ALICE Technologies uses AI to solve construction scheduling challenges.
By Tim O’Connor, Senior Editor at Knighthouse Media
Planning a construction project is a monumentally complex task. It can take months to match the building sequence with the availability of contractors and materials – and even once the schedule is in place there’s no way to tell whether it’s truly the most efficient way to meet all of the project’s goals.
While pursuing his Ph.D. at Stanford, Rene Morkos began to think about how artificial intelligence could be applied to the problem of construction scheduling. The son of a builder, Morkos grew up around construction sites and understood the challenges contractors faced. He had a strong interest in the industry from the time he was young, eventually becoming a foreman during college and later a construction engineer. Through those roles, Morkos learned every facet of what goes into building a project and where the most likely bottlenecks can be found.
New equipment, modern materials and innovative construction techniques have enhanced the constructability of projects. Today, almost anything an architect can image can be built. But even in the midst of all those advancements, one thing that hasn’t changed is how contractors lay out the project. “Scheduling has been the same thing for 56 years,” Morkos says.
At Stanford, Morkos built a tool to solve the scheduling problems he had faced on construction sites in the United States, The Netherlands, United Arab Emirates, Lebanon and Afghanistan. The software used construction information to automatically generate a project schedule in a fraction of the time it would take a human. The potential was impressive and Morkos demonstrated it at a Stanford start-up competition in 2013, where he beat out 115 other entries and took home top honors.
The competition signaled the beginning of the company that would become ALICE Technologies. The next year, Morkos finished his doctorate and in early 2015 he formally started the company. At the time, Morkos had the core pieces of the product, but he still lacked an easy-to-understand user interface. It took another two years of tweaking and refinement before he was ready to begin beta testing ALICE – short for Artificial Intelligence Construction Engineering – on real-world projects. The product did not go into full release until early 2018.
Applying AI to construction scheduling required a number of breakthroughs before it was ready to use on a real project. The first was figuring out the building sequence that the schedule should follow. ALICE needed to break down the individual components of the building to understand the scope of the project. Fortunately, existing building information modeling (BIM) tools already contained most of the data that ALICE required, such as the location, quantity and types of materials that were going into an individual project. “There’s a lot of useful information encoded in BIM,” Morkos says.
Beyond the information available in BIM, the software understands the more subtle complexities of scheduling. For instance, it is aware that it takes time to move a crane from one location on the job site to another, and it accounts for those realities as it produces the project timeline.
By accessing the BIM, ALICE could determine the pieces it needed to build a project, but it still had to figure out everything that went into assembling those pieces. That’s where Morkos hit on his other big innovation: parametric scheduling.
Instead of telling ALICE how to build every individual column, window or girder, parametric scheduling uses a “Recipe” system. Engineers only need to tell ALICE how to build a single column and the program will then extrapolate that process to all other columns in the BIM. “You don’t tell the software how to build the whole project,” Morkos explains. “You tell it how to build a slab, then to replicate for all slabs.”
As a result of that breakthrough, ALICE is the world’s first parametric construction planner. “We cracked a problem that no one has been able to crack for 50 years,” Morkos boasts. “It’s 43,000 hours, eight years of my life.”
Because it is parametric, contractors can easily adjust recipes. It takes only a few seconds to adjust a planning parameter such as the number of available crews or crane locations and see how the change would impact the overall schedule. With that power, contractors and engineers can weigh priorities such as cost savings or expediting work to deliver the best possible project at the most cost-effective price.
In fact, clients can test all kinds of variables – what happens if they double the number of cranes or change to a less expensive material, for instance – and see the results in only a few minutes. That gives them the ability before they even bid on the project to explore all the what-if scenarios that were previously too time-consuming to calculate. Early adopters, like Rick Kahn Sr., director of innovation at Mortenson, are enjoying the benefits too. “The integration of our building expertise and ALICE enables us to evaluate hundreds of potential scenarios to determine the best possible plan for our customer,” he says. “ALICE is changing our industry and we are proud to be an early adopter of this leading technology.”
The software also benefits clients by freeing up their construction engineers to spend more time on project planning and internal system installation and less time crunching scheduling numbers. While ALICE still needs input from expert planners, it takes the pain out of scheduling. “We sit down, we create the plan,” Morkos says. “The plan is basically this list of tasks that feeds into the schedule. ALICE generates the different schedule options at the push of a button.”
That savings does not only come from the time it takes to put together the construction schedule but the efficiencies ALICE can discover in the process. A human can take months to put together one schedule, but in only 20 minutes ALICE can analyze hundreds of millions of schedules and identify the most optimal plan based on the client’s criteria. ALICE Technologies estimates that users can save up to 33 percent of construction time and costs. “We go through 600 million options and take the best one,” Morkos says.
The work going on in the background to analyze those hundreds of millions of schedule possibilities may be massively complex, but it all appears simple and straightforward on the front end. The resulting schedule is color-coded to make it clear to read and review and ALICE is designed to be easy to use and quick to learn. It only takes an hour of training before an engineer can set up a schedule for a simple project and after four hours users should be able to apply it to any kind of job.
The software itself is web-based, so it can be accessed at any time from any device anywhere in the world, allowing contractors to better track job site progress against the schedule. That’s helpful for when the timeline needs to be reworked because of outside factors such as a bad snowstorm or an unexpected equipment breakdown. “If something changes or is delayed, we can add that in,” Morkos says.
Since ALICE went into beta testing in summer 2017, ALICE Technologies has continued to improve the software based on user feedback and requested features, such as the ability to track the schedule’s history. “We’re getting some real stickiness,” Morkos says. “People are logging in and using it.
The software has been utilized by about a dozen projects, including four large-scale multibillion-dollar headquarters projects for well-known companies. Most of the projects have been located in the United States so far, but ALICE Technologies is also working with clients in Poland, Norway, Bangkok and Dubai. “We’ve tried this out on everything we can think of – oil and gas plants, hospitals – this stuff works,” Morkos says.
ALICE will only get better and more robust as time goes on. One of the next features Morkos wants to introduce is the ability to link construction crews directly with ALICE so that they are actually getting their instructions from the software each morning. That can save time when workers arrive at the job site. “We’re always adding layers to the artificial intelligence,” Morkos says. “That’s the name of the game, and it’s learning.”
To support ALICE’s future, ALICE Technologies is rapidly adding to its staff. The company now has 18 employees, most of who were hired in the past year. The expansion has surprised even Morkos. “If you asked me a year ago if I’d be at [nearly] 20 people today I would have said ‘no,’” he says.
Morkos takes a great amount of pride in reaching that milestone. Creating the technology behind ALICE has consumed his life for most of the past decade. Seeing it finally launch into the market and ALICE Technologies become a fully realized business are major successes for the engineer whose love for the construction industry began as a 10-year-old visiting his father’s work site. “I’ve put a lot into this,” he says. “The construction industry needs better software tools.”
Morkos believes the world is at a critical mass of innovation – universities and businesses from all over are working on exciting new technologies every day. Applying AI to the construction industry is a new concept for most people, but once they see how it works Morkos believes adoption will occur rapidly. “This is the easiest thing to sell,” Morkos says. “You can explain this until the cows come home, but just give her a project and run it and see for yourself.”