How to migrate into “future PLM platform”?

December 6, 2014

plm-platform-migration

One of the topics I touched in my yesterday post about future PLM platforms is platform migration. The ability of customer to make a move is significantly dependent on how existing environment can be migrated. You can catch up on some of my earlier thoughts about PLM migrations by reading the following posts – PLM upgrades, release cycle and legacy software; PLM migration and product data rock-n-roll; PLM cloud and future of upgrades.

Most of large manufacturing companies (and even smaller companies) already made some sort of investment in PLM products. What is ROI of move to a new platform? How to calculate it? How not to get troubled by supporting multiple versions of applications and environment? These are good questions. Customers and PLM vendors are equally interested how to manage it in a right way.

My attention caught Dassault Systemes’ 3Dperspective blog post – Top Three Considerations for Planning Your Move to the 3DEXPERIENCE Platform. It speaks about how customer can migrate into new 3DEXPERIENCE platform. Here is an interesting passage:

The same data model and business process rules that power the 3DEXPERIENCE platform also powered the ENOVIA platform. In fact, the same basic approach also powered the MatrixOne platform. This is why so many of ENOVIA’s current customers have been able to successfully upgrade since their first implementation in the mid to late 1990’s.

The following picture shows the history of 3DEXPERIENCE platform evolution. It basically means that the say foundation platform used by all MatrixOne and ENOVIA customers and migration is effortless. I’m not sure if I’m happy to know that the same data technology used by all generation of systems from mid 1990s. However, it is clear benefit for customers looking how to migrate data between different versions of MatrixOne and ENOVIA V6.

3D-experience-platform-evolution

Dassault System’s rival – Siemens PLM and its TeamCenter platform also has long history of transformations. I didn’t find specific public references on compatibility between data models and application among TeamCenter versions. However, the following article from Tech-Clarity blog by Jim Brown presents an interesting diagram of TeamCenter evolution – Siemens PLM vision 2014+.

TeamCenter platform evolution

More information about evolution of TeamCenter can be found in the following CIMdata document – TeamCenter “unified”. The following passage speaks about “migration” issues:

Siemens PLM will continue to support Teamcenter Engineering and Enterprise for those customers that have them in production. Importantly, with each release of these older products, they have updated the underlying architecture and technology so that when a customer decides to change, the transition to the unified Teamcenter solutions will be easier. They have also developed a robust suite of migration tools that can be used when moving from earlier versions of Teamcenter products to the unified platform.

What is my conclusion? The migration is a complex topic. It is probably one of the most important topics that will define ability of large vendors to move into bright future of next generation PLM platforms. Regardless on what platform customer is going to move, migration will have cost that must be calculated and validated. The idea of “federated platforms” brings some promise of minimizing of migration cost. However, the mechanics of this process is not very clear. At the end of the day, data must be brutally dumped out and transferred. Application migration is even more complex. Users must be re-trained. All together, it is not a simple task. Just my thoughts…

Best, Oleg


It is not easy to add “social” to PLM

December 1, 2014

plm-social-universe

My recent post Social PLM: How to pull a trigger? became a trigger for me to think and discuss “social” topic again. I found certain level of disagreement with Jim Brown’s position about how to put social in PLM. Here is Jim’s comment:

@jim_techclarity: @olegshilovitsky There is a lot more low hanging fruit elsewhere. Easier to add social to PLM than PLM to social. Talk to you in 2019

I’m sure Facebook is not coming after PLM market. Getting back to my Facebook at Work post, my hunch is that Facebook is trying to crack the complexity of social behavior at work. And… this is not easy. And, therefore to add social to PLM is probably not easy too. We’ve seen so many established vendors and startup companies that tried to “add social” to CAD and PLM tools. Even to develop new tools. So far, it ended up mostly with Facebook copycats and collaboration around comments in CAD models and drawings. Is it all about social at work? I stuck with the statement of “easier to add social to PLM”. I don’t see “social” as feature or add-on that you can add to existing tools or technologies. In my view, social is a reflection of people relationships, behaviors and fundamental needs. It took me more than few days to come with some thoughts about that and I wanted to share it with you.

I came back to one of my posts from the last year – Why Social PLM 1.0 Failed? My main conclusion back there was about value proposition of social tools. The fundamental interest of people in social tool was driven by variety of social behaviors. Facebook added true identity into social networking and it gave a spike in usage among people interested in other people activities, social status, relationships, photos, etc. LinkedIn helped to develop our professional identity and find beneficial connections for our carrier and business. Twitter gave us ability to follow variety of information stream in a very easy way.

Adding social into PLM, ERP and other enterprise tools means to improve communication. In my view, this is a place where all enterprise social tools have some sort of mixed achievements. Let me bring some examples to explain what I mean. I can see two fundamental communication patterns in a company – enterprise transaction system (ERP, PLM, CRM) and email.

1- Transactional communication. The main focus of this type of communication is about business processes. It is about adding new position in your organization using CRM tool, creating new Part Number in ERP system and approving ECO using PLM system. There are many other examples. But, in all these cases, people need to perform a transaction – this is a main purpose of their communication.

2- Email. Everything that doesn’t fit transactional communication goes into email. We have many complains about email overload. At the same time, we like email. It creates communication record, can prove what you did and in many situation serves a document. The ubiquity of email and mobility makes email a convenient tool to communicate.

Enterprise social tools are pretending to simplify communication and make information in the organization more transparent. It is true. I can see few situation where social tool can be successfully used by sales and marketing team to update company and synchronize about their activity. I also can see social collaboration as a good tool that can help to find an expert in a company based on previous records and projects. However, it doesn’t replace the fundamental capabilities and functions of email and, as a result, turns out to be another communication channel.

What is my conclusion? Similar to successful social networks, enterprise social tools need to find a single purpose for people to use it. Not as a vitamin, but as a painkiller. It is like a Facebook that gave you a way to watch people activities online. What is that function for enterprise organization? How it will be able to replace email or other tool? How it will make people to justify “another communication channel”? All these questions need to be answered. As a result we will invent “social tool for enterprise” that will become powerful as Facebook or LinkedIn. In my view, enterprise social tools and specifically engineering (PLM) social tools didn’t provide enough value to people to change their traditional communication…. yet. Just my thoughts…

Best, Oleg

photo credit: Nimages DR via photopin cc


Social PLM: How to pull a trigger?

November 19, 2014

plm-social-trigger

In my yesterday blog, I shared some of my thoughts about “Facebook at Work” and potential impact on engineering and manufacturing software. It made me think again about all discussions and stories related to social software trend and social PLM.

Social was trending topic 3-4 years ago. Many new companies were founded back those days to realize the idea of “social enterprise”. No doubt it was heavily influenced by the large popularity of social networks and web. Facebook, LinkedIn, Twitter and many other companies supported that trend. In enterprise domains, companies like Yammer and few others made a successful swing by introducing their products in business domain.

CAD vendors reacted on social trend by introducing “social platforms” or “social products”. It was mix of new development (Dassault 3DSwYm) , partnerships with other vendors (Microsoft SharePoint – PTC Windchill Social Link) or technology acquisitions (Autodesk – Qontext). In my view, none of these products and technologies changed a way people used to work in organization. Vendors blamed engineers for anti-social behavior. Customers blamed vendors for non-intuitive user behavior, missed features and bad integration with engineering content.

I put some of my notes about why I think social PLM trend lost the opportunity: Why Engineers Don’t Like Company Private Social Networks? and Why Social PLM 1.0 failed? I guess,despite all attempts, email remains one of the most visible collaboration tool for most of engineering and manufacturing companies.

At the same time, I believe, it is a time to rethink ideas of how to bring social into business eco-system. This is not an easy task. Some of key elements can be – structured enterprise communication, integration with existing messaging and communication system and thinking about social in connection to task management.

I had a short tweeterstorm with Jim Brown yesterday, following my blog about “Facebook At Work”. Jim kind of dismissed my points about usefulness of Facebook as a business social tool. Jim stands behind his old article – What I Learned: We are not Going to Design an Airplane on Facebook!

@olegshilovitsky: Maybe @jim_techclarity will rethink his 5 years old statement about airplanes design on Facebook

@jim_techclarity: @olegshilovitsky No, not rethinking it. Re read the post and you will see my predictions are pretty spot on. This time, at least

@jim_techclarity: @olegshilovitsky There is a lot more low hanging fruit elsewhere. Easier to add social to PLM than PLM to social. Talk to you in 2019 ;-)

jimbrown-olegshilovitsky-tweetstorm-social-software

Most of the time, I’m in agreement with Jim. However, I disagree on his view on social software add-on role. We should not view “social” software as an addition to something else. This is a mistake that many enterprise software vendors did in the past few years. Social is a reflection of people behavior. I believe, this is a fundamental thing in what Facebook did – to emphasize the importance of “real person” identification in social network. All social networks before dismissed the importance of identification. Facebook made it as a core function and won. Of course, it doesn’t mean the same thing will work for social enterprise or PLM.

Facebook at Work is coming to capture attention of people during work time. Today, many places are banning Facebook as a destruction. I guess, Facebook wants to figure out how to offer social value for people during their work time. To me it means to crack “social behavior at work”.

What is my conclusion? The main point of “social” is not to become a nice addition to enterprise software to improve collaboration. Unfortunately, this is how most of enterprise software vendors (PLM included) understood it. Collaboration is important and existing PLM products need to focus how to improve user experience. But, social is not about that. Social is about “behavior” of people at work. Today, it is mostly around email and important business systems. Not much social. We don’t know how to pull the trigger of social software in a company. In order to do that, we need to think about “behavior”. We need to think how to make people addicted to business social function through the value it brings on everyday basis. Just my thoughts…

Best, Oleg

photo credit: Ian Ruotsala via photopin cc


Future CAD file management trajectories

July 1, 2014

cad-file-management

CAD data is a core and one of the most fundamental parts of every manufacturing design. It all starts and dies from how your product looks and feels. In the past, 3D CAD was most focused on mechanical elements of design. Today, CAD systems are meshing into a complex conglomerate of data about shape, assembly, simulation and many other aspects of product design. Thinking even more broadly, CAD files are representing a significant part of engineering and product knowledge.

Despite overall significance, many manufacturing companies and engineering organization are missing the point of CAD files management. It is not unusual to hear that 60-70% of companies manage CAD files on shared network drives. Engineering.com article – The Risks of Manually Managing CAD Files speaks about what is a danger of keeping your CAD files not managed. The following passage makes it very clear:

Perhaps the most common way to manage CAD files is on a shared drive with a directory structure and file naming conventions. That can work in some situations, but it carries significant risks and limitations. "It was easy when it was just me, but when we added a second person it was difficult to have the same file structure," recalls Andy Homyk, the lead mechanical engineer at medical device company HemoSonics. "It was hard to get updates from his computer onto mine and ensure I had the right revision."

These manual approaches are better than nothing, but in all but the simplest scenarios lead to errors. As complexity and number of engineers increase, unmanaged approaches fall apart. Relying on individuals to consistently follow manual rules eventually leads to problems. This approach frequently results in the errors discussed earlier, specifically overwriting each other’s work, using the wrong version of a file, multiple people working on the same file, and lost productivity.

Article references e-Book written by Jim Brown of Tech-Clarity with more detailed discussion about how to choose write level of available CAD file management solution – from keeping CAD files on shared drives and up to full PDM system. Jim mentioned possible solution in between, which is CAD file sharing on the cloud. The article and e-book made me think about potential trajectories of future CAD file management solutions.

1- Cloud File Systems.

Quite a few companies these days are trying to virtualize file system and make transparent between on-premise and cloud storage. If it turns into reliable, fast and cost effective solution, engineers can just use this cloud file system to save files. I can see a good opportunity for cloud file systems to support revision history. So, it is almost PDM and these companies can start eating PDM lunch.

2- PDM with cloud file storage

The complexity of CAD data can make option #1 not very reliable. In that case, we can see a next turn in the evolution of existing PDM system – turn them to the cloud via IaaS and / or cloud hosting. Technologically, these solutions can be very similar to any existing PDM system. It might require some tuning to work with low latency and cloud file storage. But underlining idea will remain the same.

3- Engineering data platforms

This is one of the most interesting trajectory for me. Somebody would like to re-think the way engineering data (include CAD data) stored and managed in the cloud. The process of re-thinking can touch also technological aspects (databases and storage) as well as logical and functional aspects related to collaborative design and engineering and more.

What is my conclusion? One of the biggest challenges these days is how to leverage cloud system advantages on top of massive amount of CAD files. Every engineering organization is struggling to find an efficient solution to manage engineering data accumulated on desktops and network drives. Security, cost and scale – these are three most important elements every manufacturing company will be assessing to find an appropriate CAD file management solution. Just my thoughts…

Best, Oleg


What makes “cloud” a good alternative for PDM system?

June 20, 2014

cloud-pdm-selection

It has been a year since I published my How to select PDM system in 5 simple steps? Engineering.com article The Difference Between Cloud-based and Traditional PDM made me think it is a good time to re-evaluate my writing and see if I need to correct my recommendations. Note, Engineering.com article is paid promotion by GrabCAD. However, as stated in the bottom of the page – GrabCAD haven’t had an editorial input to the post.

The article makes a comparison between traditional PDM and Cloud PDM for three different customer segments – small, medium and large. It brings some very good characteristics of these environments in terms of how companies work, IT capacities, process maturity, etc. Nevertheless, with different pros and cons, after all, the conclusion is simple and straightforward – Cloud PDM maybe an option for all these companies. Here is a passage which explains that:

Whether you are a part of small, medium, or large business, it is clear that cloud-based PDM has a lot more to offer than file system management and in some cases more than traditional PDM. While traditional PDM has been around long enough to be a mature product, there is no such thing as an off-the-shelf installation. Traditional PDM requires experts trained in the front-end and back-end administration. It also requires significant investment in hardware and infrastructure.

The article is long and contains comparison tables. It took me some time to review all of them. I’ve been looking how to capture the enhancement of my simplified PDM selection process. After few reads of the documents, I finally got what I need. Here is my +3 points to PDM selection process:

1. Global access to CAD data: If global access to CAD data is important, you can get significant advantage from cloud PDM system. You will have much simpler access including mobile application option.

2. CAD / PDM integration. Cloud PDM still requires integration with CAD environment and this is important for all types of companies (in my view, this is a missing point in the article, which points on integration needs only for medium-size companies). So, in case, there is no specific CAD plug-in for cloud PDM, you might prefer traditional PDM environment.

3. Cost. Cloud PDM will shift your PDM expenses into operational cost. It is hard to say something about TCO and ROI, but clearly, you will be able to run PDM environment with much lower upfront cost.

If you are interested to read more about Cloud PDM alternative, I can recommend you the following article – Lightweight CAD Management using the Cloud by Jim Brown. The following passage is my favorite:

Living in an unmanaged, manual environment is highly inefficient and prone to errors. For companies that have outgrown chaos and are tired of crossing their fingers and hoping they don’t order or produce the wrong part, it’s time for a practical solution. Fortunately, there are more options available today than ever before, including new cloud-based tools. It’s time for a rational discussion to explore the basic requirements for CAD data management and discuss whether they can be achieved without the cost and complexity that make traditional solutions impractical for many smaller manufacturers.

What is my conclusion? If you have reliable internet connection and not very complicated IT landscape you can find cloud PDM gives you overall advantage in terms of cost and accessibility. If you have complex integration requirements, traditional PDM will be a better option for you. Larger companies might decide to implement multiple PDM systems anyway to match multiple CAD systems environment and/or facilitate work with suppliers via cloud PDM options. Just my thoughts…

Best, Oleg

Picture courtesy of GrabCAD.com


How to eliminate PLM customization problems?

March 28, 2014

plm-customization

I’m following strategic visions of the major PLM vendors 2014+ publication by Jim Brown – well known analyst and my blogging buddy for last few years. It started as a publication covering Autodesk, Dassault, PTC, Siemens (vendors listed alphabetically). Last week, Jim expanded his PLM vision publications by adding Aras Innovator to the list. Navigate here to read about Aras 2014+ vision. Aras is well known by their Enterprise Open Source strategy. One of the interesting differentiation I captured in Jim’s article is related to Aras’ strategy to break rules of PLM customization. Here is the passage:

Aras has decided to break the rules [of PLM customization]. They aim to become the PLM company that defies the conundrum, allowing manufacturers to customize their software and still upgrade to future releases without major disruption. They can do this because customers can update the data schema, business rules, workflows, and forms without jeopardizing the integrity of the system. How does this work? Aras’ XML-based, model-oriented approach coupled with their willingness to provide customers with the business flexibility and tools to make it feasible. Aras has effectively morphed themselves into a PLM Platform with solid core functionality with a built in ability to be extended by customers and partners. To put this strategy into action, they have told me they are “putting their money where their mouth is.” They now include upgrade services as a part of their subscription service. I haven’t seen that from anyone else anywhere, particularly while encouraging people to enhance and modify the package. This is a clear differentiator and makes Aras unique in the PLM market.

PLM customization is a tricky deal. Honestly, nobody is dreaming to make PLM implementation with zero customization effort. It all starts from flexible data modeling, which imply certain level of data customization. Time ago, I posted – Is PLM customization a data management Titanic? Earlier this year, I’ve been discussing options and reasons on How to de-customize PLM? The story of PLM customization is tightly related to PLM system flexibility data modeling. Typically, every PLM implementation contains some portion of customization that usually done by service organization and/or internal IT department. Lifecycle rules, data import, workflows, integration with other enterprise systems – this is only a very short list of customizations done during PLM deployment. Another huge aspect of customization is related to system upgrades. That one is actually mentioned by Jim Brown in his Aras’ review.

So, is there a way to solve customization problem? In my view, the answer is – it depends. In my view, you cannot eliminate specific implementation activities. Adding of new features and infrastructure technologies (eg. RDBMS) will require certain upgrade activity to happen. However, if you are selling services, the interest will be to optimize this work. Cloud vendors have similar incentive to optimize infrastructure upgrades and maintenance, otherwise operational cost will go up. So, smart technology can optimize cost and customization efforts.

What is my conclusion? Business and technology are going together. To have good business incentive to optimize technologies is always helpful and can put pressure on development organization to optimize cost of infrastructure upgrades. Service based offering (open source and cloud) are two great examples where business interests of vendors and customers are going at the same direction. Just my thoughts…

Best, Oleg


The Future Unbundling Strategies in CAD/PLM

November 6, 2013

unbundle-cad-plm

Engineering software has a long history of aggregation and disaggregation. In CAD, it was long time about new features vendors added to CAD tools. Ultimately, it was a long run towards new features. Later in the history, CAD industry came to the idea of integrating of various add-on systems into CAD environment. So, we got simulation, CAM and many other features and systems. In addition to that, data management tools (PDM) drove their own road – first independent from CAD and later on more integrated. Later on, PLM ideas came and provide an additional set of tools, features and systems. The latest history of design and PLM tools development introduced even more "integrated product lines" with CAD, PLM and many other systems. These days leading vendors are providing product suites with huge number of tools and subsystems bound together.

I was reading Unbundling: AOL, Facebook and LinkedIn article and reviewed a very fascinating diagram about Craiglist. Here is an interesting quote:

There are a swarm of services, often mobile first or mobile only, trying to peel off parts of the Craigslist offer, or do things Craigslist should have been doing. AirBnB is only the most obvious. Chris Dixon has a good note about this here, and Andrew Parker produced this great graphic back in 2010.

craiglist-unbundled-services

Major CAD/PLM vendors are selling big product suites, which can be considered similar to Craiglist. CAD / PLM product suites are providing lots of functionality. However, each function or service is probably not "best in class". PLM analysts and duelists Jim Brown and Chad Jackson discussed it in their video blog – CAD: Granularity vs. Integrated Suites.

Another interesting interesting perspective was presented by Adam O’Hern in his blog – Why I will pay $1372 for a fillet tool? Adam is writing about tools diversification and CAD subscriptions value prop. Adam is bringing the example of Autodesk Fusion 360 as cost effective tool that can be used as a substitute for some of features of SolidWorks. Here is an interesting passage about CAD tool diversification:

I use whatever combination of tools offers me the best cost-benefit ratio for whatever task I’m trying to achieve. In my various corporate jobs I’ve used CATIA, UGNX, Rhino, and Alias for design work, and each has its own advantages. For the last five years or so I’ve found that a combination of SolidWorks and MODO–along with various plugins, scripts, and sidecar tools–has provided the best price:utility ratio for my specific 3D design needs.

The unbundling article made me think about what will happen if we will follow similar strategy for CAD and PLM. Here is my top 3 assertion:

1. CAD and PLM is too big to sustain as a one big aggregated solution provided by a single vendor. This is a polystate diversified space that needs to be covered by multiple solutions, features and vendors.

2. Vendors are never good enough to see what exact problem customers want to solve. Especially when it comes to large manufacturing companies and complicated supply chain eco-systems. That’s way armies of consulting services as well as diversified products must be applied to provide a final solution.

3. Customers often don’t know what problem to solve. For most of the situations product development is a complex problem. It requires the team of people to work on. In addition to that, large organizations are involved into politics and confrontation related to usage of different enterprise software and tools.

What is my conclusion? If my assertion is correct, future unbunlding of features and services from CAD/PLM product suites can provide a potential opportunity to diversify market and solution options available for manufacturing companies. Cloud and SaaS models will be very helpful in this process. It is much easy to use unbundled cloud services compared to old set of CAD/PLM tools. Just my thoughts…

Best, Oleg


Follow

Get every new post delivered to your Inbox.

Join 262 other followers