## Predicting pile driving times

For many buildings in the Netherlands, a deep foundation is required to make sure that the construction load can be transferred to lower and more stable ground, as massive soil layers that can absorb large foundation loads are often located deep below the surface. The foundations of large structures, such as distribution centers or traffic overpasses, can often require the installation of thousands of piles at these depths. For both cost estimation and planning it is important for [Voorbij Funderingstechniek](https://www.voorbijfunderingstechniek.nl/) to be able to estimate how much time this process requires.

Currently, estimates of pile driving times are done based on information about the soil composition (sand, clay, etc.) of a location, obtained from CPTs. Because the registration of pile data (pile measurements and CPT data from customers) has recently increased significantly, it has now become possible to make (and improve) predictions of pile driving times for future projects by applying machine learning techniques to the available data.

## Machine Learning

Part of Voorbij Funderingstechniek’s service is their ability to invest their time efficiently and predict the progress of projects as accurately as possible. By applying machine learning to the CPT and pile data, they can improve this process – and by that their service as well. 

### Database

Pile driving speed mainly depends on soil properties. By applying a machine learning approach, it becomes possible to predict pile driving speeds. The only requirement for this is that sufficient pile-driving and CPT data has been collected (at the right depths). For this reason, it is very advantageous that Voorbij Funderingstechniek has their own database, called ([Base FT](https://www.voorbijfunderingstechniek.nl/expertise/advies-en-ontwerp/base-ft)), for logging all project data, including pile driving and sounding data.  

### The model

The machine learning algorithm that Voorbij chose for this is a supervised machine learning model that uses an artificial neural network (ANN) to predict pile driving speeds based on CPT data. The algorithm is implemented within the VIKTOR platform, where it is trained on previously measured pile driving speeds and the associated probing data. During the training of the model, the ANN adjust itself in such a way that it can predict the correct pile driving speeds with data from CPT values. In the VIKTOR app, engineers can use training data to train the machine learning model themselves and apply it to projects.

### The application

Voorbij used their training data so that the machine learning model can be applied to two relevant real-world scenarios within their VIKTOR app. The two different scenarios, which will be explained further below, yielded different results in terms of accuracy. The predicted pile driving times were compared with the actual recorded times to determine the accuracy.

#### Scenario 1: Individual projects

In the first scenario, Voorbij first presents the model the results of a test dataset at the start of an individual project. Then they use some pile measurements (10%) as a training data set. Based on this data set, the ANN can then predict pile driving speed for the other piles (90%). The predicted pile driving speeds adapt in this scenario to the speeds of the test data set and can be used to improve the estimation of the total project time. After training, the model can predict pile driving speeds at certain locations as a function of depth at the locations of the test datasets. For example, for one construction project, the model was able to predict the pile driving speed on the test site set with a margin of error of only 10%.

#### Scenario 2: Multiple other locations

In the second scenario, the machine learning model is trained with a complete dataset of pile driving speed measurements from one or multiple other locations. The model also receives a test dataset with pile-driving speed measurements from yet another location. These predictions can then be used to estimate total project time even before the project starts. Because in this scenario the predictions of the model are adjusted to the measurements of the training dataset, this caused the performance of the model to decrease significantly compared to the previously mentioned scenario.

The results of the predictions of the total pile driving times in both scenarios are shown in the graphs for two separate projects with different soil compositions.

![pile foundation application](https://cms.viktor.ai/uploads/AIML_10_de922172d2.svg)

</br>

![Pile foundation application](https://cms.viktor.ai/uploads/AIML_11_ba985d7abf.svg)

*left: scenario 1, right: scenario 2*

</br>
</br>

The graph to the right presents the velocity measurements of the less accurately predicted project (scenario 2). Lack of accuracy can be explained, among other things, by the fact that the speed may or may not have been equal to zero at certain moments. The pile driver cannot register this, so the number is rounded to 0 meters per second. Evidence for this can also be seen in the scatter plots. These show the registered pile driving time against the calculated pile driving time. The black dots indicate a speed of 0 meters per second or more. The green dots indicate a speed of 0 meters per second.

## Using the app yourself

An open-source sample version of [the machine learning application](https://www.viktor.ai/blog/63/pile-drive-prediction-with-machine-learning) is now available as well. With the app you can predict pile driving times based on the entered GEF files by applying one of the two machine learning models. The app automatically generates visualizations of the entered GEF files for inspecting the soil structure, classified using the Robertson method. In addition, the app generates force estimation plots as well as a map view of CPT locations.

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/690862336?h=97c73422c7&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Sample app Pile Drive Prediction with Machine Learning"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

</br>
</br>

Using the app is simple: in only three steps you can predict pile driving times with one of the two machine learning models. 

1.	First, you upload the GEF files. 
2.	Second, you define the parameters.
3.	Third, you check the automatically generated results. 

The app can be used with or without VIKTOR. For those who want to implement the functionality themselves, the open-source code is available via the [GitHub environment of VIKTOR](https://github.com/viktor-platform/sample-machine-learning-pile-driving). You can try the app directly via the [VIKTOR demo environment](https://www.viktor.ai/demo-environment) or in your own VIKTOR environment by using the [free version of VIKTOR](https://www.viktor.ai/try-for-free).


Infrastructure & Construction

small_voorbij ft richard klein.png

thumbnail_voorbij ft richard klein.png

voorbij ft richard klein.png

See how machine learning is used in a VIKTOR app to predict pile driving times. Read more on VIKTOR.ai

Machine learning to predict engineering processes

Predict pile drive times with machine learning

1. An introduction to Artificial Intelligence
2. Artificial Intelligence vs Machine Learning
3. AI and ML in organizations
4. Building a Machine Learning application
5. About VIKTOR

artificial intelligence AI machine learning ML engineering construction

medium_A4 - White Paper COVERwp ai cover.png

small_A4 - White Paper COVERwp ai cover.png

thumbnail_A4 - White Paper COVERwp ai cover.png

WP_green

artificial intelligence machine learning engineering construction

large_artificial intelligence machine learning engineering construction

medium_artificial intelligence machine learning engineering construction

small_artificial intelligence machine learning engineering construction

thumbnail_artificial intelligence machine learning engineering construction

Blog_AI_and_Machine_learning_engineering_construction_1

What exactly is artificial intelligence? 
What are the four types of AI? 
How does artificial intelligence differ from machine learning? 
And what are the different types of ML?

In this White Paper, we will answer all of these questions and more about the topic.

Find out how you can use AI to improve and automate (engineering) processes within your organization and learn how to build a machine learning application yourself as we guide you through the following chapters:

1. An introduction to Artificial Intelligence
2. Artificial Intelligence and Machine Learning
3. AI and ML in organizations
4. Building a Machine Learning application
5. About VIKTOR

Find out how you can use Artificial Intelligence and Machine Learning to automate engineering processes!

Download the White Paper and get INSPIRED

Artificial Intelligence in Engineering and Construction

## Automating the design process

The [parametric design](https://www.viktor.ai/blog/46/parametric-design-cloud-engineering-construction) tool allows J.P. van Eesteren to generate the optimal designs of residential towers: As high as possible while using as little surface as possible. Visualizations of the parametric models are shown on a standard map in which all kinds of elements, such as apartments, elevators, and staircases can be seen as well. It is easy to adjust aspects such as dimensions or to add elements like additional walls by simply changing the input for the parameters. The impact of the design changes is immediately shown on the map. 

See here how engineers at J.P. van Eesteren use the application to create the optimal design for residential towers:

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/670120716?autoplay=1&loop=1&autopause=0"h=7298f72eb4&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="J.P. van Eesteren - LinkedIn demo (1 min 25 sec)"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

> “It is a design tool that can be used to draw and calculate simultaneously. Regularly, we come up with an idea from which the architect creates a design. Naturally, this is not immediately sufficient. So we start calculating again, and again, and go on. With our application, we can combine drawings and calculations, which means we can iterate much faster and cheaper!” – Max de Vries, Project Developer at J.P. van Eesteren

## Securing rules in Python

How did they establish the parametrization of residential towers? The answer is: By securing rules. According to Max, digitizing is all about rules: “These days, there are many requirements and rules that need to be tended to. What we need is a system to secure them. When we developed our tool, we started looking for ways to incorporate all of these rules, both written and unwritten. If we were not able to program them, we were not being sharp enough.”

For each project, a residential tower is put together in the same way. Nevertheless, there is an entire design process that precedes the construction every time, and every time it ends with the same result. That is why engineers at J.P. van Eesteren figured that securing rules for the automation of this process was the most logical thing to do. Now, their work starts with a result: The design of a residential tower. What follows is the optimization of these designs.

## Unique designs

A lot of people think that standardization leads to the mass production of a lot of identical-looking residential towers. However, J.P. van Eesteren has a whole other opinion about it: 

> “Every person is unique. There is a certain structure in every body that contains a skeleton, heart, liver, and go on. These are all kept in logical places for a good reason. If we start shuffling them around, something will go wrong. The same goes for residential towers. There are certain parts of a them that are unique, the same as with a person: The length of the arms and legs, hair and eye color, these all grow differently and can be changed if needed". - Max de Vries, Project developer at J.P. van Eesteren

With their application, J.P. van Eesteren has found a way to secure certain parts (the skeleton, organs) while also having the freedom to change other aspects (height, hair and eye colour).
 
## Transparancy

For the engineers at J.P. van Eesteren who use the application, it's important that their tool is transparent. Instead of using a black box that simply tells them whether a design is feasible or not after inserting certain parameters, they want to be able to know the ‘why’ behind it all. 

> “We have specifically chosen that the person who is using the application is also the one that is really behind the steering wheel.” – Max de Vries, Project Developer at J.P. van Eesteren

Because their application is so transparent, engineers can guide the project in the right directions where and whenever needed and decide for themselves what the best alternative would be in certain situations. This way, they can first tell the architect about their rational approach, only after that he starts making it into a fancy residential tower. 

Max: “Once we combine our rational approach and their design approach, we can review whether the tower can be made into a feasible design. Because the architect doesn’t have to think about the rational approach anymore, we gain a lot of time. The building is like a body: We make sure that the organs are there and in the right place and the architect decides on the length, eye- and hair color.”

## Realistic environments

A nice addition to the application, is that J.P. van Eesteren is not only able to quickly visualize the design of their own projects but of the surrounding environment as well. First, they place a building on the map by selecting a location and defining the area. Then, they can select the surrounding buildings and visualize these as well by using data from the open-source data model [3D BAG](https://www.viktor.ai/blog/48/open-source-data-building-visualization), by the TU Delft. This way, they are able to [create realistic environments](https://www.viktor.ai/blog/48/open-source-data-building-visualization) to match their realistic-looking residential towers within the app.  

![parametric design residential tower realistic environment](https://cms.viktor.ai/uploads/CC_JP_van_Eesteren_3_D_Bag_fef59394d3.png)


> “Building our own parametric design tool enabled us to generate the most efficient design for the available space, and save a considerable amount of time while doing so.​" - Martijn de Rog, BIM director at J.P. van Eesteren

For example, we were able to re-calculate the design for a  residential tower that could host the same fascilities, only with two floors less!”​ - Martijn Rog, BIM Director, J.P. van Eesteren​


Find out more about [Cloud-based parametric design](https://www.viktor.ai/parametric-design-software) on the VIKTOR platform. 

[![CTA_try_for_free](https://cms.viktor.ai/uploads/CTA_try_for_free_6af1b46cf2.svg)](https://m.viktor.ai/free_version)

Buildings & Real Estate

Cloud-based parametric design tool optimizes residential towers

medium_CC_J.P._van_Eesteren_residential_tower

small_CC_J.P._van_Eesteren_residential_tower

thumbnail_CC_J.P._van_Eesteren_residential_tower

CC_J.P._van_Eesteren_residential_tower

Discover how real estate firm J.P. van Eesteren build their own parametric design building generator tool on VIKTOR.ai

Parametric design app for optimized building | VIKTOR.ai

Parametric design tool generates the optimal residential tower 

1. What is parametric design, and how is it evolving
2. Collaborative parametric design: working together for better results
3. How to make tools available to others
4. Active collaboration on parametric design tools
5. Parametric design in a billion-euro project
6. What collaborative parametric design means for you



large_Viktor White Paper The future of parametric design is_ COLLABORATION.jpg

medium_Viktor White Paper The future of parametric design is_ COLLABORATION.jpg

small_Viktor White Paper The future of parametric design is_ COLLABORATION.jpg

thumbnail_Viktor White Paper The future of parametric design is_ COLLABORATION.jpg

WP_white

large_parametric_steel_structure.png

medium_parametric_steel_structure.png

small_parametric_steel_structure.png

thumbnail_parametric_steel_structure.png

WP_white_2

The adoption of parametric design grows exponentially thanks to friendly programming languages like Python or visual programming packages like Dynamo or Grasshopper.
 
Our perception is changing. We no longer see parametric design as a model to solve specific technical problems, but rather as a new way of working, in which stakeholders collaborate to find the best solution for a project.

Learn about the latest trends and the future of parametric design in this white paper

Collaborative parametric design models allow you to work together for better solutions

Download the white paper and get INSPIRED

Did you know parametric design models are no longer just used to solve specific technical problems, but rather as a new way of working? Read here how stakeholders use parametric design to collaborate and find the best solutions for their projects!

The Future of Parametric Design is COLLABORATION

A few years ago, [VolkerWessels](https://www.volkerwessels.com/nl/) started looking into ways to arrange parametric design for the automation of their processes. That is when they started using the VIKTOR platform as one of their standard interfaces. With VIKTOR, they could easily develop their own web applications written in Python code without worrying about things like user-management and interface, while at the same time being able to write their own logic, adjust things quickly, and remain close to their end-users. 

In this article, we will dive deeper into the specifics of two of the latest Cloud-based applications that VolkerWessels developed on the [VIKTOR platform](https://www.viktor.ai/platform), used for road surfacing and sheet pile construction.

## The Road Pavement application

The Road Pavement application is developed for the design of asphalt constructions. The application uses input form the CROW, an umbrella organization for multiple market parties, who developed a web tool called OIA (OntwerpInstrumentarium Asfaltverhardingen) with a database that provides all information that is necessary for the calculation of asphalt constructions. 

### Road pavement engineering

Even though this information is very sufficient, running through all calculations that are needed to create a design took road pavement engineers at VolkerWessels KWS@vise a lot of time and manual work. Therefore, they used the VIKTOR platform to develop a Cloud-based Python application that uses [parametric design](https://www.viktor.ai/parametric-design-software) methods to automate and optimize these calculations. 

### Batch calculations 

With the [parametric design](https://www.viktor.ai/blog/46/parametric-design-cloud-engineering-construction) application, they can import data from OIA to create batches of information that are used to automatically calculate designs at the push of a button. It is easy to optimize the designs since height and amount of asphalt can quickly be adjusted. Additionally, results can be found in a clear and visual overview of each road section. This takes a lot of manual labour from the road pavement engineers and leaves more time to analyse and optimise.

See how the application works in this video: 

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/664657264?h=ad5a774a5a&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Road pavement app demo - VolkerWessels (short version)"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

## The Sheet Pile application

Because many underground construction projects keep getting more complex, a lot of data needs to be processed and calculations performed in order to generate the designs. Currently, all of this data is manually put into the system. Calculations are performed manually as well. These processes are both time consuming and prone to error. Therefore, VolkerWessels build the Sheet Pile application for automation and optimization.

### Geotechnical engineering

Like many other VIKTOR applications, the Sheet Pile application is used to easily import and read GEF files of CPTs so geotechnical data is centralized in one place and can be used for adjustments. Furthermore, within the web application, geotechnical engineers can easily enter a sheet pile structure and define the phases of construction, such as excavation, load, and groundwater level, on a user-friendly and intuitive interface.

### PLAXIS integration

For these calculations, the high-end finite element program [PLAXIS](https://www.viktor.ai/blog/34/integration-plaxis-python-parametricdesign) is used through an integration with Python and the VIKTOR interface. The input for the PLAXIS calculations can be sent in the background and the result is returned automatically in Excel tables and PLAXIS graphical pictures. This can be done in batch as well. From here, you can see whether there are any adjustments needed. If so, all calculations can be run again (in batch) at the push of a button.

See how the application works in this video:

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/664659594?h=02cd1bd6ec&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Sheet Pile App Demo - VolkerWessels (short version)"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

## Future aspirations

VolkerWessels is far from done developing their own Python applications and is working hard on proving people the potential they have, for example by adding new features on demand to their existing tools and developing entirely new ones.

### Digital transformation

According to Guido Bongers, Lead Developer parametric design at VolkerWessels, their applications have a big potential. Currently, adoption is the biggest challenge. Not all intended users are fully convinced of the benefits and prefer to stick to their familiar and traditional ways of working. Therefore, the next step is to start with a professional plan to promote their web applications and train potential users so they can truly be adopted.

> “What I keep noticing is that building an awesome application works out every time. The technical side of the story is all fine. However, what we are really investing in right now is the adoption of these technicalities. We want people to start using the product we made. That is the truely difficult part.” – Guido Bongers, Lead Developer parametric design at VolkerWessels.

### New features

Regarding the Road Pavement application, there is a future version 2.0 in sight which will facilitate the collaboration between more disciplines. When it comes to the Sheet Pile application, VolkerWessels has some new features in mind as well. However, incorporating more disciplines is not one of them. 

> “The Sheet Pile application is for one discipline only: The Geotechnical engineer. This is done to ensure user-friendliness. Sometimes, it can be beneficial to build a large application that brings multiple disciplines together and sometimes it is more beneficial to keep it small and focused. That is the choice the Lead Developer needs to make, keeping the end-user in mind at all times.” – Guido Bongers, Lead Developer parametric design at VolkerWessels

VolkerWessels still has a backlog with all kinds of functions and features that users would like to add to their applications. For the Sheet Pile application, one of these features is for example the addition of the underwater concrete floor. Guido: “That is why it is so great to have your own team to develop your own application. This allows you full freedom to add whatever you want to your app!”  

### Upcoming projects

Besides the previous projects, VolkerWessels is currently working on three other Cloud-based applications as well. And they have more projects on the way. 

> “These are all medium sized projects that last about 3 to 6 months to complete. After that, they can be used constantly. That is 3 to 6 months work from which you can reap the benefits for years!” – Guido Bongers, Lead Developer parametric design at VolkerWessels. 










Road pavement sheet piling parametric desing apps

large_CC_VW_two_apps

medium_CC_VW_two_apps

small_CC_VW_two_apps

thumbnail_CC_VW_two_apps

CC_VW_two_apps

Engineers at VolkerWessels build two parametric design apps for the automation and optimization of processes. Learn more on VIKTOR.ai

Optimize structural design processes with Cloud app | VIKTOR.ai

Parametric design apps VolkerWessels contribute to sustainable society

## Larger and more complex

On their previous projects, Heijmans each time developed an application with the purpose of automating and optimizing (design) processes for a specific department. With the Wind Turbine application however, they decided to shake things up a little bit.

“This project was different from our previous applications”, says Tom Nillesen, Lead Developer [parametric design](https://www.viktor.ai/blog/46/parametric-design-cloud-engineering-construction) at Heijmans Construction and Geotechnics, “because this was our first application in which multiple disciplines where ultimately brought together, which made it not only different but much larger and more complex as well!”. 

## Library with building blocks

Whilst building the Wind Turbine application, Heijmans adopted the logic they previously wrote for some of their earlier applications. Tom: “You can see this is very efficient, since you can simply reuse the logic you have already written once instead of starting completely from scratch.” Parts of the Python code that the developers at Heijmans had already written for their previous applications was thus used as digital building blocks to shape the Wind Turbine application. This helped them save a lot of time since these code parts were stored in the Heijmans Python code library where they could easily be retrieved from where and whenever possible.

## A multidisciplinary app

The Wind Turbine App was the first of Heijmans’ application that was built to be used by people from multiple departments. The application was ultimately created for the purpose of automating and optimizing the design process of <a href="https://www.heijmans.nl/nl/verhalen/hoe-staat-het-met-windmolenparken-bouwen/" target="_blank">wind turbine foundations</a>. However, for the design of a wind turbine foundation, data from multiple sources is required.  Therefore, the decision was made to incorporate three of Heijmans departments.

Because a wind turbine foundation design project follows a quite complex process, we will explain it a bit more. What happens on wind turbine foundation design project, is as follows: Heijmans Wind Energy receives the assignment, which is a tender request. For this tender request, a quotation needs to be composed. That is when Heijmans Construction and Geotechnics comes in to build a tender design commissioned by Heijmans Wind Energy. After that has been done, Heijmans Calculations will use this information to calculate the design and propose a cost price, which is then send to Heijmans Wind Energy who issues the quotation to their client. When the work is accepted, Heijmans will then carry out the realization of the wind turbine foundation.  

## Running the app

The Wind Turbine application is created on VIKTOR, a <a href=" https://www.viktor.ai/parametric-design-software" target="_blank"> platform for parametric design in the cloud</a>. It has integrations with several (calculation) software packages (IDEA StatiCa, SCIA Engineer, and D-Foundations) and requires input from various data sources (soil research, turbine supplier, and pile types). Because there are so many processes involved, operating the application may seem complex. Therefore, we will explain exactly what steps need to be followed to create a design.

![wind turbine foundations design automation](https://cms.viktor.ai/uploads/Heijmans_article_img_5c0ad1101c.svg)

*Structure of the Wind Turbine application*
 
## 1. Input Data

Before a design can be calculated, a lot of information on a lot of aspects is needed, such as the bearing capacity of the underlying soil, data from the turbine supplier, and data on different pile types. After you have acquired all this information, it can easily be imported into the application, where it will all be stored in one place. However, before calculations can be run, the following things need to happen first:

-	**Uploading data from soil research.** The Wind Turbine application automatically classifies these CPTs upon uploading them.
-	**Creating a turbine database.** Here, data from the turbine supplier can be stored, such as stiffness requirements of the foundations, forces acting on the foundation, and axis height of the turbine.
-	**Creating a pile database.** Here, data on different pile types (precast-, or cast-in-situ concrete piles) that may be applied underneath the turbine foundation can be stored. 

## 2. Creating a tender design

After determining all information on the soil research, turbine supplier, and pile types, the application is ready to be used to create a tender design, which goes as follows:

-	First, you need to determine values for a few input parameters, such as diameter, thickness, and number of piles underneath the foundation plate. 
-	Within the application, it is then possible to perform geotechnical calculations using D-Foundations and perform a structural analysis using SCIA Engineer. After these analyses, the results on whether the design is sufficient or not are presented to the user. 
-	If the design is sufficient, you have the possibility to create a reinforcement design for the piles. Otherwise, input parameters should be changed, and calculations should be ran again first.
-	To create a reinforcement design, IDEA StatiCA is used to test the section of the piles and see how much rebar is needed for the tender design to be sufficient. 
-	After the calculations, you can download the intermediate results from the software packages to check what kind of calculations were performed. 
-	At last, some very specific information is generated which is used to calculate the cost price of the design. 

<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/641122891?h=6b1e4f46ac&amp;badge=0&amp;autopause=0&amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Wind Turbine Application impression"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

## Access and usage

Because the Wind Turbine application is a multidisciplinary app, all departments are allowed to access and see the information within. However, to keep the process structured, Heijmans decided that only the designers that are concerned with performing the calculations are the ones to operate the application. They also provide the other departments with the right information. This means that while the other parties are able to access the application, in reality they probably never have to. 

## Taking control

Benefits of the Wind Turbine application do not stop at the automation of parts of the design process for Heijmans. Before they created their application, Heijmans did not design wind turbines themselves. Now, with a lot of research and the ability to store this newfound knowledge directly in the application, they are able to deliver an integral design of a wind turbine foundation. This way, the application gives them more control over the entire process. 

> “Before, we did not do the design of wind turbine foundations ourselves. For this, we had to do a lot of research: How do you calculate a foundation? What do we need to bring to the market? Now, when someone asks something about a project, we have much more knowledge on all aspects. This way, we can communicate much faster. There is no third party involved that we have to consult first whenever something is unclear. When consulting starts, the clock starts ticking and prices start to rise immediately. Being able to do this ourselves allows us to generate a design much faster and it saves us so many resources.” – Tom Nillesen, Lead Developer [parametric design](https://www.viktor.ai/blog/46/parametric-design-cloud-engineering-construction) at Heijmans Construction and Geotechnics.

Due to the full control that the application provides, Heijmans expects to lower the costs of a project and achieve a higher hit rate in tenders. 

![engineering construction wind turbine foundation](https://cms.viktor.ai/uploads/Blog_Heijmans_Wind_Turbine_building_fff006cd4f.jpg)

## Upcoming developments

Heijmans has worked on the Wind Turbine application for a couple of months and is currently applying it to tenders. Even though the application already holds a lot of promises, there are still some improvements that they want to work on. 

## From automation to optimization

Currently, there is not yet an algorithm for the optimization of the design. If the design is not sufficient, you now need to manually change a few parameters and run the calculations again. Tom: “This way, you still get to an optimal design pretty fast. We chose to store our geotechnical calculations in a database, so we have a huge set of information on pile point levels and pile to pile distances that we can use to adjust designs quickly, but we want to automate this too!”

## From structural design to structural model

Additionally, Heijmans wants to be able to automatically build not only structural designs but geometric models as well. These models can be used to create blueprints and reinforcement prints, something that normally takes a lot of time and multiple full-time modellers per project to do so. Being able to build structural models at the push of a button would save them not only lots of time, but money as well.

## No more iteration 

Furthermore, Heijmans wants to automate that last part of repetitive work that still needs to be done to minimize chance of errors and make sure no one has to spend time on iterative processes anymore. This way, they have more time left to take on the exceptional cases where they can apply their expert knowledge and create real value.

> “I think that when it comes to design, ‘Automate the boring’ is a slogan that we could start adopting as well. Of course, we cannot (and don’t want to) automate everything, some cases are just too exceptional to automate. Furthermore, as Heijmans, we still have to go outside and build our designs of course. That part cannot be automated with web applications. However, looking at the design process of wind turbine foundations, we are really getting there regarding automation.” – Tom Nillesen, Lead Developer [parametric design](https://www.viktor.ai/blog/46/parametric-design-cloud-engineering-construction) at Heijmans Construction and Geotechnics. 

![wind turbine foundation engineering construction parametric design](https://cms.viktor.ai/uploads/Blog_Heijmans_Wind_Turbine_foundation_project_0229cb1810.jpg)

## Umbrella application

Building the Wind Turbine application, Heijmans especially focussed on the communication and collaboration between the three departments that where involved. However, Heijmans is a large company with many more departments, waiting to be linked as well. Departments such as Soil and Roads are really close knit to Construction and Geotechnics regarding the type of work. Heijmans wants to involve these disciplines in their applications by automating parts of their work processes as well. There are already some aspirations to create a larger ‘umbrella application’ that involves several more departments. For this, they want to use the same soil data and other important information when working on projects that share many characteristics. 

> “When Heijmans Soil and Roads builds a piece of highway that also contains a viaduct, Heijmans Construction and Geotechnics gets involved to work on that part. It is like having companies within a company. They each work on different designs, but within one big ‘umbrella project’. If we are able to use *one* application to create one integral process, this could not only save a lot of time and money, but it would make the overall design better and more efficient as well. That is the ultimate goal I am working towards: Building a network of applications that can be used throughout the entire organization, everything dependent on each other. If you change one thing, everything that depends on that will change as well. But before we can do this, we first have to start somewhere small: Automating a few tiny steps, combining those into a bigger application like the Wind Turbine application. If that appears to be successful, we can start going bigger.” – Tom Nillesen, Lead Developer [parametric design](https://www.viktor.ai/blog/46/parametric-design-cloud-engineering-construction) at Heijmans Construction and Geotechnics.

See how you can [develop your own parametric design tools](https://www.viktor.ai/blog/46/parametric-design-cloud-engineering-construction) as well.


Wind turbine foundation parametric design engineering

medium_Blog_Heijmans_Wind_Turbine_Foundation

small_Blog_Heijmans_Wind_Turbine_Foundation

thumbnail_Blog_Heijmans_Wind_Turbine_Foundation

Blog_Heijmans_Wind_Turbine_Foundation

See how engineers at Heijmans used digital buildings blocks to create a parametric design app for wind turbine foundations on VIKTOR.ai

Parametric design of wind turbine foundations | VIKTOR.ai

Wind turbine foundation tool created quickly from Heijmans' digital building blocks 

## Features of version 1.0

Voorbij created a web-based [parametric design application](https://www.viktor.ai/blog/46/parametric-design-cloud-engineering-construction) on the VIKTOR platform to optimize the (re)design process of pile foundations, which helped their clients save a lot of time and money. 

![figure 1.png](https://cms.viktor.ai/uploads/figure_1_49c8585c84.png)

*Map view of soil investigation*

Before, Voorbij optimized pile foundations based on their experience, by grouping CPTs, and including soil variances in a smarter way. 

![figure 2.png](https://cms.viktor.ai/uploads/figure_2_0de8bde2ac.png)

*Soil interpretation of CPTs*

Now, they use their application on the VIKTOR platform to speed up this time-consuming manual process. Their application allows for the automated grouping of CPTs based on a given set of rules and generates optimized pile plans according to aspects such as cost, soil structure, grouping, pile tip levels, and pile types. The app also provides Voorbij with rich visualizations of results for quick interpretations. 

![figure 3.png](https://cms.viktor.ai/uploads/figure_3_1908f64262.png)

*Visualization of calculation results*

Using their own calculation core, Voorbij is able to indicate upper and lower limits of pile tip levels, easily generate designs, see interactive load-bearing charts, and visualize piles in combinations with the CPTs based off one that information in their application.


## From 1.0 to 3.0 real fast

Now, Voorbij has added some new features to their application. 

### Load carrying capacity

In the first version, Voorbij was only able to calculate on pressure loaded piles. With the new version, they are also able to calculate on pull loaded piles. These are very common in large projects and construction pit projects, which means Voorbij can take on these projects from now on as well. 

> “With the application, we can now consider the whole construction pit integrally.”  - Mark Remijnse, Head Sales at Voorbij Funderingstechniek

### Cost calculation

Initially, optimization was done by reducing material. As of more recent, the application also allows for calculations and linking of actuals costs to provide engineers with a clearer picture of expenses and early on the project already guide the design in the right direction based on budget. 

### Work execution

Also, a somewhat more logistical feature has been added, which is the improvement of work execution. The optimization of a project was previously done separately from the actual construction. Now, the execution and construction parts of the project have been added to the application as well. For the work execution part especially are some preparations required that have been added to the application now:

- **Smart grouping of elements:** Which allows for the optimal design of foundations. Since the soil that a construction project is build on often has a lot of differences in structure, piles vary a lot in size, such as length, diameter, and/or reinforcement. By clever grouping of these piles (putting the right pile in the right place), you can logistically save a lot of time and reduce chance of errors.

- **Bulk orders:** If you want to order material, it is most beneficial to order in large numbers with low variety, based on the [smart](https://www.viktor.ai/blog/40/artificial-intelligence-machine-learning-engineering-construction) grouping you did before to avoid and reduce unnecessary costs.


![figure 4.png](https://cms.viktor.ai/uploads/figure_4_ed64713811.png)

*Visualization of combined pile rebar cages*

Read more about version 1.0 [here](https://www.viktor.ai/customer-cases/7/digital-transformation-python-pile-foundations)

### Increased calculation capacity

In the early version of the application, Voorbij was already able to use their application to do batch calculations of CPTs. Back in version 1, they could run an analysis of about 150 CPTs at the same time. With the latest version however, they are able to calculate up to 1000 CPTS simultaneously, increasing the calculation capacity.

## An environmentally friendly application

The addition of new features was not only beneficial for Voorbij and their clients, it also brought some environmental advantages along. 

### Less crane days

With the release of the latest version of their application, Voorbij was able to reduce days of crane usage by 40 in one case. This meant 40 days of proximality 250 litres of fuel less, which is good for 10.000 litres reduction of (CO2) emissions.

### Less concrete

Additionally, the new features allowed for the creation of a design that required about 6.000 to 6.500 m3 less concrete, resulting in Voorbij using 26.500 m3 concrete for their design were another company calculated about 33.000 m3 concrete for their design. Every m3 of concrete consists of 350kg cement, meaning every 3 m3 of concrete is about 1.000 kg cement, which makes up for about 2.275 tons kg less cement. This also equals 2.275 tons less CO2 emissions. For this design, that would be over about 2.000 tons CO2 emissions less! On top of that, it also reduces transport of approximately 500 truck movements of ready-made mixed concrete.

## From developer to advisor

Before working with automation applications, when taking on a project Voorbij would receive a definitive design that they had to make into an execution design, meaning they could only adjust some details. Since they started using their applications, they are pulled into the process from an early stage on, mostly somewhere between preliminary design and definitive design. In this stage, there is still a lot of room for advisement, improvement, and optimization.
 
According to Voorbij, their engineers are the ones who are now talking to the constructors, which works very well, as they understand each other best. 

> “Before, we were mostly doing calculations. Now we are able to put a lot of effort on advising. You can use your cleverness and really add value to a project this way.” – Wiebren Koers, Geotechnical Advisor at Voorbij Funderingstechniek

> “Thanks to our applications, our engineers are now only doing 4 up to 8 hours per design calculation instead of 20 to 40 hours per design. This, I think, is still too much however, since they can add the most value by sharing their knowledge!” – Mark Remijnse, Head Sales at Voorbij Funderingstechniek

## The BASE FT tool

Voorbij has been working hard on collecting all their project data and creating a database for the purposes of comparing their data and applying Machine Learning techniques to improve and predict future on future projects. 

The BASE FT tool consists of three pointers, which are the applications, their machines, and a database. Currently, there are three applications, but Voorbij is aiming to expand BASE FT’s network and incorporate more external applications from other companies as well.

> “It is important that BASE FT can be used to link other applications and disciplines to your processes” – Mark Remijnse, Head Sales at Voorbij Funderingstechniek

Furthermore, the machines that are part of BASE FT constantly track data when used on a project, this data is gathered and saved in the BASE FT database, where they use data engineering to optimize and better generate future designs.

## More to come

Besides expanding BASE FT’s network, Voorbij wants to work on improving their current applications as well. For example, by adding a nitrogen module to the pile application. They also want to start using the data from BASE FT’s database to predict the course of a building project, analyse execution data, and improve their designs. 

Read more about how they do this [here](https://www.viktor.ai/blog/40/python-artificial-intelligence-machine-learning-automation-application-development).

And see how you can [build your own parametric design applications here](https://www.viktor.ai/parametric-design-software).


Engineering & Design

medium_Voorbij palen.jpg

small_Voorbij palen.jpg

thumbnail_Voorbij palen.jpg

Customer_case_parametric_design_tool_fast_generation_optimal_design_1

Discover how engineers at Voorbij FT build a web app to automate and optimize design of pile foundations on VIKTOR.ai

Web app for optimal pile foundation design | VIKTOR.ai

Optimal pile foundation design saves tons of CO2

The [tunnel](https://www.cowi.com/solutions/infrastructure/northern-europe-largest-construction-project) is 18 kilometres long and will consist of a four-lane motorway and two electrified railway lines. This new tunnel will reduce travel time for from more than an hour by ferry to 10 minutes by car and 7 minutes by train. Besides reducing travel time, the tunnel will also lead to a [reduction in traffic emissions](https://femern.com/en/Benefits/A-greener-corridor). This is due to the transfer of freight from road to rail and the fact that some lorries will take a shorter route as they will no longer have to cross Storebælt, Denmark, which will result in an overall decrease in CO2 emissions. 

In cooperation with COWI a [parametric design application](https://www.viktor.ai/blog/46/parametric-design-cloud-engineering-construction) will be developed to optimise the design of the tunnel. The tunnel consists of 79 tunnel elements, each 217 metres long. Each element has slightly different loads and soil conditions, so the optimal required steel reinforcement in the concrete is different. The total amount of steel used in the tunnel is equivalent to about 50 Eiffel Towers and each tunnel element weighs 73,000 tonnes. 

The VIKTOR platform will be used to development the [parametric design application](https://www.viktor.ai/parametric-design-software) which will be used to design the tunnel, which will help COWI find the optimal design in an automated way, so they can engineer the awesome! 

 

.   

 

 

large_COWI Fehmarnbelt.jpg

medium_COWI Fehmarnbelt.jpg

small_COWI Fehmarnbelt.jpg

thumbnail_COWI Fehmarnbelt.jpg

Customer_case_automating_longest_immersed_tunnel_world

Discover how parametric design is used in an application to automate and optimize the design of the world's longest immersed tunnel on VIKTOR.ai

Automated design of world's longest tunnel | VIKTOR.ai

Automating the design of the longest immersed tunnel in the world 

Original source: [Land+Water](https://mocs698.sharepoint.com/sites/Marketing/Gedeelde%20documenten/Marketing/Land%20+%20Water_%20Artikel%20module%20damwandconstructies%20%20ism%20Viktor.pdf), nr 3 – March 2020. K. Nijs / R. Reurings.

</br>

Companies in construction and infra are experiencing a turbulent transformation to higher levels of digitalization. The reason for this being that this sector is lagging behind quite significantly compared to other industries. Just like in other sectors, digital transformation is happening on all kinds of aspects and throughout all phases of a project. This means that digital transformation is a relevant topic from tender and engineering to execution and maintenance.
 
Voorbij Funderingstechniek noticed the potential of this transformation in their industry and decided to act upon it. How? By using VIKTOR to automate and optimize part of their processes to be able to generate the best designs fast!

## Process analysis determines potential

For the first step towards building their own app, Voorbij Funderingstechniek started with a process analysis. The goal of this analysis is to gain insight into the impact of automation on business processes and what this means for the current processes. 

This sparked some opportunities. Currently, a sheet pile calculation is still an intensive and iterative process. To come up with a first design, a significant number of variables needs to be considered, such as the layout of the soil, the desired excavation depth, environmental influences, loads, and desired construction phase. This results in one or more normative sections with corresponding starting points. 

After the first design, every small change requires a new model and a new calculation. That means that all files need to be copied and everywhere one parameter, such as the type of sheet pile, the length, or the drainage, needs to be changed manually. In a typical design process, the second design also does not meet all requirements yet. The process of copying and adjusting keeps repeating itself, for every defined section, multiple times. This is exactly where digitization makes a difference; by automating repetitive tasks a lot of time can be won. 

Another complication is that for projects with multiple sheet piles often a ‘normative’ design is used for all sheet piles. Here, a sheet pile is designed as well as possible (within the stated time) and duplicated to all other sheet piles.

![Blog_smart_application_contributes_sustainable_future_construction](https://cms.viktor.ai/uploads/Voorbij_areal_a17012b79f.jpg)


## Cost reduction

By automating the design process, a lot more design iterations can be performed, resulting in cost reduction with regards to materials used and time savings for engineering. Risk of errors becomes close to zero as well because processes are automatically copied, pasted, and adjusted. Additionally, and maybe most importantly, automating processes gives engineers more time to apply their specialist knowledge on more challenging projects to create a real added value. 

With their process analysis, Voorbij identified a couple of opportunities that function as input: 

- Repeatedly copying models
- Inserting variables
- Getting results per section

With a custom application on the VIKTOR platform, these processes can be automated and linked to different steps, such as analyzing CPTs, calculating designs in calculation software, and iterating until the results are sufficient. 

## Web-based

VIKTOR is a web-based platform for the development of online applications with Python. In Voorbij’s application, Python is used to program and guarantee professional and business knowledge, specifically by integrating all relevant data sources and software programs. 

With Voorbij’s VIKTOR app, data and analysis results are centralized and made available through a dashboard to all relevant (authorized) people. This ensures that everyone always has access to all insights and the correct up-to-date information.
 
To automate and later optimize sheet piles designs, the process first needs to be standardized in the application. To achieve standardization, the right ‘recipe’ needs to be discovered that always delivers a good result. Questions asked here are: “How do we process a CPT?” “How do we build the model and how do we process the results?”. Only after everyone agrees on this, automation can follow. 

![CC_Voorbij_sheet_piles](https://cms.viktor.ai/uploads/CC_Voorbij_sheet_piles_180f2cab82.png)

*Result output of the BASEFT Sheet Pile application*

</br>

## Automation

The online application is designed as a central entry point of all variables. It creates the models itself, performs the calculations, and generates a summary of the results. 

If we follow this route, there is no need to perform any repetitive tasks. Do all sections roughly have the same excavation depth? Does the engineer want to iterate sheet pile lengths? In that case, the input only needs to be adjusted in one spot in the application to recalculate all the sections! Fast iterations like this, combined with the results gathered from multiple sections, generate lots of information that can be used to make connections between, for example, drainage and required sheet pile length. 

Automation is of course not restricted to inserting variables and generating models. Other processes, such as drainage, determining head heights (instead of manually inserting these in D-SheetPiling), and more, are automated as well. 

![CC_Voorbij_sheet_piles2](https://cms.viktor.ai/uploads/CC_Voorbij_sheet_piles2_b90017729c.png)

*Parameter input of the BASEFT Sheet Pile application*

</br>

## Optimization 

Now that it is possible to automatically perform multiple sheet pile calculations, optimization can start. The most important question is: What is the goal of the optimization? Is this the cheapest, lowest CO2 emission? Or the easiest executable sheet pile? By specifically defining this goal, a couple of conditions will naturally follow. Based on these conditions, the application calculates the optimal design. This can be in either tens or even thousands of iterations. 

Let’s say that the cheapest sheet pile is your goal. In that case, the application works as follows: If the previous design is more than adequate and has a certain price, the next design can be repeated but for a shorter and lighter sheet pile with a lower price. If this design is no longer adequate, you can take a step back and make the sheet pile a bit heavier, but for a lower price than with the first design. This can be repeated until the optimum is found between price and requirements. 

In essence, digitization of the construction and infra sector should create a distinctiveness because the automation of repetitive work leaves engineers more time for the real and fun thinking. 

Automate the boring. Engineer the awesome! 

## Interested?

Are you interested in this development as well? Visit [www.voorbijfunderigstechniek.nl]( https://www.voorbijfunderingstechniek.nl/technieken/grondkerende-constructies/stalen-damwand) or contact Richard Reurings at r.reurings@voorbijft.nl for more information.


large_CC_sheetpile_voorbijFT

medium_CC_sheetpile_voorbijFT

small_CC_sheetpile_voorbijFT

thumbnail_CC_sheetpile_voorbijFT

CC_sheetpile_voorbijFT

VoorbijFT uses VIKTOR to automate and optimize part of their processes to be able to generate the best designs fast!

Customer cases

Optimal sheet pile design within reach