The rise of the digital age is making it easier for manufacturing companies to access and share data, but even with all modern applications, cloud-based systems, and integration platforms, manufacturing, and supply chain is still a very fragmented and siloed space when it comes to the question about data access. The question of integration and data sharing is one of my favorites and here you can catch up to speed with some of my previous articles:
When considering Product Lifecycle Management (PLM) systems, a lot of questions related to data integration come up – how to share data between multiple companies and users, how to bring the data from one system to another system? Is it possible to create a universal standard to share and retrieve data between systems coming from multiple vendors? How can complex product information be visualized so that everyone understands what’s going on?
With innovations in technology, PLM vendors will be challenged to answer these and other questions about how to create a web of data or enterprise knowledge graph, etc. (the names my different as the marketing departments of the vendors are working hard to make it loos unique and specific for each vendor. The goal to create a web of data is a good one to make production faster, easier, and more efficient. What do these changes mean for manufacturing strategies today?
Web of Data, Digital Thread and OSLC
Lionel Grealou’s article OSLC Fest 2022: Making Digital Threads Real made me think one more time about achieving a digital nirvana of seamless data access and integration of data silos. The article was written as an afterthought for OSLC Fest 2022. Read the article and draw your own opinion. Lionel brings up the question of OSLC as a solution to make a web of data and compare it with Google’s web of data. Here are a few interesting passages.
This one is a dream or BHAG goal for enterprises in general and manufacturing in particular with the ability to connect multiple data silos together.
Building on OSLC principles, the event aimed to answer questions such as: “How can we achieve a Google for data? How can we query data from different databases seamlessly as if it were in one big global database? How can we establish links between data of different databases and how should we manage them? Similar questions have already been solved for documents by the World Wide Web. But what about data?”
The following article clearly states the current status quo of Digital Thread. Companies are buying into the idea of the digital thread, but apply it with the vision of “many digital threads”.
Simply put, building a digital thread is about connecting data across authoring sources and consuming applications. Many argue that there are multiple digital threads as data continuity is not linear across the product lifecycle, organizational functions and broadly speaking, across the extended enterprise.
Lionel also nailed the most critical question about how to enable data linkage between applications.
“Determine how OSLC enables MBSE by leveraging data linkages across requirements, functional and logical architectures, and physical product parameters. This (last) question is critical when considering how to drive system engineering traceability across multiple data sources, building PLM to PLM, or PLM to ERP interfaces. Avoiding data duplication and working with linked data is a core principle of effective product lifecycle management, independently of the system or technology.
Digital Thread is Not Yet Another Standard
The convo about the digital thread is interesting, but slowly starting to remind me of the discussions about interoperability and standard. Almost every time the standard initiative was trying to create a new and universal standard the industry ended up with a new standard. However, to be fair, the industry developed a decent amount of standards (among them STEP and a few others) to be able to exchange data between applications made by different vendors.
The difference in a digital thread is to stop “syncing data” between multiple applications and start sharing and linking the data in real-time. The difference here is big and I think the analogy of the web can help here to understand the difference.
From WWW to Enterprise Web of Data – What’s The Difference?
It is very common to see how digital thread projects are comparing themselves to the “google of data” vision and with the world wide web (www). The way we can get all data from multiple documents connected together is powerful. But, in my view, we should remember these three fundamental differences.
- Website creation aims to share information about business and anything else coming from these websites with everyone. Whatever information we publish online we either personally or from a business perspective want to get the data out. We don’t control (for most of the parts) how this data is used. Even, though many websites have copyright language prohibiting the copying of the data from this website, the goal of these websites is to allow users to get this data and use it or website services. Opposite to that, business applications were not built purposely to share data. Actually, the goal was to control the data and organize processes.
- Google technically built a database of all information turning it into a mechanism that allows easy access to information in a semantical form. But, the main business and financial enabler behind this development was the advertising model, which provided an enormous amount of funds to build the technology and Google power. On that, the business of enterprise software vendors is made based on data locking and growing the number of applications that can be used and leveraging the data. The principal value is vertical integration between apps is one of the strongest reasons customers are buying into a “single platform” even today.
- Creating a set of shared semantics in applications like CAD, PLM, ERP, and others is extremely hard compared to enterprise Although the world wide web is notoriously complex, the main challenge is to create a shared set of semantics is hard for search vendors like Google, Microsoft, Yahoo, Yandex, etc. In fact, these vendors started schema.org to facilitate the process of tagging information online to simplify the process of data understanding. Even so, the process of tagging is not even coming close to the level of complexity that is needed to allow to use of a piece of data from one system in another system (the dream proposal of digital thread)
What is my conclusion (and what is next)?
In my next article, I will talk about what I think companies can do to facilitate the creation of a common digital thread and how to get started. The idea of the world wide web of data is fascinating and powerful. Created by Tim Berners-Lee as an idea of the semantic web, it started the initiative and creation of standards such as RDF and OWL. In fact, OSLC roots and some tech is coming from there too. But before any of these initiatives start, the industry has to figure out how the solution for business models behind managing data originated in multiple applications. Without that, nothing will move. We need to find an alternative to the Google Ads model of the enterprise world. Just my thoughts…
Disclaimer: I’m co-founder and CEO of OpenBOM developing a digital cloud-native PDM & PLM platform that manages product data and connects manufacturers, construction companies, and their supply chain networks. My opinion can be unintentionally biased.