Big-Bang PLM Dreams – should we move away from monolith approach?

For years, PLM and not only PLM, but also other enterprise systems like ERP supported idea of a single backbone system that can solve all problems. Enterprise PLM platform, in general, follow this approach by proposing platforms and systems focusing on supporting concept of single point of truth, common view and unified modeling. This approach clearly stated in all last major releases of PLM systems coming from both CAD-based and ERP-based origins.

So, discussion I want to have is quite opportunistic. I want simply put – is it really good and beneficial to follow “a centralized approach” and believe a single system will solve all problems business have around Product Lifecycle Management? Let me explain what I have in my mind by figuring out few important trends:

1. Modern enterprises have tendencies to agile organization and outsourcing for product development and manufacturing.
2. Single vendor platform will be “a functional compromise” in terms of specific niche application and services.
3. Establishment of a single system is a source of very significant investment and business organization.

In addition to these organizational trends, there are few important technological trends I want to mention.

1. Development of OnDemand services and applications
2. Usage of Internet as a platform for social and organizational communication

All these trends together bring me to the point to ask – what will be next platform for PLM? How this platform will be able to adopt to fast changed word of business and communication? How future enterprise and PLM platforms will be able to manage “change” concept and become transportable for changed organizational and technological landscape?

I don’t think I have all answers to state how change will happen, but I see PLM system  different from monoliths applications and platforms we’ve built today.

What do you think? How do you see future agile enterprise PLM platforms and applications?

Best, Oleg

Share

Share This Post

  • V6?

  • Chris, Almost expected question :)… thanks! Every PLM and ERP today is providing approach focusing on single big enterprise foundation. SAP, Oracle, TeamCenter, ENOVIA, Windchill. Talking specifically about V6, I do see huge value in building a significantly improved “collaboration” on design level as a result of CATIA back up by ENOVIA Application Server (you can read in many comments – CATIA on top of database). This is an architecture change that allows much more compared to what many other systems today can do. However, with regards to ‘big-bang approach”, I think, with obvious benefits, it brings huge cost of implementation and even more important huge cost element. So, future need to be different….
    Regards, Oleg

  • My understanding and maybe it is wrong is V6 is BIGGER BANG than anything else in the market. I remeber presentations that talk about “One BIG Server” that everyone will use… Certainly PLM has been all about BIG BANG from an implementation point of view, but I do not think the use of PLM has been very BIG BANG. I would say use of PLM has been much less monolithic.

  • Chris, the presentation you probably mentioned, are about V6 environment (server) used to control all PLM apps (for design, engineering, manufacturing etc.) Until V6, these apps were separate. So, such unification, is positive to reduce complexity. But other side (I mentioned it in my post), granularity, flexibility, SOA should be much stronger compared to what we have today. This is what I called “go away from monolith approach”. Best, Oleg

  • Andy Finkbeiner

    Oleg,

    I’ve come to the conclusion quite recently that the existing monolithic PLM solutions can be replaced with Microsoft based solutions. (Aras Innovator comes to mind) I’m sure that hundreds of sales people will weep if that is proven true but based on the results we’re having in our environment I’d say they need to start buying tissues.

    What is so great about paying millions of dollars to have your data held hostage by your PLM vendor? How do you explain to your end users that they need to change their process in order to fit inside of the PLM tool? If your engineering team needs 6 cost fields but your PLM vendor thinks you only need 4 cost fields then what are you supposed to do?

    Why should we pay millions for the privelege of moving all of our data into a central location when we could rather connect it together using SOA based tools. The fact that some of these SOA based tools (think Aras Innovator) are free changes the game even more doesn’t it?

  • Andy,
    Thanks for your comments! I think you exaggerate when speak about 6 vs. 4 cost fields. I think most of today’s PLM tools are flexible enough to solve this issue. However, I agree, PLM is not coming to the level flexibility Excel provides. So, how do you see MS based PLM solution today? Is it something just developed using MS technologies or product based on MS Office/SharePoint product family?
    With regards to your Aras related question. I indeed see Aras is feeling a very interesting domain of OSS (Open Source Software) and I had chance to discuss it in the past on plmtwine. With all respect to Aras Innovator, I have to say, TCO of this solution can on the same level as some of other license-based PLM products. Of course, zero price entry level is very interesting as an approach in PLM.

    I’m looking forward your future comments.
    Best, Oleg.

  • Andy Finkbeiner

    Yes, you are correct that is an exaggeration with the cost field example. But I don’t think you would have to think too hard to come up with lots of examples where existing monolithic solutions are too rigid to meet business needs. I’ve sat in many meetings where the system folks tell engineers “sorry, we can’t modify the system to allow it to do that for you”. At some point you start to wonder who is in charge, the business or the software vendor.

    We’ve seen many business applications (PLM included) where the user interface is dreadful. But changing the user interface is very difficult if not impossible as so many elements of the code need to be reworked.

    Other examples would be any situation where you want a change to the underlying schema. Such a change is very difficult with a monolithic PLM tool but quite easy with a tool such as Innovator.

    I’m not so sure that the next generation of PLM has to be MS based, but I do see Innovator as an excellent example of how to do it. Basically a PLM tool kit with pre-built modules that can be quickly configured to meet business requirements. So if you have a workflow module, a security module, a lifecycle module, etc that a user can assemble as needed then you have a very flexible system that can quickly respond to changing business needs.

    What do you think of this pre-built module concept for the next generation of PLM as opposed to the current model of large amounts of compiled code?

  • Andy, pre-build modules and full flexibility is an everybody’s dream. SmarTeam and MatrixOne are very close to this target, but for different period of time were either closer to toolkit approach or modules approach. The latest enterprise PLM (but not only PLM) buzz is to provide ready to use apps. I can agree, for today’s software, this is the only reasonable way to down implementation cost and TCO. But such OOTB approaches are very dangerous when you want really customize some core behaviors of the system (you mentioned it in your notes about user interface). So, back to my main point of this post – we are in the end of big-bang… I believe TC Unified and V6 will be latest monolith apps in enterprise PLM. Time will show what will be next… Regards, Oleg.