Predicting the future with machine learning in construction

In our latest Trailblazers discussion, we speak with Tessa Lau, CEO and co-founder of Dusty Robotics.

Lau talks about her career background, her insights into the state of innovation in construction, and how she applies her knowledge about robotics to make a difference in the industry.

She also reveals why she’s excited about what data is going to do construction and how her company wants to ensure that data is no longer getting lost. “Once you bring robotics onto job sites, robots collect and publish that data in the cloud. You can start doing analytics over that data once it’s available,” she says.

Dr. Burcin Kaplanoglu, vice president, Oracle Industries Innovation Lab, led the discussion.

What is your current role and how has your career evolved since you started in the industry?

My name is Tessa Lau and I am the CEO and co-founder at Dusty Robotics. I've been in robotics for about 10 years now. I have a background in computer science, technology, and AI.

I got into robotics with a company called Savioke that I co-founded with several other people. We were building robots for hotels. So, if you ever stayed in a hotel and had an R2-D2 style robot bring you room service, that was probably one of our robots.

After five years of that, I founded a new company which got me into the construction industry. We did a lot of research about the industry starting with a house remodel. This is where I learned about construction, including all the different pain points people have.

Construction is still manual labor, and that is what is limiting the industry's productivity. I figured it would be a great place to apply my knowledge about robotics and create a difference in the world. That's how we got to where we are today.

What are the biggest challenges to innovation when it comes to construction engineering?

The industry has a history of being slow to adopt innovation and technology. In my experience with Dusty Robotics, we’re seeing people get really excited about bringing new technology into the construction environment.

Some of the challenges we've been able to resolve are things like old-school superintendents being very skeptical about anything that touches their job site. But the reality is that, in our experience, as soon as these superintendents see our product, they get big smiles on their faces and say, "Yes, this is absolutely what I want."

The challenge with innovation is coming up with the idea and product that will make a difference in people's lives, including understanding what the needs are and how to address them. Once you've got those nailed down, innovation can take off rapidly within this industry.

Where do you see successful innovative technologies in terms of adoption?

One of the philosophies we have in our company is not trying to boil the ocean. We're not trying to solve everything in one product. We’re creating point solutions and drop-in replacements for existing workflows that do things 10 times better than people are doing it today.

This includes understanding the structure of the construction industry and figuring out how to meet people where they are today without selling them on this grand vision that requires changing everything they’re doing.

Instead we’re saying: "Look, we're going to get you there, but one step at a time. And each step that we take, you'll understand why we're doing it. There’s a clear ROI just for taking this one step with us. We’re ultimately going to get you where you want to be in terms of revolutionizing and transforming the industry."

This approach has allowed us to get our foot in the door on a lot of projects that we might not have otherwise done if we were trying to take a much bigger, more grandiose vision to industry transformation all at once.

When you look at the industry, what emerging technology do you see presenting the biggest opportunity for the industry?

My background is AI, but to be honest, the biggest thing I'm excited about is what data is going to do for the industry. Construction is a manufacturing process. It's about building buildings, and each building is a unique snowflake that is built differently every time. There is so much data that gets generated throughout this entire process. Right now, most of that data gets lost, and no one is collecting, monitoring, or analyzing it.

What I would love to do, and what our company is advocating for, is ensuring that data is no longer getting lost. Once you bring robotics onto job sites, robots collect and publish that data in the cloud. You can start doing analytics over that data once it's availabale.

That’s where AI and machine learning will come into play, but you can't do that until you have that data in the first place. Collecting information about what's happening in the field—what are people doing; how long it took them to do things; what pieces are they touching, what tools are they using; what parts are they installing; why are they doing it; and who's doing it—all that data is information.

That data is going to empower the next set of innovations that will leverage a lot of the AI, machine learning, and other innovations we're seeing in other industries.

"Construction is a manufacturing process. It's about building buildings, and each building is a unique snowflake that is built differently every time. There is so much data that gets generated throughout this entire process. Right now, most of that data gets lost, and no one is collecting, monitoring, or analyzing it."

-Tessa Lau, CEO and Co-Founder, Dusty Robotics

Where do you see the biggest impact to construction engineering in terms of AI and machine learning?

The biggest impact is going back to data. Machine learning is about inferring patterns from data. I would love to see more information available about what's happening on job sites. Once that information is available, we can start recognizing patterns and using those patterns to predict the future. That’s really where AI is coming into its strength across the broader industry.

For example, we're installing parts in the field, and this manufacturer's parts take twice as long to install as another manufacturer's parts. We can start analyzing and using this information to optimize the entire design process.

This will help us figure out, “How do I optimize the entire construction process based on information that we're starting to gather in the field and learn from past mistakes to prevent making them again in the future?” That’s where the real opportunity lies, starting to instrument the workflows that are happening in construction. This will enable the next generation of AI and machine learning.

Oracle Construction and Engineering, the global leader in construction management software and project portfolio management solutions, helps you connect your teams, processes, and data across the project and asset lifecycle. Drive efficiency and control in project delivery with proven solutions for project controls, construction scheduling, portfolio management, BIM/CDE, construction payment management, and more.

Read more Trailblazers articles here.