/ Profile / Future Forward: Advancing Construction Monitoring

Future Forward: Advancing Construction Monitoring

Todd Danielson on August 1, 2016 - in Profile

Mani Golparvar-Fard is associate professor of Civil Engineering and Computer Science as well as the director of the Real-time and Automated Monitoring and Control (RAAMAC) lab at University of Illinois at Urbana-Champaign. His award-winning work in the area of automated construction performance monitoring using video and imagery inputs and 4D building information models has been strengthened by partnering with construction companies to put these approaches into practice.

Organization

The RAAMAC lab focuses on answering fundamental research questions in visual data analytics for monitoring performance on construction projects. The analysis of video and images is tuned to recognize performance related to productivity, safety and quality.

The team takes inputs from any source data such as time-lapse cameras, smartphones and, more recently, camera drones.

“We do basic research, creating prototypes in the lab, and then partnering with construction companies to pilot these solutions to understand the value they add to construction processes,” says Golparvar-Fard.

Communicating Problems

A key application area of this research is to improve communication and understanding of actual construction performance and problems among contractors, subcontractors and owners.

“There are a number of reports—the most recent from McKinsey—showing that understanding on how much progress is being achieved on the jobsite every day is different among these groups,” notes Golparvar-Fard.

The main issue is the lack of timely and intuitive reporting on performance problems. It may take a week or more for these problems to be reported, and it can take longer to address them, therefore resulting in flawed performance management on job sites.

Improved Planning

Construction plans provide a helpful baseline to compare against actual construction progress. Detailed plans are combined with visual sensor inputs that use machine-learning algorithms to detect what’s taking place on the job site.

“Companies are achieving really good planning when it comes to creating a master schedule, and planning for that two months and even two years down the road,” says Golparvar-Fard. “When it comes to short-term planning, planning for subcontractors to execute tasks on a weekly or daily basis, there’s a gap in the ability to create reliable work plans.”

A major part of the gap has to do with measurement techniques and the fact that planners typically aren’t the people on the job site to understand the actual productivity rates. Also, the measurement techniques of Earned Value Analysis and Plan Percent Complete offer retroactive metrics where we only enable learning from past performances and do not directly offer any insight on the reliability of short-term and long-term plans.

“We need metrics to proactively identify potential problems before they surface on the jobsite,” says Golparvar-Fard. “We’ve been working on predictive data analytics and location-driven risk assessment techniques to make sure we can bring transparency to where the problems are and where they are likely going to be. We can also bring accountability by showing who was expected to perform a specific task at a given location on a given day.”

Increasing Inputs

The amount of imagery being captured is only growing through the use of apps (Latista, Plangrid, Fieldwire) and the companies that monitor sites with cameras and video (Multivista, EarthCam, JobsiteVisitor, OxBlue). Laser scanning also is growing, with some workflows scanning every concrete location before and after placement.

“There’s a huge amount of visual data that gets captured, mostly for liability issues and of course documentation of work in progress,” adds Golparvar-Fard. “We are tapping into the existing visual data to bring intelligence to it and map the images into project control workflows. We’re also linking planning to reality capture from camera drones to bring intelligence to it and offer metrics that are meaningful for project execution.”

To date, drone use is dramatically on the rise, with more than 40 percent of companies with FAA 333 exemptions using drones on construction sites. Research is focused on getting more from drone inputs, as their impact so far has largely been just an increase in image volume.

Quantified Return

The RAAMAC lab is focused on bringing research to pilot tests in real-world conditions that then can be integrated into construction project workflows. Careful monitoring and execution of pilot studies then are assessed for the value they brought to the project.

“In academia, the expectation is that we contribute to science and the body of knowledge,” notes Golparvar-Fard. “Industry expects us to have working solutions that have practical significance to them. To bridge this gap, my personal interest lies in bringing entrepreneurial thinking to my classes and research to help both graduate and undergraduate students take prototypes out of the labs and bring them into practice. This cycle does take time and requires funding from various sources. At the end of the day, it’s a validation of my own work if I’m contributing to the body of knowledge and creating something novel that has application and makes a difference in construction.”

ManiGolparvarFardVisit Informed Infrastructure online to read the full interview.

Todd Danielson

About Todd Danielson

Todd Danielson has been in trade technology media for 20 years, now the editorial director for V1 Media and all of its publications: Informed Infrastructure, Earth Imaging Journal, Sensors & Systems, Asian Surveying & Mapping, and the video news portal GeoSpatial Stream.

Comments are disabled