Is PLM the New PDM?

By Martyn Day, editor, CADserver, January 7, 2004
 

See Also

·  Dassault Systèmes website  ·  PLM features - at CADdigest.com  ·  EDM/PDM Products & Companies list - at TenLinks.com

The marketing frenzy around badging everything “PLM” (Product Lifecycle Management) in the engineering market continues unabated - even though many financial analysts, journalists and most importantly, users, have still to work out exactly what PLM means. While companies like Dassault, IBM, PTC, SAP and UGS PLM Solutions have lead the PLM marketing charge, many other software developers have leapt onboard the PLM bandwagon to promote their products or services.

There's a quote saying, “An industry gets the analysts it deserves” and the CAD market has a collection of industry watchers that has been all too happy to join in the creation of PLM hype for their customers -- namely, the software vendors; between them they have managed to produce many millions of jargon-driven sentences that mean nothing to anyone. These documents are usually accompanied by all the graphs and charts you could possibly ask for, stating (or should that be predicting?) that the PLM market is huge, is growing, and is led by one of their clients.

It became pretty obvious that a lot of this marketing and the keen interest of the PLM coat-tailers, derived from the fact that the PLM messaging was gaining some traction with the financial analysts. Press releases appeared to be more geared for analysts than for journalists or users; I guess in the hope that promises of big PLM revenues (through the combination of a number of previously identified revenue streams, CAD, EDM (Electronic Document Management), PDM (Product Data Management), etc., would increase the share price.

Being at the receiving end of this PLM marketing tsunami has been a rather unpleasant place to be. It's annoying to have to write articles that attempt to try to decipher and explain someone's obtuse take on marketing, as opposed to tackling the real issues, but that's what CAD journalism has descended into -- a translation service!

Our company hosted a PLM event a month ago and I went along to hear the opinions of the attendees, as well as the speakers, who were all fairly impartial, with some brutally honest end-user case studies. During the day there appeared to be a running theme: PDM (Product Data Management). Nearly all the presentations were about deploying PDM systems and the trials and tribulations that go with that task. There was talk of cultural problems, technical issues, problems with vendors, lack of support and inability to connect various systems. This is not new. In the past PDM managed to acquire a bit of a bad reputation for itself, being expensive and complicated to deploy. Installations never ran as scheduled or budgeted and in many cases the systems had been oversold to the customer, leading to an expectation / delivery mismatch.

That PLM was the new PDM seemed to be the underlying message from some of the other vendors hyping PLM. Autodesk, for instance, told me that PLM is essentially the new term for PDM and for awhile the company happily used the PLM messaging, even though it didn't have a PDM solution, just a Web-based document distribution system, or portal.

Reading the press releases from the CAD vendors muddied the water even further. In many cases upgrading a CAD system between revisions, or changing from a competitor’s CAD system to theirs, was heralded as a “PLM win” or “showing a commitment to PLM” by merely buying more seats of CAD. The same hype "win" hype had been announced back when those customers had purchased or upgraded their PDM systems -- SMARTEAM, TeamCenter, and Windchill.

Perhaps compounding hype, IBM recently launched the PLM Express bundle, which appeared to comprise CATIA P1 seats (CAD), SMARTEAM (PDM), and some IBM middleware. Surely this is PDM Express? The mixed messaging around PLM has not helped anybody outside of these companies to fully grasp what is meant by the term. It appears that most software developers who are playing the PLM game match their definition of PLM to whatever services and products they currently supply, which is a bit limiting if PLM is allegedly all-encompassing. Surely there must be more to PLM than this?

I contacted the main PLM protagonist, Dassault Systèmes, and asked Pascal Daloz, Vice President R&D Strategy: "Is PLM just PDM?"

"PLM is really what we call the relational design," Daloz responded. "The fact that each time you make one modification, you are able to propagate the change along the lifecycle of the product and to measure the real impact of the change you are doing. If you are just approaching the problem with a simple PDM system, the only thing you will be able to do is to send a message to others that you are making a modification and they have to take into account the modification you have done. PDM is not sufficient for PLM. PLM equals Information System rather than a suite of applications.

"Let me give you an analogy. How do you define ERP (Enterprise Resource Planning)? ERP is only a collection of applications that deal with the financials, the human resources and so on. It's an information system. You have a back end where you store all the data and can have access to this to generate whatever you need to make decisions and manage your company. PLM, for me, is the same.

"If you approach the problem with a pure PDM positioning you will never be able to capture all the semantics of the application and you will never be able to implement a relational design scenario. You will capture the geometry, the topology, the product structure -- but what about the tolerances or the dimensioning or all the application data? We build our model on the PPR (Product, Process and Resource) model and it delivers significant levels of abstraction to manage all the data between the key components. With a pure PDM approach you will never achieve that because you are managing files and documents and you cannot manage more information than you have in your document."

I asked if Dassault recognized the confusion with the PLM marketing messages, where purchases of individual products become touted as “PLM commitment.”

"For me PLM is related to V5," replied Daloz. "PLM is not related to CATIA, ENOVIA or SMARTEAM because PLM is about business transformation. If you want to transform your business, you need to improve the process validation and the lifecycle management and for that you need a common architecture and V5 is really the foundation to support PLM. We used to have CATIA in V4, without the PLM acronyms, so that's why I agree in some way with what you are saying. It's not because someone is buying CATIA that they are doing a complete PLM transformation. Each time I meet a V4 customer who wants to migrate to V5 and use it like V4, I tell them that they are wasting their time. If you want to migrate to CATIA V5, you have to keep in mind that in some way you will want to re-engineer your processes and the missing piece today. If I compare PLM with the ERP business, and especially SAP, they were able to create a complete ecosystem or system integrators to carry out the business process re-engineering inside large companies. And this is something we have to put in place to have a similar approach.

"The main difference between PLM and ERP is the complexity of data that you are managing. ERP is just numbers, it's what we call flat data. In our case we are not managing numbers, we are managing smart objects. By that I mean objects that really have the capability to capture the semantics of the application. When you design a door, for a mechanical guy, this means a structure. For a designer this means a shape, for a software guy this means a control piece. But at the end you are talking about the same product. It's just that tthe way you express it is different because it's a different discipline with different semantics. So each time the mechanical designer makes a modification you need to understand what will be the implication for the others, which means you need to create the interrelation between those semantics. This is really why we believe in our architecture. And with V5 we have put in place the enablers to manage those smart objects. We are not managing and not storing numbers, they are smart objects."

I asked if Dassault's recently launched industry-specific PLM Solutions was intended to address the specific semantics to every vertical market.

Daloz commented, "In CATIA today we have over 10 million lines of code developed in C++. Boeing, over the last two years, has developed more than that, just to “verticalize” what we are doing for their specific needs. Sometimes people consider that we (Dassault) are not “open” but it's completely wrong and this is a really good example. Can you imagine how Boeing could develop 10 million lines of code on top of what we are doing without the right level of “openness"?

Openness is a term that's been overused in our industry. I said to Daloz: "There are two types of openness, between you and your customer, and between you and your competitor. In addition there appears to be two ways of doing PLM -- using a customer's existing systems (PDM, MRP, etc.) and building connectors between them, or buying a complete solution from one vendor. Dassault appears to favor the monolithic approach."

Daloz replied by explaining a little of Dassault's hub technology. "Today we have two engineering hubs. We have the engineering hub and the manufacturing hub (as well as the enterprise hub). For the level of integration that you are discussing we are doing this at the enterprise level, not at the engineering level. The reason is because we are managing the complexity of the data in the engineering hub and the manufacturing hub.

"If you want to draw on the maximum benefit of V5 you need to implement at least the V5 back-end, V5 infrastructure. It does not mean that you cannot have a PDM system with V5. For example, this is the situation that we have in our largest accounts. Sometimes they are using another PDM system but they are also using CATIA, plus ENOVIA. If you want to manage the complexities you have to have ENOVIA."

Thinking about managing and creating data, there appears to be two ways of deploying a PLM mindset. I commented to Daloz, "It seems to me that UGS, PTC and Dassault have a similar outlook on PLM, approaching it with a data-centric methodology. SAP appears to come from the 'knowledge of the process' side of the business. So why should PLM be tackled from a data-centric perspective?"

"One is data-centric, the other is workflow-centric," replied Daloz. "We believe we are mastering the virtual world and our motto is to be able to define, control and monitor the physical world. It means that the key technology we have is the modeling technology and capability. On the ERP/SAP side, it is totally different; they are optimizing the physical world, the real assets, and all their capabilities are related to optimization. For me, this is the real difference. We are providing the modeling capability and the reason is because in the discreet manufacturing industry, 80% of the cost is engaged during the design phase. So without a modeling capability, how could you justify some time to take this decision if you are not able to realize the complete impact. This is the main difference between what we are providing and what SAP is providing. With SAP you can optimize what you have, but you will never be able to optimize what you don't have!

"I believe that there is room for those two approaches inside a big company. On the one hand, the key difference is that you have the data- centric companies that are directly related to the modeling capability and a requirement to manage complex data, and on the other hand you have optimization and you need to put in place very straightforward workflow."

I have concerns over the applicability of PLM to small- to medium-sized firms and asked Daloz, "Is PLM just for large enterprises? I don't see the PLM need of smaller companies, unless they are in a large supply chain. Many are still wrestling with PDM."

"I agree," answered Daloz, " but I meet more and more companies that want to put a data re-use strategy in place. And if you want to do this you need to have knowledgeware capability at least in the authoring tool and inside the PDM -- and today this is one of the key fundamentals we have put in place with V5. To design the door of the new 7E7 took one guy 32 minutes. To do the same thing in V4 for the 777 took 20 people three weeks! This is really the value to have “knowledgeware” capability inside. In this case it's not only productivity gains; by reducing the numbers of people, we can also use a lot of knowledge that was pre-developed and it’s really at this point that PLM is becoming more and more important for the mid-sized market. That's why there's the new PLM Express offering and why we will be able to penetrate this market more deeply."

IBM’s PLM Express concerned me when I read about its launch. I asked, "It's called PLM Express but isn't it just CATIA and SMARTEAM? Is that PLM?"

"SMARTEAM has the capability to capture the knowledge that CATIA can create," Daloz replied. "The complete integration of CATIA and SMARTEAM took us two and a half years in development and we had the development teams for both products in-house. It's what we call 'entry PLM'. PLM Express is a brand managed by IBM and it includes a lot of 'websphere' functionality and it's a bundle. But I agree, we have to be careful to differentiate what is PLM and what is not PLM. Inside Dassault Systèmes we have a clear segmentation between process-centric and design-centric, so that splits SolidWorks and CATIA, for example.

"We are ten years behind ERP in terms of maturity. I am not referring to the market size, more in terms of industry maturity. Today the scope of PLM is not very well defined. For example, the CRM (Customer Relationship Management) capability is getting more and more integrated into the PLM environment; the reason is because this industry is restructuring. It used to be a CAD/CAM/PDM/CAE market. Now it's the PLM market. What is the difference? There is a tight integration, and also the scope is not exactly the same and the value proposal is not the same. Not at all.

"In the software business, we are all sharing the same vision. What's different is how you implement the vision. All the players in this field are talking about PLM but with very, very different implementation. For us PLM means business transformation."

My final question related to a specific technology that was delivered in CATIA V5. Morphing appeared to offer the “fantasy” of being able to design one product and morph it into the next generation using rules and templates. At the same time, it should update all the manufacturing jigs, moulds, etc., drastically reducing cycle times. I asked if Daloz if Dassault's morphing technology was just a “wizzy” new feature or a PLM differentiator?

He replied, "Right now people are deploying morphing technology. It's a process transformation. The most advanced morphing company in the world is in Japan where it is applying the technology along the lifecycle of its product, not only during the initial design phase. This company not only uses morphing to remodel the vehicles it creates, but it also remodels the complete production system and process. This company is aiming to create a division specifically to prepare the data for morphing, rules and templates for new data sets and legacy data!”

The Dassault view is that the accepted vertical engineering areas are merging together and getting more tightly integrated -- CAD, CAM, PDM, ERP and CAE. It's a model-centric approach and, for Dassault, the creation tools capabilities are paramount. It's also essential to have back-office systems that can handle the complex data and interface with other engineering systems. The combination of intelligent systems allows users to automate many tasks through change propagation methodologies, which have been proven by companies like Boeing, BMW and others. However, it has to be noted that there is a lot of work required to set up these systems and prepare the data. There was also some recognition that much of the marketing is not helping clarify a widespread understanding of the term PLM.

Using the Dassault definition, buying, deploying, owning or changing a PDM system is most certainly not PLM and I would have to agree. Although there does appear to be different levels of deploying PLM, IBM's PLM Express is perhaps closer to traditional PDM than it is to Dassault Systèmes’ morphing definition of PLM.

It also appears that some form of PLM can be achieved by connecting the existing systems within an engineering enterprise to share flat data, bills of materials, MRP and CRM. Again, this automates many engineering processes and can be achieved with any 2D or 3D CAD system, and perhaps this is where PLM has a future for smaller companies. Meanwhile, I can see companies like UGS PLM Solutions and Dassault Systèmes optimizing the larger manufacturing companies (automotive, aerospace, manufacturing) with more radical technology -- and perhaps some amazing results.

About the Author

Martyn Day is group editor of MCAD Magazine and AEC Magazine. For more information, visit the CADserver website.

More PLM Select Articles