PDM. Pre-configured? Painless?

PDM. Pre-configured? Painless?

I want to get back to the topic of Out-of-the-Box PDM systems. The topic isn’t new. The following post by Jim Brown was a trigger for me to get back and discuss it. Jim posted: Can PDM Value Being Achieved Quickly? Painlessly? You can go to Jim’s blog and read about the research he did. In addition, you can navigate to the following link to upload the executive summary of the research for free. You can get the full report, but you need to be registered with Siemens PLM that provided their sponsorship to make this research done. Choose your download option and make your opinion, first.

I’ve been blogging about Out-of-the-box PDM and PLM before. If you haven’t had a chance to read my previous blogs on this topic, I can recommend you the following few links.

Out-of-the-box: Misleading of Focusing?
Why do I need to change my Out-of-the-box PLM?

In my original thinking, “out-of-the-box” option was created to simplify the initial introduction and implementation of PDM/PLM software. So, it was like a anesthesia you need to provide to a customer to help him to digest your application. Another concern that I was raising all the time with out-the-box option was about the ability of manufacturing companies to align with the environment. From my standpoint, engineering and R&D (majority of PDM implementations is going there) is the environment that hardly can be standardized. Joking about it, I’m saying “when you have two engineers, you have a potential to have three different opinions :)”.

How To Out-of-the-Box PDM function?

There are four major components of PDM system, in my view: Data Model / Vaulting, Revision, Identification, CAD / Integration. The question I’d like to ask if these functions can be predefined and used by any customer without modification? Let me put some analyzes about each of them.

Data Model / Vault. This function helps you to organize data. Usually, this is one of the most fundamental functions, and it defines the level of flexibility that implemented by the system. I’d be thinking about two aspects of data models in the context of PDM. One is related to CAD data. Another one is related to the management of Bill of Materials. The discussion about these two models and how they are related very often makes PDM implementation complicated. Vaulting function is straightforward and standard for many PDM systems these days.

Managing revisions comes next after vaulting. Companies have a temptation to customize the revision schema. However, the number of customization here is limited, in my view. On the other side, identification can be more complicated story. I posted before about document numbering, part numbering and identification before. These topics can be complicated and requires implementation skills and methodology.

The last one is simple and complex at the same time. You want PDM system to be embedded into CAD user interface. It started as a trend (10-15 years ago), and now it is considered as “must” requirement. However, this is still a challenge. The function normally worked well for PDM systems coming from the same CAD providers. At the same time, supporting “other CAD” systems can be problematic and require additional implementation skills.

PDM + CAD = ?

I can see a trend towards embedding PDM functions into a CAD system. First of this kind innovation came out of Dassault with the release of CATIA V6. PDM functions such as vaulting, revisions and some others are embedded. CATIA uses Enovia V6 to manage it. In parallel, I think other CAD vendors create successful bundles such as SolidWorks/EPDM, SolidEdge/TC Express, Pro-E/Windchill, etc. I wrote about this trend some time ago. Navigate your browser on the following link to read about it.

What is my conclusion? Engineers normally dislike PDM functions. They are trying to avoid it as much as possible. Therefore, PDM systems are not as popular when it comes to implementations. It requires time, cost and affect CAD functionality. However, the industry perception is that you need to have PDM to control your data. CAD vendors are trying to embed PDM functions into CAD packages and improve vertical integration between CAD and PDM packages. Can it be completely pre-configured and painless? I’m not sure. I think, the best PDM engineers can think about is the “invisible PDM”. Just my thoughts…

Best, Oleg


Share This Post

  • Brian Roepke

    I think you’re spot on Oleg. OOTB is something we have been doing and really push to strive for. However there are always a couple of things that happen when you deploy those things like Vaulting, Security, Revisions, Lifecycles, BOM, ECO, etc…

    The further you get down list the harder it is conceptually for company to fit their paper or Excel models into the electronic world. The Vaulting part is easy. This is typically a day of effort for people. Clean up and load up your data and train your users. No biggie (your mileage may vary depending on how disorganized you were to begin with 🙂

    Revision – Like you say; again… Pretty easy. “What’s your scheme? Okay… Let me make that in the Vault and we’re done…”

    Then is security and lifecycle. This starts to get into that process discussion. While we can configure just about anything under the sun there are always discussions about whether or not you should implement what you have or tweak it to perform better now that you have this new tool. This again isn’t tough but it’s harder.

    And BOM and ECO. Yikes. These things affect a lot of downstream systems so you have a ton of “choices” for the customer to make. How does this fit in? How do we integrate? What departments get to use the new system? etc… It’s not hard but in my opinion it’s a lot of questions and discussions to get there.

    I guess for me it comes down to (1) you need an out of the box system so people don’t start with a blank sheet of paper; that’s too open to figure out what you need to do (2) it needs to be SUPER easy to configure. Don’t make me write code or deal with esoteric configuration files in XML or some other ancient mess to configure (because I should want to change it over time as well!) and (3) Bite it off in chunks. Get small wins and share them with a lot of people. Don’t wait till you get to the end of the trip to share your first photos from day one…

  • beyondplm

    Brian, Thanks for your insight! I agree with your analyses. When vaulting and security are easy, moving towards BOM and changes is a lot of process discussions in companies. The two important characteristics I envision in future systems are “granularity” and “flexibility”. This is probably what you are calling “bite it off in chunks” :)… Good discussion. Thanks! Oleg

  • Jim Cooke

    I agree with your “original thinking” that it comes about as a means to simplify the initial introduction. Some are looking to optimize Time to Value. Others seek a low cost entry point. But on the adoption side I think the request is driven by fear – don’t make *me* less productive with too much complexity and too many choices and options.

    I think OOTB is a cry for “I recognize I need it, I want it quickly, cheaply, and non-intrusively”. Oh, and by the way I don’t want invest in something that cannot scale to the full richness of PLM as I mature and get comfortable with incremental benefits.

    I liked your “invisible PDM” conclusion, but I’m not sure this is really what they “want”. They want visible benefits, they just don’t want to have to do a lot of engineering and training to get there. I really like the way Brian’s summed up what they need to see and equate with OOTB.

    But today – I think the “Box” needs to come with a consultant, and people have a natural tendancy to want to be able to do it themselves. It’s sort of like the natural resistance to stop and ask for directions when you’re lost, or to read the manual. We want to be able to do it on our own, perhaps because it makes us feel smarter or more accomplished. As Brian noted, the process to get PDM actuallyy takes “a lot of questions and discussions to get there”, and this challenges us to know and identify what we want. Asking for help is not as comfortable as being able to do it yourself.

  • beyondplm

    Jim, thanks for your insight! I liked your comparison between “Box” and “consultant”. This is exactly what happens today. My “invisible PDM” is not a product I can show :). I believe somebody can come with this. However, never seen it before :)… Best, Oleg

  • I have customers that have seen / heard of wildly expensive implementation of PDM/PLM. They don’t want their careers caught in an implementation like that, yet they see the value of managing their data. Or more importantly, that their data isn’t managed and the problems that it causes. So, time to value is a big concern. Get it functional, get value and the next workflow, the next add-on will naturally surface.

  • beyondplm

    Brian, thanks for your comment! I think, the “people-factor” is important, and you touch an interesting point in PLM implementation practices. I always see “a person” behind any PLM implementation. This person (manager or even CxO) becomes your sponsor. The value needs to be clear, but what is mostly important, it should be aligned to the reasonable implementation timeline. Per-configured PLM concept is trying to achieve this goal, but the success is doubtful. Just my thoughts… best, Oleg

  • Pingback: Future CAD won’t require PDM()