If AI would design a bridge, would you dare to cross it?

I don’t know about you, but the math teachers I’ve met would scold us to study hard because we would need to build a bridge one day, and if we messed it up, people would die. That’s a perfect point for all the professionals who rely on math and have huge responsibility, not only construction engineers.

But now, AI is everywhere, including in industry where public safety is a critical issue. According to Ankin, the impact of AI in construction reached $22B in 2022 and is expected to grow at 20% of the compounded annual growth rate through 2023 based on improved efficiency and productivity through optimization of project scheduling, resource allocation, and task management, leading to shorter timelines and reduced costs.

So, AI is swiftly pulling its weight in the construction industry but it’s definitely not going to build a bridge on its own. However, it can very well perform all the calculations on its own and there are already some AI bots which can help.

Can I build a bridge using AI?

You can start off by getting some figures about the budget you will need for the project. For instance, Civis.ai has a bridge design calculator where you can input the length of the bridge, its type, and the region in which you want to build it.

You will get an estimation on the quantity of materials you will need to use, how long will it take and, most importantly, how much will approximately cost. Then, if you scroll down, you will see a list of examples of similar projects and you will be able to research about their development as well.

However, it’s less known that a pedestrian bridge was already designed based on AI, as reported by Medium. In fact, researchers from ETH Zürich developed a method to use Artificial Intelligence (AI) to change the traditional design process.

They rely on first studying the performance parameters and then go to the design process instead of the classic mode that is reversed. This approach, apparently, cuts from the costs and time needed for the whole process.

The swiss, then, developed a whole construction engineering toolbox based on the Conditional Variational AutoEncoder (CVAE) learning model.

How can AI improve the construction industry?

It’s easy to understand how AI can help with the design and costs part, but there are a lot of other aspects it can improve on.

For instance, the AI can analyze building codes, zoning regulations and environmental obstacles that could hinder the project. Furthermore, it can assess the costs and distribution of labor in the area where your project will be developed.

Lastly, it can underline the possible safety risks on the construction site and even simulate any problems within the construction before you even start the project.

We don’t think we need to hold our breath until we see the first 10 mile bridge designed entirely by AI, but it already helping engineers all over the world building them.

So, would I cross that AI-generated bridge? Yes, I would, but only if expert humans still supervised and approved every detail. How about you? Write down your thoughts in the comments section below.