Digital Builder Ep 14: 3 Things We Learned About AI & Machine Learning in Construction

ai construction

There are plenty of misconceptions around artificial intelligence (AI) and machine learning (ML). For some people, these technologies conjure up thoughts about self-aware robots who’ll take over the world. Then there’s the more common notion that AI and ML are here to replace humans and take their jobs. 

Both are far from the truth. The reality is that artificial intelligence and machine learning can help keep job sites safer, improve the productivity of construction teams, and help firms understand their businesses better. 

On Episode 14 of Digital Builder, Josh Kanner, Founder and CEO of and Pat Keaney, Director of Product Management, Intelligence at Autodesk Construction R&D, join us to demystify artificial intelligence and machine learning in construction.

We discussed:

  • What machine learning and artificial intelligence really are 
  • The common misconceptions about this type of technology 
  • How artificial intelligence and machine learning are used today in construction, including real-world examples 
  • What we can expect in the future (surprise — the future is already here!)

“Over the next 5-10 years, AI in construction is going to be all about augmenting people — making people more efficient, making them smarter, making their lives better.” — Pat Keaney

New Episodes Every Two Weeks 

Digital Builder is hosted by me, Eric Thomas. Remember, new episodes of Digital Builder go live every two weeks. You can hear more episodes like this one by subscribing to Digital Builder on Apple Podcasts, Spotify, Stitcher, or wherever you listen to podcasts. You can also subscribe to our email list with the form below so you’re the first to hear when new episodes are released.

Here are the top three things we learned about AI and machine learning in construction. 

1. There Are Some Common Myths About AI and Machine Learning

The realms of artificial intelligence and machine learning are rife with misconceptions, myths, and fears — most of which are largely unfounded. 

“One myth is that AI and ML are buzzwords or that they are science fiction. The reality is that we use AI and machine learning every single day,” says Pat. 

Smartphones today have image search functionalities that rely on artificial intelligence. Streaming services like Netflix use AI and ML to power its viewer recommendations. 

These technologies are also used every day in construction. According to Pat, construction professionals are increasingly relying on artificial intelligence to streamline various processes and improve safety. 

“People are using AI to find the right subcontractors during procurement. We're using AI to automatically generate submittal logs by looking at plans and specs.”

Pat continues, “We're also leveraging AI to automatically identify root causes of RFIs and to continuously identify high risk issues by tracking progress, predicting safety incidents, etc. This stuff is happening literally every single day in construction.”

She then dispels the fear of robots taking jobs away from humans and says that AI and ML are here to help workers do their jobs, not replace them. 

In fact, AI can give construction professionals “superpowers,” says Pat.

“Over the next five to 10 years, AI and construction is going to be all about augmenting people, making people more efficient, making them smarter, and making their lives better. We’re definitely not going to be replacing people in construction in that timeframe.”

Another misconception is that AI is always right. But it’s important to remember that there are also error rates in artificial intelligence. The key, says Josh, is to not fixate on those error rates and instead focus on how AI can help surface the insights you need. 

“The point I would add here is that it's not about error rate. It's about, ‘Does AI help us run our business better?’,” adds Josh.

“Can we surface those areas of insights, the workflow automation, the pieces where it can actually help us move forward with confidence? There are pieces of AI that we will have more confidence in than others, and those can help move the business forward.”

2. AI and Machine Learning are Driving Advanced Business Insights

AI and machine learning have several practical and powerful applications in construction

One area where these technologies can truly shine is safety. Tasks related to safety observations and inspections — which are traditionally done by people — can be time-consuming and difficult to scale. 

Enter AI and machine learning. 

“What if you could train a set of computer vision models to identify these situations, just like a person could do, but now you can do it on a massive scale?” asks Josh. 

“You don't need to have people walking around, you can actually have the machine help you understand what those scenarios are. So now you're able to get hundreds of additional observations a week, and thousands a year.”

He adds that with more and more construction data available in digital form, companies are in a better position to use all of that information to create safer jobsites.

"There's a massive amount of digital data that exists in systems today... and beyond that, there's daily reports with manpower logs, you've got field observations, you've got employee timecard data, locations of jobs, and more."

When layered with structured data from AI, construction firms can start predicting incidents and reducing risks — ultimately helping keep workers safe. 

“If you start being able to combine all that stuff together, which wasn't possible before, you can now start creating business insights and predictive insights,” explains Josh.

Beyond safety, AI and ML can streamline various construction management processes

As Pat puts it, “AI can help you accomplish things that you really want to do but you just don't get around to.”

One example is identifying the root causes of RFIs.

“People are so busy, they want to know the root cause, but they really just don't have time to think about it… So, we said 'Well, what if we could just create that data for them? We can analyze the RFIs, classify them, categorize them, then find the root causes,’” explains Pat.

Pat says that by using AI and machine learning, they were able to validate the root causes of RFIs and surface those insights to the right team members. 

According to her, the RFIs example illustrates how artificial intelligence and machine learning make team members’ jobs a lot easier. 

“There's a lot of places where AI is actually doing things that people have wanted to do, they know it's the right thing to do, but it's just too hard and time consuming.”

Pat adds, “those are some perfect applications where AI can step in and say, ‘Hey, you know what? We can do that for you. Like in the story of the Elves and the Shoemaker; we can do it while you're sleeping and it's just going to be there for you.’”

3. Getting Started with AI and ML Doesn’t Have to Be Scary

Getting started with AI and ML can feel overwhelming, but there are simple strategies and tools that can facilitate the process. It’s also important to note that digitizing data collection and analysis should be done ASAP, as the data you gather today will benefit you tremendously in the future. 

“The first thing I would tell any construction company is, digitize, digitize, digitize. You will not regret it, and every piece of data you collect today is going to be really valuable to you not only today, but five years from now and 10 years from now,” says Pat.

She adds that there’s nothing to fear with AI and ML.

“Don't be afraid of using AI. It’s there to make your lives better. By using AI and engaging with products that analyze your data, you’ll learn more about your data than you can any other way. It'll help you see problems and what you can start to correct. It then becomes a virtuous circle, because you will be making your data richer.”

The good news? You can begin that virtuous cycle right away. 

As Josh points out, “It's really not hard to get started. If you're an Autodesk customer, for example, you can use our pre-built integrations to BIM 360 and start analyzing data that you're gathering already. It takes less than 90 seconds to get the integration set up and the data to start flowing.”

The next step is to understand the scope of your data and see what you can do with it. 

"We call it a health check, so you can see if you might be a good fit for predictive safety modeling at your company,” Josh explains. 

All in all, Pat and Josh both agree that AI and ML can unlock major advantages and these technologies should be adopted sooner rather than later. 

“Just get in the game. Digitize, and do not be afraid of AI,” reiterates Pat. 

“If AI can help you automatically create a submittal log that you used to spend two weeks doing, why wouldn't you use it? If AI can prevent one death in your company, why wouldn't you use it?”

Listen to the Full Episode of Digital Builder

Hear everything that Pat and Josh have to say about artificial intelligence and machine learning in Episode 14 of Digital Builder. Tune in to get a better understanding of these technologies and how you can start using them to your advantage.

Eric Thomas

Eric is a Sr. Multimedia Content Marketing Manager at Autodesk and hosts the Digital Builder podcast. He has worked in the construction industry for over a decade at top ENR General Contractors and AEC technology companies. Eric has worked for Autodesk for nearly 5 years and joined the company via the PlanGrid acquisition. He has held numerous marketing roles at Autodesk including managing global industry research projects and other content marketing programs. Today Eric focuses on multimedia programs with an emphasis on video.