Guest blog: We all speak the same language, don’t we? by Jonathan Eyre, High Value Manufacturing Catapult

We all speak the same language, don't we?

Jonathan Eyre, Senior Technical Fellow in Digital Twins, AMRC and Digital Twin Lead, HVMC

And what does industrial common language mean for data interoperability?

In conversation we use language to express thoughts and our point of view, but are the same words being used and understood by everyone in the same way? At the moment, this is most certainly not the case for a term like digital twin at the moment, especially with use cases all being so different to each other. This misalignment can be manageable for a small group discussion, but for larger world-wide collaborations the question becomes how can you trust and perhaps even validate that everyone is using the same language in the same way?

These issues are already deeply embedded in current information systems; information is actively consumed across organisations, supply chains and even across human languages that are all trying to exchange information without any loss of quality. Common acronyms can have different meanings; SME stands for both "small and medium-sized enterprise" and "subject matter expert" which even a small misunderstanding like this can cause major downstream issues. Ensuring language is being used in the same way for every end user is difficult, but not impossible as we’ll discuss.

So how do we create an industrial common language?

We live in a complex world where manufactured goods are produced for other sectors (like the built environment), that are then transported around to let other sectors like healthcare to provide services to society. A sprawling system of systems.

Creating consistency in this interconnectedness is not to be understated in its difficulty. As with most things, there is prior work such as “The pathway Towards an Information Management Framework” [1] where this report and other supporting outputs detailed key principles and captured common language formally as “Industry Data Models & Reference Data”. The scale of the overall challenge is overwhelming for any individual; however individually we don’t need to solve everything. The framework critically enables experts in their fields to create consistent language to support everyone in managing information quality all the way to the top.

This is what the Apollo Protocol [2] is empowering by having a method for convening forums to solve problems, establishing a consistent language for them and justifying 'why?' with evidence along the way. Language is an ever-evolving process and creating an industrial language is no different with on-going efforts required.

I'm convinced, but what does it really give us? 

Chiefly, data interoperability. This often gets mentioned superficially as being crucial to enable digital twins and cyber-physical systems, however, to get there it is common language that is a key step towards having it.

The next thing is producing reference data libraries (RDLs) which are a “particular common set of classes and the properties we will want to use to describe our digital twins.” [1]. These will define the underpinning common language structures that will enable a click of a button exchange of information between information systems without data loss.

Data interoperability and RDLs together provides a new layer to build upon for managing quality information ensuring overall consistency. Importantly though nothing is technology (or vendor) dependent and is simply a methodology to analyse the world backed with evidence. This, itself, has a lot of advantages but overall allows a much greater agility to enable its development in distributed environments, thus avoiding silos of information that are typically controlled by a dictated single source.  Critically, this consistent, but distributed, approach enables continual open extensions to improve and innovate the structuring of the language we build up, even in different data management systems in different ecosystems.

So, what next?

Consistent language is critical for data interoperability and requires input by everyone. Being able to agree on language though doesn’t mean that everyone needs to unanimously agree, but by creating understanding where we can in our respective areas will enable the success of the transformations we are all making. With this approach, other areas of opportunity naturally open up such as mapping reference data libraries, ultimately enabling us to solve the wicked problems we face together worldwide.

The Apollo Protocol and its approach is enabling convening to develop unified common language for industrial data. If you are involved in initiatives and events also trying to enable interoperability and data sharing, then perhaps consider what you can do to enable consistent language as a starting point?

Jonathan Eyre is a member of the DT Hub Advisory Board, Senior Technical Fellow for Digital Twins for the Advanced Manufacturing Research Centre and Digital Twin Lead for High Value Manufacturing Catapult. Contact Jonathan via the DT Hub. 

Links:

[1] The pathway Towards an Information Management Framework:

https://www.cdbb.cam.ac.uk/what-we-did/national-digital-twin-programme/pathway-towards-information-management-framework

[2] The Apollo Protocol: https://theiet.org/apollo-protocol

Join the Apollo Protocol network discussions:

https://digitaltwinhub.co.uk/networks/29-the-apollo-protocol/