Smart Tech, Smarter Construction
Data-driven insights from AI (artificial intelligence) and BI (business intelligence) is enabling construction workers to design and execute their skills more precisely, quickly, and accurately on-site. The technologies have replaced old-fashioned measuring tape, best estimates, and after-the-fact readjustments.
Indeed, the emerging technologies used will soon be ubiquitous on construction sites, says Vicki Satran, vice president of marketing for Computer Guidance Corp., www.computerguidance.com, Scottsdale, Ariz., developer of construction-site cloud-based ERP (enterprise-resource planning) solutions.
The construction industry will start seeing even more robots, AR (augmented reality)—the ability to visualize the real world through a camera lens—BIM (building information modeling), cloud and mobile technology, wearables such as Fitbits and 3D glasses, and increased prefabrication and modularization using prebuilt and off-site components, Satran says.
That’s because cloud, mobile, and Web-based technologies outmaneuver traditional client-based and on-premises systems. They’re less costly, and open the opportunity that the future demands with accessibility to data, information, and collaboration anytime and anywhere. Also, they are flexible, scalable, and ubiquitous, and that drives automation and intelligence, Satran adds.
Likewise, during on-site execution, being able to predict bottlenecks and slowdowns based on performance is incredibly powerful. Construction planners can use AI to see trends in historical performance and productivity rates, for example.
The biggest changes in the next few years will be around improving site efficiency, such as when to execute work, which work phase to give priority to, and how to enable as-late-as-possible material ordering and delivery, adds Daniel Patterson, chief design officer of InEight, www.ineight.com, Scottsdale, Ariz., which provides solutions that span projects from design to estimate and from field execution to turnover.
The upcoming efficiencies will be “responsive” to the ever-changing characteristics of the jobsite. “Today, project-execution planning is a very static process,” Patterson says. “The plans we develop don’t truly take into account the fluid, always-changing nature of the project. Being able to predict the impact of site changes and then plan accordingly will bring massive efficiency improvements to the construction industry.”
Yet Satran points out that workers’ tech skills need continuous upgrading, given the rapid change in technological applications. The future of work and the digital transformation will likely require more advanced skills, and workers must be prepared to meet the demand.
To get there, Satran says some employers will need to team up with outside parties, including their ERP trainers and consultants, universities, cities, and nonprofits to educate their workers.
The result will be huge improvements in efficiencies. Yet as AI becomes more prevalent, it generates massive data volumes.
“We will soon need to make a decision as to whether we archive these huge volumes of data or focus more on capturing and storing the patterns that AI can deduce and make inferences from,” Patterson says. “It is my belief that the real value from an AI perspective here is in the inferred patterns. Think of it as knowledge versus data. As AI gets smarter, knowledge will become more of a valuable asset.”
Still care is needed in what happens with that information, Patterson continues.
“We need to provide meaningful information to the relevant people at the right time,” he says. “Inundating a project team with fancy dashboards and statistics doesn’t help a project. In simple terms, does the nugget of information in question help a project stakeholder make a better decision or not?”
It’s also important to avoid just swallowing the data whole.
“The only word of warning would be that we should continue to question and challenge data just as we always have—before computers started making the suggestions,” Patterson says.
Data from the Site
A more informed and connected jobsite may well keep construction office managers and on-site workers from drowning in data transfers.
“The old-fashioned way, you’d do your best to lay out a point correctly,” Tim Jones, product manager for advanced layout at Hilti North America, www.hilti.com, Plano, Texas, says. “But you might not find out whether it’s accurate until you put up a steel column.”
Advanced sensors, such as those resulting from Hilti’s partnership with German-based Seuffer GmbH, www.seuffer.de, BadenWürttemberg, Germany, let contractors track variables by collecting data on how workers are using tools and vehicles.
“We can ensure construction partners that we’ve measured usage times, calculated torqueing, and taken into account other critical factors,” says Henrik Zetterqvist, business unit manager for direct fastening at Hilti North America. “We record the information and send it back to the user.”
One of the challenges on the jobsite is making sure the information is transmitted into the cloud, since some sensors have no beacon function or must have the function activated.
The tools also must have enough battery power to transmit the data.
Yet the goal is that the computer understands the context of a project’s location, type, and order of magnitude so it can make more informed suggestions about expected costs, durations, and productivity rates. This is a huge leap forward from simply doing a lookup in a database and pulling standard rates.
Taking it a step further, Lisa Duncan, director of vertical construction for Topcon Positioning Systems, www.topconpositioning.com, Livermore, Calif., adds construction companies are already using AI (artificial intelligence) on-site to visually tag data and analyze it for safety violations, potential hazards, and to mitigate many risks.
Other applications may include sorting notifications, identifying potential issues such as conflicts or missing materials, tagging, and organizing documents—and even piloting drones, running machinery, and helping in design.
“As you can see, the possibilities are wide-ranging for a construction firm to mitigate risk before it impacts that company’s project margins and performance factors,” she says.
Another way to use data efficiently is through 3D modeling, Duncan says. In vertical construction, companies use a 3D model to store or catalog data, whether that’s gathering data on a jobsite using a data collector with a robotic total station or inspecting photographs or point clouds from a terrestrial or drone scan.
“The key is to link the data from the field to the office and vice versa as quickly as possible, and with the highest amount of precision and accuracy,” Duncan says.
The cloud plays an essential role, too. Data sent to the cloud for processing lets both the field and office staff access it immediately, Duncan notes.
“The data gathered is based on the needs at the time of collection such as dashboard layout, notifications, parameters, and units of measure,” she says.
For the contractor, owners, and managers, analytics make it possible to compare what has been installed on a jobsite and if it is the correct location or to evaluate existing conditions on a jobsite to determine if the project is on schedule, Duncan says.
Even more, connected solutions let contractors get comprehensive monitoring of conditions, operations, and the work site through sensors and data.
“This means monitoring knows no boundaries,” Duncan says. “Assets can be monitored from everywhere, wherever the contractor happens to physically be. It enables realtime communication and task management with machines and crews on the work site. Additionally, the entire fleet of machines working at the site can be connected.”
Ultimately, construction companies can see “enormous productivity gains and start outpacing the industry” by adopting the latest digital tools, Jones says. If they fall behind, they won’t be able to bid competitively on future jobs. Once they get behind, the learning curve is steep.
Instead, construction companies and workers can use technology to interpret patterns from large volumes of data, rather than being content to simply generate large amounts of data.