Megatunnels You've Never Heard Of: Data-Driven Strategies for Infrastructure Owners

YANISSA DE JONGHE: Welcome, and thank you for being here today. We appreciate your time and interest. Today, we are here to discuss the Oosterweel link Project, a crucial project for Antwerp, Belgium, and, by extension, Western Europe. Oosterweel link is a project currently under construction. And we want to take you to the subsequent Operate and Maintain phase and the vision we have for this phase.

First of all, let's introduce ourselves. So I am Yanissa De Jonghe. And I am IT Manager at Lantis.

BAS VAN EKELEN: Hello, everyone. My name is Bas van Ekelen. I'm Project Manager, Asset Management at Lantis.

GUILLAUME NEYRINCK: And I'm Guillaume Neyrinck, Senior Customer Success Manager for ACC in the Benelux.

YANISSA DE JONGHE: Thank you, guys. And this is the agenda for our session today. We start with a commentary on the Oosterweel link, followed by social trends and how we interpret the concept of smart infrastructure.

Then, we will explain to you the strategy of infrastructure owner organization that we are building to then come to a demo of our technical implementation and conclude with a short overview and some time for questions.

The Oosterweel link, what challenges do we face? We start with an introduction from the well known YouTube channel, The B1M, which nicely explains the project we are realizing today. We will share the link of the full B1M video about the project at the end of our slide deck. It elaborates on the civil construction and the challenges we face over there.

[VIDEO PLAYBACK]

- Antwerp in Belgium, a place known for its fine arts, fashion, and a busy port that made it the center of the world's diamond trade.

- What do I know about diamonds? I come from Antwerp.

- And yet, it's not all glitz and glamour here. And you've only got to go for a short drive to find out why. [HORN HONKS]

- Yes, this beautiful old city has a big traffic problem, one that's not just impacting local people and industries, but the rest of Europe too. But thanks to some incredible underground construction methods, Antwerp's now on track to complete a massive piece of infrastructure, one that it's been trying to finish for decades.

- It's big.

- The area around the city is becoming one epic building site, creating a lineup of the most impressive engineering this country has ever attempted.

- Sites like this are why I love construction.

- From sunken river tunnels to routes stacked inside a canal and massive new parks, sit back and get ready for a deep dive into one of Europe's most important and lesser-known megaprojects.

There aren't many European cities better placed for international trade than Antwerp. It has the Netherlands to the north; Germany to the east; Brussels to the south; and to the west, that big, old patch of water known as the North Sea.

- Antwerp has one of the world's major seaports. Huge highways with other countries and cities all lead here, converging in a single stretch of tarmac in the outskirts known as the Antwerp Ring Road.

- But hold on a minute. Isn't a ring road supposed to be ring shaped? Well, yes. And because this one isn't, getting from A to B by car can be a frustrating experience.

- In 2022, drivers spent an average of 61 hours stuck in traffic, thanks in part to that unfinished ring road. The existing tunnels under the river have become bottlenecks. And the congestion has spread out to residential areas.

- If all of that weren't bad enough, the road is also a crucial element of the Trans-European Transport Network, connecting Paris and Amsterdam along the North Sea Mediterranean corridor.

But things are about to change. Antwerp is finally putting the last piece of the puzzle in place with the fantastically-named Oosterweel link project, a set of building works that should be simple but really aren't, creating hassle for construction teams and residents and great material for a construction YouTube channel.

A lot of the challenge here is down to geography.

- First off, there's a river in the way. And any attempt to complete the ring would have to contend with another water obstacle, the Albert Canal.

- It's why this is being called "Belgium's project of the century." It was first put forward way back in the mid 1990s. And it's now being developed by Lantis on behalf of the Flemish government.

[END PLAYBACK]

YANISSA DE JONGHE: Lantis, the company Bas and I work for, is currently building the Oosterweel link, a complex project in a very dense environment, which only adds to the challenge of completing this project.

In this image, you can see an overview of the project. Everything at the top marked in orange and yellow will be built in the Oosterweel link project. Orange indicates the tunnel work on the bottom left, you can also see a piece of orange. This is the JF Kennedy Tunnel. It is an already-existing tunnel crossing the river Scheldt. It is scheduled to be completed by 2033.

Meanwhile, parts of the Oosterweel link are already being delivered and managed by Lantis. To this end, part of Lantis is transforming itself into a new kind of organization, an operate-and-maintain organization, an infrastructure owner instead of a client.

Today, we'll take you through how we're going to meet this challenge and how we see modern infrastructure-owner organizations. Spoiler alert, it will be data driven.

Before we go further into the presentation, I still don't want to withhold this animation from you. This is one of the tunnels we are building. The dense environment compels us to build this complicated piece of infrastructure, quite challenging. This example was created in Navisworks, by the way. We can show this because the data-driven strategy we will discuss later on has also been translated into a strong BIM strategy.

The parts of Oosterweel now being delivered were unfortunately not delivered according to this data-driven BIM strategy. The contracts for the first part had already been signed before this choice was made.

Anyway, smart infrastructure, what does it mean to us? In our opinion, smart infrastructure is directly related to the future-proof asset management. As we've seen earlier, building the Oosterweel link already takes a long time, but the infrastructure also must last a lifetime after it.

Therefore, the infrastructure and platforms around it must be super agile and flexible beyond 2033. We need to remain technologically relevant.

To define our strategy as an infrastructure owner, we must consider a number of societal megatrends. Because of the long lifetime of Oosterweel link, we need to take these trends into account. What megatrends are relevant for the Oosterweel link?

Climate change is driving a global push towards sustainability. Reusability and CO2 neutrality are no longer optional, but essential. Our goal is to create CO2-neutral tunnels striving for a greener future. The further globalization continues to affect global supply chains, European growth, and the activities of, of course, Antwerp's Port.

The interconnectedness of our world means that local changes can have global repercussions. The aging population is putting pressure on the functioning of our working society. As the number of elderly increases, we must adapt our systems to ensure they remain effective and inclusive.

But, of course, also labor shortages-- shifts in the workforce, including skill gaps and changing employment patterns, are creating both challenges and opportunities. Addressing these labor shortages requires innovative solutions and adaptive strategies. If we look at the top 10 bottleneck jobs in Belgium, half of them are construction related. This will have an impact on us.

Digitalization is connecting machines with each other and with humans in unprecedented ways. This increasing connectedness is changing industries and daily life, creating new possibilities and efficiencies.

And, of course, urbanization-- urbanization is leading to increasingly concentrated living and working environments. This trend has a significant impact, especially around dense areas like the Oosterweel link. For instance, Antwerp is expected to grow by 10% by 2030, presenting additional urban challenges in housing, work, infrastructure, and facilities.

These six megatrends are not isolated. They interact and influence each other, shaping the future of our society. By understanding and addressing these trends, we can better prepare for the changes ahead and work towards a sustainable, inclusive, and prosperous future.

Looking at these six megatrends, we can only conclude that each of them will have an impact on us. Modeling a bridge, tunnel, or other infrastructure without society is not enough. We need to start with society at the center. And this is how we arrive at the term "social digital twin."

"Physical infrastructure on itself is not adaptive, so the platform around it must be adaptive to the ever-changing context." The context in which we will-- sorry, the context in which we work will keep changing, that's for sure. Social trends, but also the urban development of Antwerp and the political landscape, brings changes that we have to be ready for.

With all this context in mind, we have set a few goals towards 2050, "Fast Forward to 2050." "By 2050, we aim to eliminate unexpected tunnel closures." Predictability is key. We want to guarantee maximum availability. We want to achieve this by focusing on data-driven maintenance of the Oosterweel link and focus on maximum predictability and state monitoring based on data.

Moreover, by 2050, we aim to have tunnels operated autonomously. By focusing on data science, AI, and vision technology, we want to work towards tunnels where there is no longer a need for human interaction, for standard traffic management interventions, and for technical failures.

So what choices did Lantis make in their strategy? Lantis is a data-driven management organization and intends to remain so in the next phase, the "Operate and Maintain" phase. This means that we use contractors for almost all operational tasks. This also means that we do not need a full maintenance management system such as IBM Maximo. Our contractors will probably have this type of system.

Lantis wants to manage based on technical, economical, and social data to optimize costs and achieve maximum availability of the infrastructure. We focus on life cycle management.

We distinguish four domains to manage-- traffic management, operations, asset management, and, of course, safety management. We start at the top with traffic management.

By this, we mean the operation, monitoring, and control of traffic. This domain needs smart sensors, data analysis, and traffic management technology to ensure the smoothest possible flow. Next, at the bottom, we go to asset management. The predictable nature in terms of availability of the infrastructure should ensure maximum uptime due to the data-driven nature.

On the right side, operations-- the whole operation of infrastructure is more than letting cars or cyclists pass through a concrete tube. Last but not least, safety management-- traffic flow should not only be smooth, but also be safe. There are numerous use cases for smart infrastructure.

All together, they constitute the ecosystem, the Oosterweel link ecosystem as pictured in the middle, an ecosystem that needs to be relevant and responsive to the regional, social, and economical interests and needs. Each of the four domains require its own specific data streams, while other data streams are to be shared.

The challenge and power as an infrastructure owner is to develop the platform that supports all four dimensions and the connectivity of the system with all other ecosystems in the outside world that are relevant to the mission.

And that's the occupation of the infrastructure management team today, building these clear data pipelines and insights, not only from a technical perspective, but also from the other angles of the ecosystem. And therefore, we expect to employ more it and data specialists than technical engineers.

That's what we call "platform thinking." Given that there are no doubt many unexpected events and trends coming our way during the lifetime of Oosterweel link, we must adopt a flexible and modular approach, platform thinking.

Working data driven as an infrastructure owner is still a greenfield in several areas. On the one hand, Lantis has no legacy systems. So we have little to worry about in terms of compatibility. On the other hand, we find that the civil infrastructure in general is not yet very data driven.

Greenfield is a great freedom and opportunity. But it's also a pitfall if insufficient attention is paid to change management throughout the supply chain. It's also about taking bold and potentially wrong decisions.

Thanks to the platform thinking we apply, we can overcome these wrong decisions. It is important for us to be able to work with parties who value the same platform thinking and fail fast approaches. We found a partner in Autodesk next to Esri and Microsoft.

The technical implementation-- for a technical implementation that we will discuss, we are focusing on the domain of asset management, one of the four domains as an infrastructure owner, like we've just discussed.

So thus, why should an infrastructure owner take ownership for data management?

BAS VAN EKELEN: Thanks for the interesting presentation so far. To come back to your question, we need to be sure that data is interconnected. We need to be sure that data can be shared amongst ecosystems as Yanissa mentioned before. We need to be sure that we have all the data available, not only from a technical point of view, but, for example, also from a user's demographic or predictive point of view.

Therefore, it is important as an infrastructure owner to focus on the data, data standards, and conditioning. It's crucial to be in the lead of getting the data right. Otherwise, others will decide for you on the conditioning of it.

Another important aspect why an infrastructure owner needs to focus on the data is because of the fact that the data needs to be exploitable. Therefore, it's key to think about an asset-centric database instead of files and documentation. Nowadays, as-built is contained in the EDMS, mostly file oriented without a separate database delivery. But to start the asset management phase, it's key to be asset oriented instead of file oriented.

Let's zoom in to a concrete example to make it clear. Consider the fade out of the construction phase and the ramp up of the infrastructure ownership from a maintenance point of view. From the as-built, we need to create the as-maintained.

GUILLAUME NEYRINCK: Hey, Bas, quick question. What do you mean with the as-maintained?

BAS VAN EKELEN: Well, Guillaume, thanks for your question. The as-maintained contains a print of the as-built, but only containing relevant information needed for asset management. Asset managers face the challenge. There is a lot of information spread over many different sources originated in the construction phase so you have the as-built captured in the EDMS.

There's geographical information about the built assets in GIS. At handover, there's a baseline measurement with findings and captures in point clouds. There's 360 imagery together with drone photos to have a complete overview of all assets built.

And on the other end, you're preparing for operation, which leads to new sources of information. There will be different contractors for different disciplines and areas. They provide their management schedules. All events, issues, and configuration changes on the assets will be captured over time to build up history.

The question is, how can we connect this all in an accessible way? In recent months, we, as Lantis, did a proof of concept together with Autodesk to find out how to collaborate on setting up the asset information platform of Lantis. So Guillaume, from the as-built how to start the as-maintained?

GUILLAUME NEYRINCK: That's indeed the key question. And our approach is to unify all these diverse information silos into an asset-centric data structure. In simple terms, we'll organize everything around our central asset registry database. Now, while the data itself will remain in the systems where they work best, our key focus will be on connecting this data efficiently.

Our different personas will need tailored environments and tools, of course, to effectively do their day job. In order to set up for the future, we also need to anticipate changes over time-- both changes of the current data set, but also changes of the platform.

For example, the technical specs in the as-built will unavoidably receive an update in anticipation of replacements, which requires a full-blown CDN review process, just to name an example. It's just not the current state alone that we have to cover.

BAS VAN EKELEN: And here's something important to remember. We're building a platform for a constantly-evolving ecosystem. That means we need to be flexible and ready to integrate new data sources as they emerge.

GUILLAUME NEYRINCK: Let's zoom in on the as-maintained and how it is composed. Let's start with our asset register in the middle.

BAS VAN EKELEN: Guillaume, thanks for porting out to asset-centric data management. We like to use the metaphor of a tree to emphasize the importance of OTL for data structuring. The Object Type Library, the OTL, is the fundament. The basics is standardized data structure for consistency and interoperability represented by the roots of the tree.

Second, there is location at the landscape. It's essential to know where an asset is located in the Oosterweel link. This is represented in GIS. On the next layer, relevant building information visualized, represented by the trunk of the tree.

Layer 4, well being of a tree can be recognized by its leaves. This is where the maintenance takes place. Maintenance will be organized by our contractors, as Yanissa mentioned before. Our contractors will use their own maintenance management tooling for this.

And then the top layer is the environment. We will have to unlock and collect a lot of information about our assets from the environment to utilize their full potential. Connectivity with multiple systems and partners during the life cycle of the assets is an ongoing topic.

What would be the first step to create the asset-centric database based on OTL? In order to do that, we need to have a look at the setup of our registry. We're doing this with ACC build assets. And we are leveraging 3 ACC Connect recipes to create this repository, as per the OTL standards set out by the regional authority.

The very first step is to create your categories, or classes. This will also define the hierarchy and the permissions that are given to our stakeholders. Secondly, we will create the custom fields or the custom attributes in bulk. For this DOC, we already had over 800 custom fields. And we need to pay attention, of course, to their correct settings and their types.

Next, we then link steps 1 and 2 together by assigning the right custom fields into the categories. We use the ACC Connect script for this via a join, of course, because this is quite an important part of the setup where you can link the classes and subclasses and then define which exact parameters are given to that class or inherited to the subclass.

As you see here, we use a low code environment for that, which is ACC Connect, which makes it quite easy for all our teams to quickly integrate these systems. And the last, but most important, step then is to then populate all the data of the assets and the metadata into this OTL structure as you see here on the screenshot.

We then have the relevant fields available to us to create this asset either manually, through import of a model or Excel or via the APIs. They are then accessible in the background of this screenshot where you can see all of the parameters listed in the background.

GUILLAUME NEYRINCK: Setting up the asset register on OTL standards is indeed the first step to take. Secondly, the handover as-built contains 2D drawings, technical documentation in PDF files, and as mentioned before, our future handovers will also have 3D models because those projects are BIM based. It's crucial to link all these relevant documentation to the right assets in the asset register. The question is how to link the as-built.

BAS VAN EKELEN: Then, we have a look at how this is linked and how this connection is established and maintained. The first step here is when we're navigating in our structure is to select, of course, the assets that are relevant. We can multi select and then create a reference in ACC to another part of the system. In this case, we reference to the document management system, which is, of course, based on a folder structure.

This is where the as-built files live, either BIM files or 2D, like Bas mentions or technical specs. It's essential to understand that managing these as-built files involves more than just attaching static documents in an archive. We anticipate replacements and updates. And the technical specifications need to be revisions reviewed and approved.

It necessitates a robust CDE including a full destruction, its own permission, naming convention, and versioning. This will allow us to maintain an accurate and consistent audit trail over the years to come.

GUILLAUME NEYRINCK: Next, in addition to the asset register, the primary entry point for stakeholders can also be the GIS environments. Whether we want to scroll through a list of assets based on the OTL structure or we want to navigate a 2D map, which is quite intuitive, it's our choice. The platform here allows us via a strong partnership between Lantis, Esri, and Microsoft to provide this ecosystem, this platform.

This is the GI environment designed to provide stakeholders with the intuitive access to spatial data and asset information. By integrating these systems, we ensure that everyone involved has comprehensive map-based view of the assets, making it easier to visualize and interact with the data.

For the average high-level stakeholder who may not be familiar with CDE folder structure or the asset codification structure, the GIS environment serves as a primary entry point. This user-friendly interface simplifies access to the crucial asset information, presenting it in an intuitive map-based format. It allows the end stakeholders to then navigate to the right model and find the relevant layers of information.

Next, we'll have a look at the top-left part of our schematic. And as Bas mentioned, at handover, we're capturing baseline measurements using point clouds. These point clouds provide us with a baseline of geometry and also with the issues at handover.

The connection with the asset register is not included in the environment itself, but it is connected in the ecosystem. Matterport, in this case, shines in capturing the point clouds. But it integrates with ACC through the issues.

In ACC, we can then-- ACC allows us to assign, distribute, and track the issues. Additionally, we'll also be able to connect the relevant issue to an asset which allows us to find it back later, coming in from the asset registry or the GIS environment.

The owner will then take full responsibility of the issue post closeout and post handover, reinforcing the message of Yanissa earlier in terms of why it is important for the owner to have a strong digital vision.

So as we just said, our point cloud is connected through the issues to the asset register again. Next, if we have a look at the top right, we'll see our 360 images hosted on SharePoint.

How do we access those? Well, if we have a look at this part of the platform, we'll again see our GIS environments. And here, we don't have a direct link to an assets and the image. But rather, it's given a GPS coordinate which allows you to interact with it as easy as selecting an asset.

Place yourself on the map. And then you are given access to very rich 360 images. We want to ensure that these assets are regularly photographed to give the contractors an option to quickly get familiar with the street views before mobilizing on sites. This gives them access to the location and gives insight to the surroundings.

In addendum to the GIS environment, which is asset based, the 2D maps includes the geolocated pictures itself. But they're not directly connected to it, as we see in our schedule. But this brings us to a major question.

BAS VAN EKELEN: Yeah. Thanks, Guillaume. Recapping, we have an asset register based on the OTL structure interlinked with the build files, point clouds, images, issues, and 360 images, as you just mentioned, all made accessible through the GIS. We are now ready to collaborate with the contractors.

Most of the data and information about the assets in the as-maintained is coming from our contractors. It's important to have history of all events issues and configuration changes resulting from the collaboration. Up-to-date and correct information is maintained is important for the contractors as well. We aim for a self-service platform that allows contractors to collect the information they need to fulfill their jobs and enrich the information with their findings, keeping an audit trail.

My question here, how to collaborate on the as-maintained with the contractors.

GUILLAUME NEYRINCK: That will be answered by looking at the bottom part of our diagram. We'll start with our contracts because this is the first step we'll take to collaborate with our supply chain is to have a contract in place. And the contract will include the cost allocations, a schedule of values, and, most importantly, the references to the assets that are covered in the contract scope of course. This setup provides us with an immediate transparency into how the costs are distributed and which assets are included.

Next, we anticipate, of course, that this contract will have some changes because things happen throughout the years and we'll have worked on these assets, which was not covered under this contract. So if we start with our assets under the contract like this fire door, we can easily see what contract has been attached, but also an add on, a new potential change.

This is shown here. If we continue with our fire protection, we can see that the asset selected can now be given a reference to a potential change order of a cost module, which is created on the fly by the relevant stakeholder with permissions, of course. This potential change order allows us to cover a complete process as well. And the contract link allows us to track the change order and cost variances that arise during maintenance in a structured way.

This connection not only helps us manage the initial budget allocations but also provides ongoing visibility into how changes and unforeseen costs affect the overall financials. Ensuring more accurate budget management and financial tracking throughout the lifecycle is key. We aim to develop a heat map at asset level of the maintenance costs to identify the areas where the investment yields the highest return.

Continuous budget and cost creep during maintenance shall then influence the prioritization of the maintenance activities, which is taken into account into next year's maintenance schedule. This brings us to the next-- this brings us to the next part where we include the schedule of the next year.

Owners can access the plans, activities of their contractors. They can then regroup those based on the zones or the assets or any metadata or custom attributes included in the plan. Most importantly, we can then start a review and suggestion process to validate the planning that is provided by the supply chain.

Importantly, we also need to think about the versioning of these plans so we anticipate new versions to come in. But we, of course, want to keep the references in place.

This is where it comes in handy because the link between our activity and our assets is made and maintained. As you can see here, I'm back in my asset registry by clicking through from the schedule activity. If we go back into our schedule activity from the assets repository, we can then also see the full history of this event.

This history is quite important, especially if we look at comparing the yearly baselines that are provided by the supply chain. As you can see here, you can connect, but you can do compare of three years worth of baselines to see where the changes happened and which assets they influence, of course. A major prerequisite here that we learned about is that we need to have the asset ID as a prerequisite towards our supply chain schedules.

But then, execution of all of this planning will lead to issues and maintenance events. Those will be locked onto the assets. It's then, of course, important to capture all of this rich metadata in a complete audit trail. For the assets, we, of course, track the status changes, as this will be a major KPI to track the operations of our assets.

Next, we'll also track the metadata fields, like the custom attributes of the OTL structure and how they change over time. When it comes to issues, we, of course, have a robust activity log in terms of when they were created, how they were resolved, and all of the changes that occurred in the meantime.

Of course, it's as important to have a robust view of other areas of our system, like, for example, the CDE environment. Many of you know, looking at the left part of our screen, that the CDE environment has quite a robust version history, giving you the metadata fields of each version, but also the content of each version, allowing you to compare what has changed over time.

On the right side, we see an example of the budget compare feature because, of course, on a yearly basis we'll have to accommodate those changes as well.

And this provides us, of course, with good insights looking backwards. So all of the modules that you see here in front of you have their own audit trail. But we also need to think about the future. I'm going to hand over to Bas now to talk about the bottom-right corner of our diagram, looking at how we will future proof the platform.

BAS VAN EKELEN: Thanks for the demo, Guillaume. Correct. So in the demo, we focused on the as-maintained and maintenance related to asset management. But the as-maintained, including all previous building blocks is, of course, to be enriched by real time IoT in the future. So for example, the monitoring of the structural health of a tunnel to be able to better predict the life cycle of the tunnel or monitoring vibrations of ventilation systems to be able to anticipate and prevent failure of the systems.

But if we zoom out to the four domains as an infrastructure owner, there's more to come. Focus is not only on asset management, but on overall infrastructure management. We would like to gain insight in alarms and signals related to traffic control and safety. We would like to gain insight in our contribution to congestion reduction. But we would also like to gain insight in impact on the environment or our sustainability performance.

And all of this together makes a step towards a digital twin.

YANISSA DE JONGHE: Thanks, Bas and Guillaume. That brings us to the conclusions of our case. So we would like to pass along three major takeaways. First, apply a user-centric approach. Emphasize putting the user of the infrastructure at the center of all decisions and design. Focus on the ever-changing needs of users.

Second, be bold. Dare to innovate. Encourage being bold and making mistakes as part of the learning and innovation process, creating a risk-taking culture. Foster an environment where taking calculated risks is encouraged and failure is seen as a learning opportunity. Promote a culture of experimentation where new ideas can be tested and iterated upon quickly. Let's say create a culture where people feel safe enough to try.

Third, apply platform thinking. Adopt platform thinking to create adaptive and resilient infrastructure systems. Design systems in a modular way so that components can be easily updated or replaced without disrupting the entire system. Ensure that the infrastructure can scale to meet growing demands and adapt to changing requirements. Build systems that can easily integrate with other platforms and technologies, enhancing flexibility and resilience.

In our opinion, these principles can help create a robust, user-friendly, and innovative infrastructure that can adapt to future challenges and opportunities.

Would you like to know more about the Oosterweel link? You can use this QR code to watch the full video of the B1M. Thank you very much for your attention. Do not hesitate to contact us. We hope to have inspired you with our case study. Thanks.