Last week I attended ICCAS 2019 – International Conference on Computer Applications in Shipbuilding in Rotterdam. This year’s unofficial theme was definitely Digital Twin and good old fashioned Data. Virtually every presentation mentioned the digital twin, to the point that if you did not mention it, you were not cool.
I have definitely seen an evolution in the presentations and the discussions at ICCAS compared to the previous event. The understanding and appreciation of the digital twin has grown as well, and I have found that many are aligning their presentation message with the digital twin ideology, which is good for consistency throughout the industry.
The conference is focused more on technology and the challenges it promises to solve in our industry. It is understandable and expected, but from my perspective, I would like to see more how that technology can be implemented, taking the realities of our businesses into account, to achieve our business goals. There were some practical implementations which did provide some insight into how our business goals could be achieved, I simply would like more. Most of us understand the theoretical benefit of many of these new technologies, but are still unsure how we can apply them. Especially when encountering some real, non-technology related, obstacles such as culture, owner mindsets, data ownership, perceived security, traditional processes, etc.
Below are a few of my current thoughts on the key topics covered at the conference: digital twin, data and relationships, the platform of platforms ecosystem, culture, production and VR.
Digital Twin for the Shipbuilding Industry
As I mentioned, almost every presentation and conversation mentioned the digital twin. I really think that this is good for our industry as it provides a common understanding and aligns many of our initiatives and strategies. Since no solution to a challenge exists in a silo, having a concept like the digital twin allows us to leverage a single implementation to solve more than one challenge and achieve several goals.
I was glad that our industry has recognized that there is not just one fully encompassing digital twin model. There are multiple digital twin models, as well as several types of digital twin, as defined by John Vickers when he coined the term. My blog post No Single Digital Twin Platform to Rule Them All elaborates more on the topic.
Shipbuilding Data & Relationships
A key element to achieving a digital twin is digital data, so it makes sense why it got a lot of our mindshare during the conference. Today, we have no issue creating vast amounts of digital data. The challenge is that the digital data that we are creating does not have the correct links to digital artifacts created by another authoring tool – whether that data is created in a different department or even from the same user.
Data consistency was also mentioned as a challenge that needs to be overcome. There was an example mentioned where a person got several different digital artifacts from someone else in another company, but those arifacts had conflicting data. On one digital artifact the dimension was ‘X’ but on another digital artifact it was ‘Y’.
To have reliability and consistent data we will need to make some changes to our business processes, strategies, culture and technology.
The Elephant in the Room – Ownership
With all this discussion around the digital twin, the question of “Who is going to own the digital twin?” came up. Also, the question of “How are we going to get the owners to pay extra for the digital twin?” was mentioned. No one wanted to answer these questions as they are tough. My perspective is that I think those are the wrong questions to be asking?.
Platform of Platforms
I have blogged several times about what I like to call the “Platform of Platform Ecosystem.” The idea is that the current and future state of any company will need to embrace a platform of platform ecosystem where the digital footprint of any asset will be encompassed in multiple platforms, rather than in one single monolithic platform.
I did not see one presentation that failed to align with this platform of platform vision. Everyone seems to understand that in the future we will have to connect systems/platforms together, even if they are from different vendors. However, I do not feel everyone understands the full realization of what this actually means. Some were just talking about creating standards to transfer information but ignored how the information in multiple platforms need to be linked/connected, it’s not just about transferring data. Also, some were talking about creating open APIs, but lost the idea that it is a two-way street and they will need to be able to leverage other platforms’ APIs.
Having a truly open platform will require unrestricted access to data, relationships and even initiating behaviors or actions. It is a fundamental shift on how many companies think about their solutions. It is good that vendors are talking about it, but the proof will be if they continually grow their solution to play in this new world. Many companies still feel (but do not externally communicate it) that a “good business strategy” is to hold their user’s data hostage. As you know, I am big on true openness. No company should compete by holding their client’s data hostage.
Mindset & Culture
There was some discussion that our industry is adopting technology a lot slower than other industries. Some thoughts were that because a majority of shipyard personnel are over the age of 50,it is hard to initiate change. However, people have been noticing that the younger generation is moving up the ranks and is more comfortable adopting new processes and technology.
Many of the current workflows were created when we had significantly different constraints. The key is that we need to think with the end in mind. What are our business goals? We cannot simply digitize our whole organization overnight, but we can look at important areas of our business and redesign the processes using today’s constraints, opportunities and technologies.
OSX and Shipbuilding Standards
I found the presentation from Ole Christian Astrup very interesting. He was talking about a standard (OSX) that DNV GL, several clients and several competing software vendors were developing to pass information to class, DNV GL in this case, for validating the ship model.
I have a love/hate relationship with standards, but the cool thing about this one is that all the companies involved have invested their own time and money to create it with little bureaucracy. It is not commercially viable at this point, but I do intend to follow this (unofficial) standard.
Production / Manufacturing
Surprisingly, manufacturing does not really get the attention it deserves when we talk about our digital information. The ability to better manage your shipyard shop floor, and the capability to drive machines and/or robots, can result in tremendous cost savings, improved quality and reduced risk.
I enjoyed that Marcel M. Veldhuizen discussed the importance of having reliable digital data with ideally connecting information from the authoring tool. Ultimately, ensuring high data accuracy and the correct level of fidelity you need.
He mentioned that if you used his biggest competitor (Microsoft Excel), the complexity of the shipyard is simply unmanageable; resulting in inefficiencies and missed opportunities.
Virtual Reality in Shipbuilding
Virtual Reality got a lot of airtime at ICCAS. It was really nice to see a lot of use cases in the real world (pun intended). There were some who implemented their own VR solution using one of the common game engines, but the majority seemed to be using a VR solution from a VR company. Game engines look interesting for quick prototypes and researchers who want to play with VR. However, for the majority of our workflows where we can benefit from VR, the gaming engine pipeline (CAD->Export to Mesh->Import to game engine->process scene->wait->VR ready) is not streamlined. However, some successful implementations, such as at Fincantieri, do use a game engine.
There have been a lot of incorrect assumptions around VR for shipbuilding, like that it cannot handle ship models. However, it’s clear that the quality and performance of VR can completely handle even the most complex ships of today. These assumptions were formed because it is easy to create a VR environment using a gaming engine that does not properly consider basic VR best practices. The result of these easy and quick solutions is a very bad VR experience, with even a modest sized ship. Almost all commercial VR solutions consider VR best practices and don’t suffer from these same issues.
I must thank David Thomson, ICCAS’ resident VR expert, on sharing a lot of new and interesting information about VR. I will potentially share some of those insights in another post.
ICCAS is a great event and one of the very few that is focused specifically on technology in shipbuilding. As usual, I learned a lot, connected with old colleagues and even made some new ones. This conference has reaffirmed several key aspects of the future of technology within our industry:
Digital Twin has definitely caught on in the shipbuilding industry and is becoming a common foundation for us to communicate, define our strategies and build our solutions.
There will be several digital twin models that will represent the different types of digital twins. Data and their relationships are fundamental to a successful digital twin, as is implementing a platform of platform ecosystem which will connect that data across platforms from different companies and vendors. The digital twin will drive several initiatives such as VR and better manufacturing.
Creating digital data is easy and not a problem these days; however, the challenge of creating consistent data that is linked to other digital data is very real.
Finally, we will need to continue to figure out how we can solve the non-technology challenges slowing the adoption of these technologies in our industry.
View the paper and presentation SSI presented at ICCAS 2019.