It is without a doubt that AI has been the trending topic for the last few years. This can be seen with ChatGPT’s “doll pictures” that were flying around at one point, or with AI-generated videos from big companies. We can safely say that AI has had a big impact on the marketing industry. But did AI affect the way civil engineering consultancies work?

Has AI Changed the Way Civil Engineering Consultancies Work?

In this article, we want to cover our experience with AI and whether it has changed the way we work.

To start with, we need to establish what AI systems are out there. Here is a short list:

• Generative AI: This is the AI that you have mostly seen “flying around” – generative text, images, video, and audio.
• AI Agents & Bots: This is the AI behind the scenes that carries out tasks with a set goal. E.g., if a new enquiry comes in, it creates a new prospect folder in the system and pings an appointed person.
• Predictive AI: This is AI that has been trained on a specific task. E.g., it is used in finance and logistics to forecast. This does not create content but calculates.

Which Types of AI Are Used in Civil Engineering Consultancies?

Now that we have established the types of AI, the question is: which AI has made its way into civil engineering consultancies? Based on our experience and speaking with peers, the short answer is Generative AI and AI Agents.

AI Agents in Civil Engineering:

Some companies have incorporated AI agents primarily to automate admin tasks, such as setting up a job folder when a new enquiry arrives in an inbox. This is great as it reduces admin time. Currently, AI agents have not made it into the design process.

Generative AI in Civil Engineering:

Generative AI, especially in the form of generative text, has definitely made its way into the design process in one shape or another. Some notable examples are:

• Uploading multiple documents and providing a summary.
• Providing meeting minutes and identifying actions by sifting through a meeting transcript.
• Summarising the results of an online search.

All the above are great, but they are still somewhat administrative tasks. This is still very much welcomed as it reduces admin time and gives us more time to focus on the engineering.

Now, generative AI is not just able to produce text but also can produce pieces of code. These lines of code can help engineers automate tasks. This is very beneficial as it has allowed us to significantly reduce time spent on repetitive tasks.

Practical Engineering Use Cases Enabled by AI

Here is how we have utilised AI to increase our effectiveness and accuracy:

Drainage Design – Pipe Clash Checking:

Whilst there is no AI-generative drainage design software out there (as far as we know at the time of writing this post), it does not mean that we should not explore other options to leverage AI. With a fundamental understanding of how to carry out a pipe clash check, we prompted AI to generate code that allows the user to speed up the process. By asking the user for pipe details and Invert Levels (IL), the code carries out the clash check. This has allowed us to increase the speed and accuracy of our checks.

External Works – Setting up Finished Floor Levels (FFL):

When our clients ask us to produce an earthworks cut & fill exercise, the process behind the scenes can be laborious. This is especially true on residential projects where we have to draw lines around each house and assign the proposed FFL to take it into account for the exercise. By prompting AI to generate a piece of code that reads the proposed FFL and assigns that elevation to the line (so it can be used in the 3D model), we have increased our efficiency and removed the repetitive task of manually typing each FFL on each line.

Feasibility Assessment – Preliminary Earthworks Exercise:

We are usually asked to produce a Preliminary cut and fill exercise to support the bidding process on a new scheme. These types of projects can become very time-consuming very quickly, especially when assessing different options. The solution for this one is not AI exactly, but rather keeping our finger on the pulse of the AEC industry by reviewing available products.

In this case, we found a Grading Optimisation plugin that allows our engineers to define key features of a site and set parameters such as max/min slopes, building outlines, and road alignments. You then specify the desired earthworks requirement (e.g., Balance?). The software then goes through an iterative process to produce a model.

Now, this is great on paper, but it runs the risk of things going south very quickly. If the end product is not reviewed thoroughly by our engineers, the software could produce a model that adheres to the parameters but results in the ground sloping towards the building—which is not ideal! A careful review of the end product is required. Our experience is that this type of application is a good starting point, and then our engineers can take it from there and finalise the model.

Our Approach to AI:

So as you can see, we have used AI in engineering, but not in the way people might expect based on other industries. Instead of letting a bot take the wheel, we have leveraged AI to create bespoke code and automate the heavy lifting for us. At this stage, we haven’t found an “all-in-one” AI software that can replace a full drainage design or a complex levels strategy and in our view, that is still a long way off. Engineering requires navigating constraints that aren’t always obvious to an algorithm, like phasing, specific flood risk zones, and site-specific nuances.

Looking Ahead: Efficiency Without Replacing Expertise:

However, the efficiency we’ve gained by using these tools is real. We are committed to staying at the forefront of this technology to ensure our clients get the most accurate and cost-effective designs possible.

If you are looking for a consultancy that combines traditional engineering expertise with modern AI-driven efficiency, we would love to hear from you. Get in touch with our team today to discuss how we can bring this streamlined approach to your next project.

Written by Ferdi Jafar, Associate.