Machine learning has been making waves in the construction industry. But what does that really mean and how are teams using this seemingly futuristic technology today?
Let’s start by unpacking the concept of machine learning, a subset of artificial intelligence. Machine learning uses algorithms that allow computers to learn from data without being explicitly programmed to solve a specific problem. Without knowing it, you probably already leverage the benefits of machine learning in your daily life. For example, consider the filters on your email that tag incoming messages as ‘spam.’ This is a perfect example where machine learning identifies certain keywords or senders and automatically knows if an email is likely to be spam. And as more people flag certain emails as spam, the algorithm becomes smarter. Over time more spam is being accurately tagged, benefiting inboxes everywhere.
So how does that work in construction?
There are a lot of great use cases for how machine learning can improve our industry. For example, companies like Newmetrix use image recognition to scan jobsite photos and flag safety concerns. Pype uses SmartPlans to scan through drawings to identify submittals or potential submittals and organizes them. Even TradeTapp ingests data to compare subcontractor risk levels and proactively mitigate against project risk.
All of these products are leveraging machine learning to help simplify and automate construction workflows. And one of the industry’s most powerful machine learning based capabilities is through Autodesk’s Construction IQ.
At Autodesk, our focus is applying machine learning to deliver better products and user experiences for our customers to make their work lives better. Construction IQ for the Autodesk Construction Cloud platform and BIM 360 does just that. With built-in machine learning technology, it helps teams to predict, prevent, and manage risk by giving them instant visibility into daily priorities, a snapshot of individual project health, and insight into company-wide performance across projects.
We spoke with Michael Murphy, Digital Construction Operations Manager at BAM Ireland, to understand how Construction IQ is helping to drive safer, higher-quality projects for its teams with less risk. Here are some great examples of how Construction IQ can specifically help teams across various workflows:
One of the most important goals on a jobsite is to ensure everyone goes home safely at the end of each day. Safety management is by no means a new concept. But with Construction IQ, teams can spend more time looking into leading indicators that can predict and reduce safety issues rather than responding to lagging incidents.
For example, 60% of all construction related deaths are attributed to accidents occurring within one of the “fatal four” categories; fall, caught in between, electric, and struck by. Construction IQ uses its machine learning technology to categorize jobsite issues that fall into these categories so that teams can quickly see, prioritize, and address the most pressing issues.
“Construction IQ gives us laser sharp focus in terms of the health and safety issues onsite, enabling our teams to focus and track leading safety indicators, whilst working with our supply chain partners more effectively,” says Murphy. “By visualising safety issues such as high-risk subcontractors or high-risk safety items, we can address issues right away, resulting in a heightened level of safety on-site.”
Keeping track and maintaining project quality is another critical part of a successful project. But it’s challenging for teams to understanding where to focus attention to address these issues.
Through the Quality Risk Factor card in both the Autodesk Construction Cloud platform and BIM 360, you can easily see a list of high risk subcontractors and issues predicted by Construction IQ. These could have been tagged if there was some indication of rework, inspection risk, or water related risk as water infiltration can be a primary cause of high-risk quality issues as construction progresses.
“Construction IQ surfaces all of the leading quality-related trends that support our teams in focusing on helping our partners succeed in their execution,” says Murphy. “By flagging an issue that poses high risk earlier in the process, we can eliminate any issues before it significantly impacts the quality of the project.”
Through internal research, we found out that over 70% of RFIs stem from design or documentation issues. Diving a bit deeper, unsuccessful projects (classified by profit margins) tend to have 50% more RFIs, with a root cause in Design & Engineering or documentation issues, than successful projects. It is critical to resolve these design issues as early as possible, so they don’t become higher cost impact RFIs later during construction.
With the Design Risk Factor card, you can see design specific risk such as code compliance issues, issues to critical building components, and errors and omissions in design and documentation.
“The Design Risk Factor card provides insight into design issue trending, which enables our design teams to concentrate on areas of the design that are on our critical path,” says Murphy. “By identifying design issues earlier in the process, we can avoid any necessary cost overruns or schedule delays.”
Teams often struggle in prioritizing which RFIs to focus on. This can frequently lead to additional rework, schedule delays, and cost overruns. Furthermore, design and construction information are often disconnected, increasing the total number of RFIs in general.
Our data experts found that projects that prioritized closing more critical RFIs faster were more successful. So how can teams better prioritize and manage RFIs?
The Construction IQ driven RFI Risk Factor card breaks down RFIs based on those tagged as higher risk, any MEP or Structural RFIs, RFIs that have a root cause in design coordination, documentation errors, or code compliance since these tend to drive the most cost or schedule related impact.
“Construction IQ provides critical insight into the status and progress of RFI's on which to prioritise by level of risk,” says Murphy. “This reduces the level of effort normally required by us to manage this process on a daily basis, allowing us to focus on other critical areas of the project.”
Harness the power of Construction IQ’s built-in machine learning and AI functionality to predict, prevent and manage risks to cost, schedule, quality, and safety. Learn more here.
We recently spoke to two industry experts on the future of AI and machine learning in construction on our Digital Builder podcast. Listen to the full episode here.
