Autodesk, Vault and Multi-CAD

February 21, 2012

I’d like to continue my "multi-CAD and PDM" story. If you had no chance to read my yesterday post The Anatomy of MultiCAD-PDM integrations, please do it before continue. I want you to take some initial context from there. I’ve been talking with Brain Roepke, director of PLM and "father of Autodesk Vault". My interest was to learn more about Autodesk Vault PDM in the context supported CAD systems, Open APIs and openness.

Autodesk, Multi-CAD and Integrations with other PDMs

According to Brian, Autodesk is investing a lot in multiCAD capabilities. It comes in various capabilities and different applications. I’d like to to strict my coverage purely to Autodesk Vault. So, this post will not cover multiple translators and ability to read different CAD formats by various Autodesk products and for variety of purposes – visualization, simulation and many others. Autodesk Development Networks (ADN) allows you to have the ability to develop applications using AutoCAD, Inventor, and other Autodesk products. In addition to that, you may use 3rd party tools that allow you to access Autodesk files and formats. I found integrations with Autodesk Inventor, and AutoCAD developed for Windchill, Enovia, TeamCenter, SAP PLM and others. Some of the integration developed by vendors and some of them, by partners.

Autodesk Vault- foundation and APIs

Let me start from core capabilities of Autodesk Vault to be integrated with other applications as well as providing a platform for development. Navigate your browser to the following link – Autodesk Dev Center. Here is what written about Autodesk Vault there:

Vault makes use of Web Service technology, which greatly simplifies client/server communication. Fundamentally, the Vault API allows a program to interact with the Vault server component, also known as ADMS. In fact, Vault Explorer and the CAD plug-ins are built on top of this API. So you have the same level of access to server data that Autodesk clients have.

Following this you can learn that Autodesk Vault SDK is installed automatically with Autodesk Vault. According to Autodesk you don’t have to be ADN (Autodesk Development Network) member to use APIs.

Another interesting functionality I learned about was Vault Attachments. Navigate to the following link to learn more. What was interesting is that Vault Attachments can be used in order to have Vault to use rules based on the association between files. The following passage, actually, explains this capability:

You can add attachments between any two or more files within the vault. Attaching a file to another in Autodesk Vault creates an association between the two files in the vault. Using attachments, you can check out all the files that are linked together and work on them as a unit, regardless of the applications that created the files. For example, you may want to attach a set of images to a Microsoft Word document if the images are linked to that file. In a design environment, you may want to assign a relationship from a document that represents assembly instructions to the actual CAD assembly files.

I can recommend you the following blog post – 6 hours of videos of Vault API, which contains quite good collections of videos you can use for training. It includes presentations and code samples by Doug Redmond.

Autodesk Vault and Other CAD systems

I’ve been searching for integrations of Autodesk Vault with major CAD systems of competitive CAD/PDM providers – SolidWorks, CATIA, Pro-E / Creo-Parametric, SolidEdge and NX. Here are few videos shows examples of these integrations.

Autodesk Vault and SolidWorks

Autodesk Vault and Pro-E

Autodesk Vault and Bentley Microstation

What is my conclusion? I found multiple ways to develop Vault PDM integration with CAD systems. Some of them requires API and development. Some of them can be done just by product configuration. I believe in such a thing as "integration", devil is in details. It sounds like Autodesk has an intent to be open. More APIs and Web Services approach supported by Autodesk Vault provides a good foundation for that. Just my thoughts… I’m looking forward to learn more from comments and discussion.

Best, Oleg

Disclosure: This post was reviewed by Brian Roepke before publishing.


PLM Integration Failures

January 13, 2011

There is one topic that always raises lots of controversy, in my view. I’m talking about integrations or even more specifically about PLM-oriented integration. I want to point on the following two articles I posted previously about PLM integrations:

PLM Integration Gotchas
PLM and Enterprise Integraton Game

I read Reasons Why PLM Integration Fails?" article on the To-Increase Blog. To-Increase is a company from Netherlands specialized in the Microsoft ERP products (Dynamics AX, NAV) and product configuration software e-Con. Read the article and make your opinion. The author is making point of various difficulties related to PLM integrations. Here is my favorite passage from this article:

A fundamental risk within any manufacturing firm, especially a firm with global operations, is the risk of information becoming siloed within individual teams. For example, if information concerning a flaw in the development of a product is available only to the engineering team, and kept from marketing, there exists the risk of gearing up product launch tasks too quickly – resulting in wheel spinning at best, and a significant loss in resources at worst.

In much the same way, if a PLM system is implemented – but not integrated with all other systems related to manufacturing processes (think Enterprise Resource Planning systems, think Manufacturing Execution Systems) the risk exists for information to be siloed in one system.

PLM Integration and Competition

The focus on PLM-ERP integration is interesting. These are two systems that very often are trying to establish a dominance in a culture of manufacturing organization. Are you PLM or ERP driven? What system "owns" Part or Bill of Material information? Who is authoring BOM? I heard such statements many times when talked to customers during implementations. The integration point is often becoming a competitive advantage. I believe for To-Increase, integration with other products is a significant competitive advantage. Manufacturing companies would be thinking twice before deciding what system will drive product development processes.

Partners and Integration Complexity

Integration is not a simple task. You need to have enough technological and process knowledge as well as technical skills to make it work. In addition, you can rarely find two identical integration solutions. Each manufacturing company will have their own practices, systems and specifics. Because of such high level of complexity, software vendors are trying to rely on partners to deliver an integration solution for end users.The ability of partner to deliver integration becomes a key in the ability to make an overall implementation success .

PLM Integration is hard. The cost of implementation is high. The cost of failure is even higher. Vendors are pushing integrations out of the scope of their deliveries. Partners, like To-Increase can provide a significant advantage by helping customers to make integration happen. These are realities of ERP and PLM implementations.

What is my conclusion? Integrations are important and complicated at the same time. Integration failures are one of the main sources that can cause overall implementation failure. The reliance on services increases the implementation cost and creates dependency of customers on implementation services. I’d expect software vendors to re-think their view how they can make integration easier. It can be a significant differentiation factor in future PLM systems. Just my thoughts…

Best, Oleg
Freebie. Nobody paid me to write this article.


PDM vs. PLM: An Integration Perspective

August 30, 2010

I’m continuing discussions about PDM vs. PLM differences. I’d encourage you to take a look on my previous two posts related to this topic: PDM vs. PLM: A Data Perspective and PDM vs. PLM: A Process Perspective. My next PDM vs. PLM comparison perspective is related to integration of PDM or PLM systems with other systems and processes in the organization.

Data Integration

For PDM, data integration is the ability to facilitate an exchange between product data records, other enterprise systems and extended value chain. The most important and very important is PDM – ERP integration. This integration provides an ability to exchange design, engineering and manufacturing data. For PLM data integration becomes more complex and needs to include an expanded set of application and data sources. Some of advanced data integration PDM projects are including federation of data between multiple systems.

Process Integration

When it comes to business and product development processes, both PDM and PLM systems can provide a significant influence on how these processes can be integrated. However, PDM system process scope is normally very limited and stay focused on engineering documents and data lifecycle. PLM process integration is focused on the ability of PLM system to have a tight connection with other processes. Most of them are ERP-based. These days it becomes more and more important to have a good support for integrated product development processes.

Social Integration

This is a relatively new aspect of integration and related to the ability of PLM system to be integration with social tools. Tools like Yammer and others represent a new class of tools helping people to collaborate in the organization. Both PDM and PLM systems provide a context for social tools. It can come as data about product, documents, changes processes and everything that helps to identify social relations. For example, by analyzing your product data, you can find a knowledge expert in a particular filed in the organization, etc.

What is my conclusion? Both PDM and PLM systems have very strong dependencies on their ability to be integrated within an organization. PDM integrations are oriented on data access, transfer and interoperability between a PDM and other systems in the enterprise. Data is the main focus of PDM integrations. It may include a different type of data integrations – manual, automatic, one or bi-directional data integrations. However, when it comes to Product Lifecycle Management, the scope of integration can be increased dramatically. It may include integration of processes, people and social context. The ability of PDM and PLM system to integrate what other systems and people in the organization are critical for successful implementations. This of the reasons why PDM and PLM implementations often fail in the organization.

Update: When I finished this post, I found a new post coming from Kurt Chen of Technology Evaluation Center. Kurt posted PDM vs. PLM A Matrix View. What I liked in Kurt’s post is the focus on the PDM/PLM need to be integrated with people and system in the manufacturing organization. This is a key in my view.

Best, Oleg


PLM and ALM: How To Blend Disparate Systems?

June 2, 2010

I had chance to read an article in SD time – Organization works to blend application, product life-cycle management. Author discussing the need to integrate two separate domain – development of hardware and mechanical components and software. I think, the message is very timely made. There are lots of software in modern products. Author brings an example of OnStar in vehicle communication. However, it is possible to bring more examples, of course.

Integration between disparate application having completely different set of data, rules and behavior is always a very challenging use case. In this case, author discussing the future of common standard creation that will help to integration PLM components and components managing software lifecycle (i.e. Rational tools). This discussion made me think about potential pitfalls and opportunities on this way.

Heterogeneous Application Environment
In the real world, many applications used during the design, engineering and manufacturing process. Mechanical, Electrical and Software teams are normally separate and relation between them quite limited from the software sides. This is the reality. In my view, when it comes to software, the disconnection comes to the top level. What can be a system that controls software build level need to be placed in the particular vehicle or other mechanical product?

Does One Standard Fit All?
The author is discussing OLSC (Open Services for Lifecycle Collaboration). I found the following video funny. The idea of community is going very much aligned with modern social approaches.

There are three key fundamental principles – URL, Minimal Schema and REST services proposed to make this solution work. I’m thinking how much time people will spend before they will agree about minimal schema that fit all. At in the end, as film states everybody wants to be a little different.

Don’t Integrate, Just Connect Dots
Here is my point. We don’t need to invent a minimal schema. It is enough to agree about to interlink different product representation- mechanical, electrical, software. Think about URL only. In my view, it will be enough to get job done. Global data identification similar to what we have in the internet can move us in the right direction. One of the examples of such technologies can be PURL. “A persistent uniform resource locator (PURL) is a Uniform Resource Locator (URL) (i.e. location-based Uniform Resource Identifier or URI) that does not directly describe the location of the resource to be retrieved but instead describes an intermediate (more persistent) location which, when retrieved, results in redirection (e.g. via a 302 HTTP status code) to the current location of the final resource.”

What is my conclusion? The landscape of application involved in this product development is very large. The number of applications is growing. The ability to absorb the requirements of all applications into one minimal single standards schema seems impossible. The new and more efficient way to interlink data need to be proposed. We don’t need to bring software build and engineering bill of materials to a single representation. However, we need to be able to interlink data related to different applications to maintain data integrity.

Just my thoughts…
Best, Oleg

Share


PLM and Heterogeneous Product Development

May 6, 2010

The issue of heterogeneous is really complicated. I think PLM vision and strategies are heavily missing this point. Most of PLM strategies are talking about a single source of truth, single data sources, single… However, in practice, companies are running multiple systems to accomplish their business goals. The following writeup about Parker Hannifin experience made by Jim Brown of Tech Clarity made me think again about options to manage heterogeneous product development.

CAD and Interoperability
The first phase in the list is an ability to exchange information between different CAD systems. I think, industry is discussing this topic for years (if not decades). There are many solutions for this problem and multiple companies are providing a different type of utilities. My last favorite in the discussion about CAD interoperability is the presentation made by Mike Payne at COFES 2010. You can see my write up here-  How to Liberate data?

PLM and Federation
The early version of interoperability between different PLM systems came with the ideas of Federation. These ideas weren’t introduced exclusively for PLM and actually discussed in the context of ERP and, in general, related to databases. The trial of PLM to establish a federated environment was very interesting one. I think, it really worked for few big customers. However, the cost to establish this environment is huge. The biggest problem is the amount of hand-wiring things you need to. I can see IBM is investing a lot in this direction discussing their PDIF program.

Single PLM Database Dreams
I can hear voices of PLM vendors to establish a single repository for all PLM information. It sounds very promising and interesting for customers. I think, currently, we are about to experience multiple tries to make these system’s works. It comes in the announcement and publications from all mindshare PLM vendors – TeamCenter Unified, ENOVIA V6,  Windchill… I see customers are actively searching for a solution in this space.

Future of PLM and Internet of Things
What do you think come next? I think, the future answer about solving problems of Heterogeneous Product Development need to come from the side of Internet technologies. All previous trials were “files and database oriented”. I think, the internet today demonstrates the capabilities of a big, single and self balanced system. Will it provide an answer to enterprise organizations and heterogeneous product development is a good question? We are going to learn it in the future.

One of the things I’m continuously thinking is the fact, you cannot change your product development systems and practices in a very short period of time (i.e. months). Even more, I think it is very hard to make a change in several years. So, companies proposing today a singe bulk of software will have a very tough time soon. I’m coming back to multiple thoughts about flexibility and granularity. In my view, it becomes a key.

What is my conclusion today? I think, future cloud-based solutions and the internet should impact product development strategies in the organizations. The connectivity, mobility, collaboration, data access – all these terms are going to impact product development practices in enterprise organizations. I think it is not simple to see it happen in the short term. However, I do see a cost of change in the existing systems as a major factor to make a movement towards a different set of systems.

Just my thoughts…
Best, Oleg

Share


PLM SOA – how to mix integrations and business processes?

August 14, 2009

soa-dangerSOA is a very oversold article these days. Even if, I think, social media and cloud successfully over hyped SOA during last one-two years, on the technological horizons SOA is still a topic that raises discussions and questions.

Wikipedia: In computing, service-oriented architecture (SOA) provides a set of principles of governing concepts used during phases of systems development and integration. Such an architecture will package functionality as interoperable services: functions provided as a service are available to be used from systems created by other organizations

So, a topic I want to touch today - what could be important components of SOA implementation in PLM. On the ground, Product Lifecycle Management should be yet another enterprise system. Why should we try to understand specific SOA topics? But, if we would take deeper look on what is going behind the scene in PLM, you probably will see what I mean.

One of the very important characteristics of PLM implementation is flexibility. Since almost every manufacturing enterprise organization is different, in the end of the days, you will find yourself making changes in your PLM environment. So, you better will be prepared and have all tools to do so. Another reason is integration with multiple external systems – design, engineering, manufacturing, supply chain. All these systems need to be somehow connected together since they provide ultimate input for PLM. Last, but not least – processes. Processes are flowing across organizational domains and boundaries. To be able to handle them, you need to have an ability to be connected to process oriented environment. It is normally, PLM environment or environment that comes from IT systems (depends on what you have in organization).

So, what is my conclusion? SOA is ultimate answer to provide PLM system configured and flexible. Two most important components of PLM SOA environment are to be able to integration with external systems and manage business processes. I think, PLM SOA having such characteristics will have very good chance to success these days.

Best, Oleg.


Follow

Get every new post delivered to your Inbox.

Join 71 other followers