' line that there was the User Need/validation element not really addressed, but also understand you were focusing on the question(s) at hand.

    Candidly I am a bit surprised at the volume of these similar questions given what I thought was fairly clear description of the Design & Development process (i.e. and relating it to "waterfall" which then tends to support the Trace Matrix use) in both ISO 13485 and 21 CFR 820, but apparently there may be a need for an overview of the whole thing.  This could be a good place for that overview, but I'd be a bit concerned that your fingers might wear out or Forum space be limited before finishing.  If you did it, I'm sure it would be instructive and helpful to those who may have residual questions.

    In any case, thank you for raising the level of understanding so far.

    ------------------------------
    Rem Siekmann BSE, MBA, ASQ CBA
    Senior Principal Engineer
    Bellaire MI
    United States
    ------------------------------



  • 6.  RE: design inputs and traceability matrix

    Posted 03-Apr-2019 15:07
    ​Many thanks Rem for the positive feedback.  Indeed, lots of very interesting questions are arising regarding design controls, and there's always value in reviewing fundamentals.  I seem to learn something new each time I do so myself.

    Based on your seconding Edwin Bills' comments about user needs, it sounds like a discussion about that is in order.  I'll put it on my list to start a corresponding thread (or others can of course do the same).

    All the best,

    ------------------------------
    Kevin Randall, ASQ CQA, RAC (U.S., Canada, Europe)
    Principal Consultant
    ComplianceAcuity, Inc.
    Golden CO
    United States
    www.complianceacuity.com
    ------------------------------



  • 7.  RE: design inputs and traceability matrix

    Posted 03-Apr-2019 16:11
    Thanks to you both, Kevin and Rem, for your kind comments.  I teach the recently revised AAMI  courses on Quality Systems, Design Controls, and Risk Management.  Whenever I see something where I think I can bring clarity or make a point that might help others, I chime in the discussion.  I have seen the "light come on" in some of our students eyes in this topic and thought it might be helpful to look at the complete Traceability Matrix for design, to expand upon the response to he question that came up.  We also use a similar Trace Matrix in Risk Management (which actually includes some of the information from the Design Trace Matrix) as required in ISO 14971, tracing from the Hazard, to Hazardous Situation, Risk Estimation, Risk Evaluation, Risk Control (Design Requirements and Design Output), Verification of Implementation of the Control (Part of the Design Verification), to the Verification of Effectiveness of the Control (Part of both Design Verification and Design Validation).  Someday, I hope to find a software package that effectively implements these requirements in a useful manner.

    I ma open in participating in this topic, so I will jump in when the time is right.  

    Thanks again, gentlemen.

    ------------------------------
    Edwin Bills MEd, CQA, RAC, BSc, CQE, ASQ
    Principal Consultant
    Overland Park KS
    United States
    elb@edwinbillsconsultant.com
    ------------------------------



  • 8.  RE: design inputs and traceability matrix

    Posted 04-Apr-2019 10:03
    Hi! I haven't necessarily seen all the related threads. Are you able to post an example trace matrix?

    Thank you.

    ------------------------------
    Corey Jaseph RAC
    Director of Regulatory Affairs
    South Jordan UT
    United States
    ------------------------------



  • 9.  RE: design inputs and traceability matrix

    Posted 04-Apr-2019 21:16
    ​​Corey, speaking for myself (can't speak for Edwin):  Sticky question... Since I have clients who've remitted payment for my assistance on that, and since this is something with a bit of customization for each client, I'd be remiss if I posted it here.  But if you contact me offline I'd be glad to discuss your question further.

    ------------------------------
    Kevin Randall, ASQ CQA, RAC (U.S., Canada, Europe)
    Principal Consultant
    ComplianceAcuity, Inc.
    Golden CO
    United States
    www.complianceacuity.com
    ------------------------------



  • 10.  RE: design inputs and traceability matrix

    Posted 04-Apr-2019 21:37
    Corey,

    I know Richard's situation, and understand his response.  I am working mostly in Product Risk Management and have published a risk traceability summary that I built from the GHTF guidance on medical device risk management at this link:  GHTF risk guidance 

    This is not the design traceability you are looking for, I know, and AAMI has one in their Design Controls course material, which I am not able to share.  The basic structure is an Excel spreadsheet, but it is more complex because as you can see in the GHTF reference, you cannot simply follow one row across the spreadsheet since, for example, one design requirement may lead to several design outputs.  Even more complicated is that two design requirements may lead to one design output.  

    I can refer you to a couple of other versions of the risk traceability summary in publications that I have done, but I think you can get the concept from the GHTF document and convert it to a Design traceability document.  My direct e-mail is on my signature block.

    ------------------------------
    Edwin Bills MEd, CQA, RAC, BSc, CQE, ASQ
    Principal Consultant
    Overland Park KS
    United States
    elb@edwinbillsconsultant.com
    ------------------------------



  • 11.  RE: design inputs and traceability matrix

    Posted 03-Apr-2019 16:37
    ​I'll second Edwin's mention about his instruction and the AAMI training courses.  If not the best of the best, then those courses are in my opinion amidst the best of the best.  (And I'm not currently an AAMI instructor, nor do I work for or with Edwin, so the Forum can rest assured that my testimonial is objective and impartial).  AAMI's materials and instructors are top notch based on my actual real-life experience being exposed to them working on compliance projects in the past.  If there is any down-side, some of the best instructors have retired, so I'm glad folks like Edwin are still involved.  Edwin, you're not allowed to retire; we need you too much. :)

    ------------------------------
    Kevin Randall, ASQ CQA, RAC (U.S., Canada, Europe)
    Principal Consultant
    ComplianceAcuity, Inc.
    Golden CO
    United States
    www.complianceacuity.com
    ------------------------------



  • 12.  RE: design inputs and traceability matrix

    This message was posted by a user wishing to remain anonymous
    Posted 05-Apr-2019 11:40
    This message was posted by a user wishing to remain anonymous

    Where can we access these courses? Are these courses online or in - person?  Thanks.


  • 13.  RE: design inputs and traceability matrix

    Posted 05-Apr-2019 12:06
    AAMI courses are in-person and are shown at AAMI Industry Courses  The courses have been developed with the cooperation of FDA and most in the DC area include an FDA instructor on-site  

    RAPS also has some excellent courses and I have been involved in the development and updating of the Medical Device Risk Management on-line course.

    ------------------------------
    Edwin Bills MEd, CQA, RAC, BSc, CQE, ASQ
    Principal Consultant
    Overland Park KS
    United States
    elb@edwinbillsconsultant.com
    ------------------------------



  • 14.  RE: design inputs and traceability matrix

    Posted 05-Apr-2019 11:56
    The AAMI course descriptions and registration can be found here:

    https://www.aami.org/professionaldevelopment/content.aspx?ItemNumber=2419&navItemNumber=2825


    ------------------------------
    Kevin Randall, ASQ CQA, RAC (U.S., Canada, Europe)
    Principal Consultant
    ComplianceAcuity, Inc.
    Golden CO
    United States
    www.complianceacuity.com
    ------------------------------