Monitoring Railway Infrastructure Sites Through BIM and GIS

STEFANO LIBIANCHI: Hi, everyone. It's a pleasure to present this class with my colleague Catello Cascone. I'm Stefano Libianchi, beam expert in FSTechnology. And in this class, we are going to explain how our workflow can improve the monitoring of railway infrastructure site through BIM and GIS.

Here is the agenda for the presentation. Just a few words about us, then we start to describe the workflow and see the use case, after show the outputs and the conclusions. And, now, a brief introduction of our company and our team. FSTechnology is the high tech company of the Ferrovie dello Stato Italiane group.

It was created at the beginning of 2019. And its goal is to strengthen and support digital innovation among the company of the group. The BIM and GIM Competence Center is a team within FSTechnology. Considering the core processes of the group, we mainly support linear infrastructure projects. And, therefore, we support Italferr, which is the engineering company of the Ferrovie dello Stato Italiano group, during the design and the construction stage from conceptual design to handover.

We also support Rete Ferroviara Italiana, the company owner of the entire railway network. And here in this slide, we can see our team. The main objective of our team is to research and implement new technologies to improve the processes and the workflow for the management of the entire life cycle of infrastructure projects.

Our first class was presented by Marcella Faraone and me in 2018. And since then, our group started to investigate how to better integrate BIM NGIS with other platforms and implement a solution for remote site monitoring. This year, in addition to this class, we are going to present a second one, how to avoid wasting water and energy with the help of BIM NGIS.

Before describing the workflow, I would like to explain the reasons why it was designed. Ferrovie dello Stato Italiane group is the leading energy consumer in Italy with a slice of around 6% of national demand. And, therefore, an investment of 1.6 billion euro will be allocated to plants installation to self-produce energy, amounting to 40% of the overall consumption of the group to be achieved by 2027.

After this date, we will continue with the reduction. The FS group believes that an organization long term success is built on a strategy that prioritizes the protection of natural balances. The FS group's greatest contribution to the creation of an environmentally-sustainable development model is that it offers increasing more efficient and sustainable transport services that maximize the benefits of collective mobility.

The workflow aligns to the objective of FS group, incorporate the protection of the environment into its strategies and priorities by promoting and implementing a more rational use of resources, the use of renewable energy sources, and the prevention of reduction of environmental risks with the aim of gradually reducing the group carbon footprint. To be able to do this, we developed different workflows that respond to specific needs but originate from the same data, which, as we see, are centrally stored on Autodesk Construction Cloud platform.

It is important to highlight the added value and relevance of standardizing and automating of processes with the help of the leading edge technologies. This is the focus of our work, set up a workflow with a continuous fine tuning depending on the outcomes achieved from the tests. So we set up this workflow based on an integration of Autodesk software like Autodesk Construction Cloud, Revit, Navisworks, and Civil 3D with Esri, ArcGIS, and GeoBIM to actively support the monitoring of the environmental impacts.

It all starts with a survey. The data is saved on Autodesk Construction Cloud and [INAUDIBLE] we would notice the central role of the [INAUDIBLE] environment is CC. From the CC, the data is read to open the linear design with Civil 3D and the design of functional element for Revit but also to create collaboration with Esri, ArcGIS. The data prepared on Esri, ArcGIS is used to create useful checklists in workflow for both maintenance and construction works on site.

Also, I like the central role of the external database in which we stored all the valuable information coming from the model and the data retrieved in the field with the checklist. The data from the checklist, therefore, use it to complete the modeling with the information taken from the sites. And in the case of water design, we create an automaton with Dynamo and Civil 3D that adds information to specific assets, for example, on the pipes.

This part of the workflow was presented last year in Autodesk University. And we prepared an automatic procedure to assess, estimate the physical progress of the works inside the construction site using the survey deliverables and an ad hoc simplified model that represent the progress of the work using Navisworks. This reaches the data with further information that we will need in the future. It is also overlaid, can be viewed and analyzed through ArcGIS Pro webmap or inside GeoBIM.

And on the field with the issue workflow in SEC, show the nullity for safety purposes or use it with augmented reality and virtual reality in the field. The use of artificial intelligence can complete the data of the dashboards as well as IoT with TANDEM.

We focused our energies on the possibility to reduce time on construction site management activities. We will explain where we started and what we have achieved so far. In July in 1996, the European Commission adopted a resolution to implement the Trans-European Transport Network.

The intent of this multi-phased project is to provide coordinated improvements to primary roads, railways, inland waterways, airports, and traffic management systems through the Europe. When complete the Scandinavia Mediterranean corridor of this project will stretch from Helsinki, Filandia, to Valletta, Malta, and the Napoli-Bari high-speed railway project is part of this corridor and start in 2015.

Now, let's explain the workflow in more detail. The starting point of our workflow is the survey but from drone and ground. This short video shows the main construction site area of the works on [INAUDIBLE] Frasso Telesino Railway. We regularly surveyed the site roughly every two months to monitor the construction works, including two viaducts, [INAUDIBLE] system, and tunnel.

And all project file are shared and stood on Autodesk Construction Cloud within folder structure that reflects the work breakdown structure of the project. At the top level, there is the project phase, for example, design phase, construction phase, and main disciplines, civil works, technologies, environment. The main overview, WBS elements, viaducts, tunnels, retaining walls at a lower energy level. We also find the surveys organized by survey date. And at an even lower level, we have the output listed by type or format. And now Catello will continue with the explain of the workflow.

CATELLO CASCONE: Once all the project file are connected, we can reach the scene with all the available base data taken from the different source. This is an example of the data.

The data related to the diagnostic investigation are stored on a group owner portal named Sigmap. Sigmap is a site made up by a DB and a GIS application.

In this specific case, the GIS is a web map that displays the input data provided by the DB part. By accessing the Sigmap, we can have three types of user, owner, same player, who enters the result of the monitoring activities performed on site or organization from the validated environmental monitoring work. By accessing the Sigmap through the username and password, we only have access to specific section according to the user type.

Let's have a look in detail. By looking in, for example, to sigma co-monitoring, with my user, I can only see the project I have access to. And by selecting Cancello Frasso, I can see the atmospheric components granted by the permission of setting.

If I choose groundwater, I can see the master data sheets over the sites being monitored or enter new ones. Each master file provided the general site's data and associated measures, which are under a validated flow between the owner and the supplier. In addition of this information, documents and images useful for monitoring the sites can be stored.

Accessing the GIS selection is [? standard. ?] Once in the project I am interested, I will have this type of visualization. Data subject to environmental monitoring, such as roads sites, area, and the road access are referenced on a base map. Zoom in in the site interest.

This site offers design, the interest possibility of insert external data, such as the data made available by ArcGIS or to be able to verify the design against environmental constraints or personal data in KML or shape format. To be able to compare them with this publisher, it's possible to change the cartographic background by selecting the satellite view to view the accuracy of the placement of the monitoring points in the countrysides.

If there are comments to share, we developed a [INAUDIBLE] with the widget to allows the user to comment on the map using a markup tool and text boxes. Once the comments has been added, it can be downloaded locally in the JSON format and sent to the supplier or to a colleague who will be able to reupload and view the comments back in the browser based application using the same widget.

The environmental and archaeological constraints, sensor data come from the Eva portal. In this portal we can see all the archaeological discoveries near the railway project. For example, here we have the area with the high potential archaeological risk in red.

We have a database containing all the information classified by type of archaeological finds. [INAUDIBLE] designs must take into account the classification of a potential archaeological risk in the area of interest. Now we are showing an example of a natural asset, preservation validated with the environmental inspection during a construction. We analyzed the [INAUDIBLE] of the environmental system to verify whether the natural asset were preserved and ensure to project of our landscape heritage.

Using the ArcGIS Pro, we can georeference the asset. Let's see how once the vegetation present in the area of interest has been referenced and classified, we obtain a map of the vegetation near the construction site. On the left, we have the 2D map with the classified point elements, while on the right the 3D scene. In the 3D scene, you can see how the data has been given a realistic symbol, which allows the territorial reality to be digitally reproduced.

From the BIM model, the starting point is the BIM Cloud Connection Connector that combines the data suite and the Esri suite. We have the possibility to upload on the map the most up data version of the BIM model of interest available on Construction Cloud directly into ArcGIS Pro. Once the model is placed on the map, we can navigate it. Select it to turn on and off the layers.

This allow us to see the model as a whole or look at specific elements in more details. We can then explore the spatial context by adding layers to the map. We change the base map, selecting the most suitable among those available in the catalog. In this case, we decided to use the satellite map. Once again, we add the layers to the scene to visualize the model within the spatial context with the build and the natural environment.

We added to the scene a reconstruction of a building, the curve of the construction site and the area of construction site, the tracks and the location. From DTM, we have developed a workflow which allows us to have a very high definition of accuracy. Starting from the same point cloud, we classified it by adding material information to tell concrete from steel, timber, as well as ground from the vegetation.

The first workflow we created is used to calculate, cut, and fill earthworks volumes. And first step of this workflow is about the necessary data of the processing that we will that we will then use in Civil 3D for an automatic calculation of cut and fill volumes. To have a usable result and limit processing times, we had to decimate the input data by using the tool that reduce the density of the point cloud, preserving the geometry as well as possible.

Therefore, in Civil 3D, we import the dams generated from Python scripts loading and the reference surveys generated from the most recent survey. Next, we load the surface of the same area and survey that the previous time-- specifically, the surface to be compared must be named the same plus a suffix indicating the survey, T1 or T2, for example.

Running the Dynamo script, it's detected the dams and in Civil 3D files lists all the surfaces, sorting and pairing of them by name. The script created the volumetric surface from the period dam surface. The data extraction process catches the name of the surface to compose the name of the process volumetric surface so that I will get the same name as the control once and get both the name in the survey comparing a suffix [ITALIAN].

We, therefore, use the solution for semi-automatic volume calculation. We have the possibility to using ArcGIS Pro to get the output for Civil 3D through the dam processing script from the point cloud and the ship pilot to calculate excavation and filling. This help us to make our digital workspace as realistic as possible, allowing us to create a dimension where we can carry out different types of analysis.

Another type of information that we use for our analysis are the building in the area of interest. We use our CT engine software from the Azure suite. We have the possibility to select the quality of our final product and identify the area of interest.

Here, for example, we are selecting Las Vegas. We can identify what to acquire. Here, we select buildings and roads. The result is a base map of the area with the selected data in 3D format. Then we can use a texture to make the buildings look more realistic. Data can be routed or moved. We select those relevant to us, and we can export them to then load them into the GeoBIM web platform, which brings together [INAUDIBLE] works.

The cadastral data are another set of information that we used. From the internal database, we can view, identify, and acquire the plot of land owned by the Ferrovia Group. Then we can use the national database to acquire the cadastral limits of the data plots. The result is an overlay of cadastral information that allows us to have a total picture of the area of interest, identifying which lands can be immediately incorporated with the construction site and which cannot.

We reach our BIM model with more information. Here, for example, we added the information from the construction program. Using the joint tool identifying the right case, we added the time information of start and end of construction on the specific section of the model. Let's see additional dates on the piers then on the bridge deck or on the pier cap.

We can then use the time feature of ArcGIS Pro to view the model according to the datas defined on the time schedule. We see that, in fact, the portion of the model not yet building is not displayed in the scene. Reactivate the navigation of the underground model. We can see by portion of the model, that this is covered by the ground. In particular, here we see the support of fields and planes.

By activating the time slider, we can see the different stages of construction on the BIM model. We have first the visualization of fields then of the planes. Analyzing at the scene with the ground, then we can see the effect of the time filter for the entire work. We can also compare the point cloud acquired on the site with the BIM model to analyze the progress of the construction site in relation to the time schedule.

We then move to the publication of our content. We decided to create the application with the GeoBIM and the shared data for analysis. We published the model containing the time schedule using the following workflow. BIM filed to the geodatabase, make building layer, and, finally, create building scene layer content that allow us to view the model on the Esri online platform.

Once we have acquired the data of interest for the area under examination, we can import them all into a single scene and proceeded with our analysis. We use GeoBIM where we see a 3D scene with the dwellings, ground beam model of the viaduct, and the classified vegetation also referenced. We can also activate it on the ground navigation to see the viaduct in its entirety.

We have identified the viaduct construction site area in orange and the area near the tracks in white. The point cloud has been uploaded. We have added some markers to our scene that lead back to diagnostic investigation portal. We can use them for monitoring the surface and the water example of data got from a pitometer.

This information on the presence of historical aqueduct was acquired from the Eva portal. By integrating the information retrieved from this portal, we can monitor and therefore better preserve the archaeological asset, such as ancient Roman roads and/or aqueducts close to the tracks. With the land register data, we can identify the best area to plan the installation of photovoltaic plants close to the track. With the land register data added, we can also use the shadows tool to perform shading the analysis, identify the most efficient location and orientation of the photovoltaic panels.

Here we see the use of the tool with the time slot as a variable and the shadows in red. By using all this data together in a 3D scene, we can make analysis and obtain results that help us to our choice. Here, however, the map contained in the GeoBIM web application-- browsing the map and zooming in on our construction site, we can observe the functionality of the time slider.

With a simple preview, we can see the work appearing according to the information we got from the construction program. We then use the layer list to show how high the portion of the model or show the entire model, just as we did earlier with the desktop application. We also have a dynamic legend that show us what is displayed on the map.

With a click, we switch from a 2D visualization to a 3D one. We can then navigate to the scene and see in transparency even the portion of the model placed below the ground level. GeoBIM offer for new tools, the ability to navigate the scene at a different time, the ability to choose the view whether we prefer, the calculation of the shadows, and the calculation on the viewing accounts.

Thanks to publication as a building layer, we can use the 3D Explorer tool to see the BIM model and the information it contains, just like the desktop tool. Among other applications that can be used at the 3D scene, we have the issue tool. Here, for example, we see in white the client's area of the tracks, which indicates the safety distance to be respected during the passage of trains. It's overlapping with the construction site boundaries in orange.

Using this tool, we can fill out a form with the user full information, name of issue, status, type, causes, and description.

Once sent, we can see it in the documents in the application and review the summary table. We can view in a BIM 360 the created issues. We are seeing in a start and target data to the resolve the issue and the user who has to manage it. Finally, we publish it. As you can see in the list of activities, there is the list of activities issues with the uploaded information.

Using GeoBIM we also have the possibility to select or to select the information visible in the scene. Let's select, for example, the layer of activity construction site. We can see at the same time the feature and the [INAUDIBLE] containing the design of the construction site. We then reach the application with the point cloud representing the physical progress of the work. We can then overlay the BIM model and the point cloud inside the build and the natural environment.

STEFANO LIBIANCHI: Thanks to Catello, we have seen some of the possible analysis using ArcGIS and GeoBIM. Now let's see the other results that we can obtain with the other workflows. The first output that we quickly show, we got by combining the project data and the artificial intelligence analysis, the orthophoto taken on site.

In this video, we can see the elements identified for a given survey, for example, plants, tracks, structural peels, and that by choosing an element, in this case, tank. The dashboard updates to show all the AI images where this object was recognized using the algorithm.

With the use of Unity, a cross-platform game engine, and ArcGIS SDK for Unity, we managed to integrate GIS data and BIM models. Resulting is a solution very simple to navigate and enjoyable by all known BIM experts. And what you see in this video is a simulation of some potentially dangerous situation while moving inside a building and in the construction site. Another output that we are going to describe is augmented reality. And as we can see in the video, it is possible to exchange information remotely or open the documentation needed to carry out the checks of various kinds.

One step forward we are taking is approaching the digital twin. We are studying how to use TANDEM and the potential. We are currently testing it to integrate the information coming from an IoT sensor and field service information to have an overall view of the data. In the future, we hope to be able to forecast and anticipate problems before they arise.

Here we are looking at how we can put together the data collected through 3D scene of ArcGIS Enterprise. We are able to have a territorial context enriched with data such as a building, roads, vegetation, and navigate the scene even on the ground to compare the differences with our model with the context. Now, the output is this dashboards built with an Esri application of ArcGIS enterprise.

In this video, we are looking at an example that shows the construction data of some sites. The dashboard is made up of various interactive parts that allow the user to query data for different purposes. In the lower section, also, there are graphs that allow the user to see the consumption trend of the sites.

In conclusion, let's go back to the strategies seen before and show how we can intervene with our workflow. Naturally, to solve the protection of environment and water reduction, we use artificial intelligence and TANDEM to have an overall view of the information.

While we use GeoBIM, as [INAUDIBLE] does-- for example, to position the photovoltaic panels in the best areas and the augmented reality and virtual reality to reach dangerous area-- moreover, we can say that with our workflow using an integration oriented approach, it's possible to achieve many benefits, work in common environment, integrate platform, use integrated methodology to have fewer errors, and use Digital Twin and utility network tool. Thank you. And I hope you enjoy the presentation and our work. Thank you for the watching.