Tag Archive

Two weeks ago, I shared my first post about PDM/PLM migration challenges on LinkedIn: How to avoid Migration Migraine – part 1. Most of the content discussed was about data migrations.

Starting from moving data stored in relational databases to modern object-oriented environments – the technology upgrade. But also the challenges a company can have when merging different data siloes (CAD & BOM related) into a single PLM backbone to extend the support of product data beyond engineering.

Luckily, the post generated a lot of reactions and feedback through LinkedIn and personal interactions last week.

The amount of interaction illustrated the relevance of the topic for people; they recognized the elephant in the room, too.

Working with a partner

Data migrations and consolidation are typically not part of a company’s core business, so it is crucial to find the right partner for a migration project. The challenge with migrations is that there is potentially a lot to do technically, but only your staff can assess the quality and value of migrations.

Therefore, when planning a migration, make sure you work on it iteratively with an experienced partner who can provide a set of tools and best practices. Often, vendors or service partners have migration tools that still need to be tuned to your As-Is and To-Be environment.

To get an impression of what a PLM service partner can do and which topics or tools are relevant in the context of mid-market PLM, you can watch this xLM webinar on YouTube. So make sure you select a partner who is familiar with your PDM/PLM infrastructure and who has the experience to assess complexity.

Migration lessons learned

In my PLM coaching career I have seen many migrations. In the early days they were more related to technology upgrades, consolidation of data and system replacements. Nowadays the challenges are more related to become more data-driven. Here are 5 lessons that I learned in the past twenty years:

1. A fixed price for the migration can be a significant risk as the quality of the data and the result are hard to comprehend upfront. In case of a fixed price, either you would pay for the moon (taking all the risk), or your service partner would lose a lot of money. In a sustainable business model, there should be no losers.
2. Start (even now) with checking and fixing your data quality. For example, when you are aware of a mismatch between CAD assemblies and BOM data, analyze and fix discrepancies even before the migration.
3. One immediate action to take when moving from CAD assemblies to BOM structures is to check or fill the properties in the CAD system to support a smooth transition. Filling properties might be a temporary action, as later, when becoming more data-driven, some of these properties, e.g., material properties or manufacturer part numbers, should not be maintained in the CAD system anymore. However, they might help migration tools to extract a richer dataset.
4. Focus on implementing an environment ready for the future. Don’t let your past data quality compromise complexity. In such a case, learn to live with legacy issues that will be fixed only when needed. A 100 % matching migration is not likely to happen because the source data might also be incorrect, even after further analysis.
5. The product should probably not be configured in the CAD environment, even because the CAD tool allows it. I had this experience with SolidWorks in the past. PDM became the enemy because the users managed all configuration options in the assembly files, making it hard to use it on the BOM or Product level (the connected digital thread).

 The future is data-driven

In addition, these migration discussions made me aware again that so many companies are still in the early phases of creating a unified PLM infrastructure in their company and implementing the coordinated approach – an observation I shared in my report on the PDSFORUM 2024 conference.

Due to sustainability-related regulations and the need to understand product behavior in the field (Digital Twin / Product As A Service), becoming data-driven is an unavoidable target in the near future. Implementing a connected digital thread is crucial to remaining competitive and sustainable in business.

However, the first step is to gain insights about the available data (formats and systems) and its quality. Therefore, implementing a coordinated PLM backbone should immediately contain activities to improve data quality and implement a data governance policy to avoid upcoming migration issues.

Data-driven environments, the Systems of Engagement, bring the most value when connected through a digital thread with the Systems of Record (PLM. ERP and others), therefore,  design your processes, even current ones, user-centric, data-centric and build for change (see Yousef Hooshmand‘s story in this post – also image below).

The data-driven Future is not a migration.

The last part of this article will focus on what I believe is a future PLM architecture for companies. To be more precise, it is not only a PLM architecture anymore. It should become a business architecture based on connected platforms (the systems of record) and inter-platform connected value streams (the systems of engagement).

The discussion is ongoing, and from the technical and business side, I recommend reading Prof Dr. Jorg Fischer’s recent articles, for example. The Crisis of Digitalization – Why We All Must Change Our Mindset! or The MBOM is the Steering Wheel of the Digital Supply Chain! A lot of academic work has been done in the context of TeamCenter and SAP.

Also, Martin Eigner recently described in The Constant Conflict Between PLM and ERP a potential digital future of enterprise within the constraints of existing legacy systems.

In my terminology, they are describing a hybrid enterprise dominated by major Systems of Record complemented by Systems of Engagement to support optimized digital value streams.

Whereas Oleg Shilovitsky, coming from the System of Engagement side with OpenBOM, describes the potential technologies to build a digital enterprise as you can read from one of his recent posts: How to Unlock the Future of Manufacturing by Opening PLM/ERP to Connect Processes and Optimize Decision Support.

All three thought leaders talk about the potential of connected aspects in a future enterprise. For those interested in the details there is a lot to learn and understand.

For the sake of the migration story I stay out of the details. However interesting to mention, they also do not mention data migration—is it the elephant in the room?

I believe moving from a coordinated enterprise to a integrated (coordinated and connected) enterprise is not a migration, as we are no longer talking about a single system that serves the whole enterprise.

The future of a digital enterprise is a federated environment where existing systems need to become more data-driven, and additional collaboration environments will have their internally connected capabilities to support value streams.

With this in mind you can understand the 2017 McKinsey article– Our insights/toward an integrated technology operating model – the leading image below:

And when it comes to realization of such a concept, I have described the Heliple-2 project a few times before as an example of such an environment, where the target is to have a connection between the two layers through standardized interfaces, starting from OSLC. Or visit the Heliple Federated PLM LinkedIn group.

Data architecture and governance are crucial.

The image above generalizes the federated PLM concept and illustrates the two different systems connected through data bridges. As data must flow between the two sides without human intervention, the chosen architecture must be well-defined.

Here, I want to use a famous quote from Youssef Housmand’s paper From a Monolithic PLM Landscape to a Federated Domain and Data Mesh. Click on the image to listen to the Share PLM podcast with Yousef.

From a Single Source of Truth towards a principle of the Nearest Source of Truth based on a Single Source of Change

• If you agree with this quote, you have a future mindset of federated PLM.
• If you still advocate the Single Source of Truth, you are still in the Monolithic PLM phase.

It’s not a problem if you are aware that the next step should be federated and you are not ready yet.

However, in particular, environmental regulations and sustainability initiatives can only be performed in data-driven, federated environments. Think about the European Green Deal with its upcoming Ecodesign for Sustainable Products Directive (ESPR), which demands digital traceability of products, their environmental impact, and reuse /recycle options, expressed in the Digital Product Passport.

Reporting, Greenhouse Gas Reporting and ESG reporting are becoming more and more mandatory for companies, either by regulations or by the customers. Only a data-driven connected infrastructure can deal with this efficiently. Sustaira, a company we interviewed with the PLM Green Global Alliance last year, delivers such a connected infrastructure.

Read the challenges they meet in their blog post:  Is inaccurate sustainability data holding you back?

Finally, to perform Life Cycle Assessments for design options or Life Cycle Analyses for operational products, you need connections to data sources in real-time. The virtual design twin or the digital twin in operation does not run on documents.

Conclusion

Data migration and consolidation to modern systems is probably a painful and challenging process. However, the good news is that with the right mindset to continue and with a focus on data quality and governance, the next step to a integrated coordinated and connected enterprise will not be that painful. It can be an evolutionary process, as the McKinsey article describes it.