NEWS: Measuring Construction Activity and AI Analysis Can Control Overtime

A study used artificial intelligence to analyze jobsite activity recorded on several construction sites and found correlations that would allow contractors to plan overtime and control its use for maximum benefit to the project

A new study by Dodge Construction Network and construction artificial-intelligence (AI) pioneer Versatile found that unexpected overtime can be predicted and controlled through regular job site activity measurement. The study found overtime predictable at an 88% confidence level, with proper measurement.

Overtime is a persistent feature of construction however it is often unplanned and unpredictable. Despite the cost of overtime, its impact on skilled workers, and its implications for safety and other key factors on a project site, it is often applied to address immediate concerns rather than planned to maximize its effects.

ALSO READ: Chinese construction firms using AI to monitor workers’ safety... but also to spot ‘loiterers’

The Dodge/Versatile study suggests that data and measurement of jobsite activities is key to understanding overtime and its impact.

“Unique insights derived from advanced data and analytics tools will empower construction crews to build better,” said Meirav Oren, cofounder and CEO of Versatile. “Overtime can be a very effective tool on the jobsite. Through the power of data, general contractors gain the ability to minimize unnecessary overtime while maximizing its strategic benefits.”

To conduct the study, jobsite productivity and performance data was collected through Versatile’s AI- and IoT-powered CraneView, a non-intrusive device mounted below the crane hook. Versatile analyzed data from a sample of project sites that deployed overtime to provide valuable insight into how measuring overtime can help improve productivity.

Key findings from the study include:

• 71% of crane processes during overtime are indirect activities to the actual project build.
• 96% of days with overtime had a higher volume of non-productive activities during the morning hours compared to the volume of total activities during overtime.
• Crane utilization was 6-7% better during overtime than for the same activities performed during standard shift.
• Overtime is strongly correlated to unscheduled breaks, with 63% of unscheduled breaks happening during the morning hours, or first part of the shift.
• “The goal of this report is to encourage conscious, intentional measurement of site activities and objective analysis to determine true best practices. This approach can help every kind of contractor with useful, actionable perspectives based on facts, rather than commonly held myths,” states Steve Jones, Senior Director of Industry Insights Research at Dodge Data & Analytics.

This initial study, Measuring What Matters: Overtime Efficiency, kicks off the companies’ co-authored series of reports on utilizing data and measurement on construction sites. Dodge and Versatile will analyze jobsite data to find other ways for contractors to improve productivity, safety and overall performance on projects, with more research expected in 2022.

The full findings of the study are available for free download at www.construction.com.