INSURANCE AND RISK MANAGEMENT Predicting Workplace Safety and


Protecting the Bottom Line General Contractors and Their Subcontractors Use Predictive Analytics to Better Manage Safety Programs and Costs


By Scott Marvell, Zurich North America


It’s no secret that con-


struction sites are ripe with safety hazards. OSHA con- siders construction a “high hazard industry” and con- sistently ranks construction as one of the most dangerous professions in America, ac- cording to Bureau of Labor Statistics in- formation released in 2014.1 California construction sites are not


immune to workplace safety hazards ei- ther; as recently as August 2014, 81citations with penalties of more than $135,000 were issued throughout California for various labor law and workplace safety violations, according to a news release from the De- partment of Industrial Relations for the State of California2


. And while the August 2014 investiga-


tion targeted contractors in California’s underground economy, it’s important to note that even the most legitimate general contractors experience fatal and costly ac- cidents at their worksites. As a result, general contractors are in-


creasingly investing in predictive analyt- ics — software driven data analysis that predicts future behaviors and events – to foresee and then prevent hazards on their worksites,


said Eric Lambert of Zurich


North America’s construction group. “Predictive analytics can be an effective


tool,” he said. “When combined with the right worker safety programs, it can make a significant difference in a project’s per- formance.”


Stopping Problems Before They Start Up to 80 percent of the general contrac-


tors on the ENR Top 400 are estimated to be using some form of data analytics to manage workplace safety, said Griffin Schultz of Predictive Solutions, a software and consulting firm focused on predicting workplace injuries.


12 March/April 2015


coming a key component of a world-class


“Predictive analytics safety


is be- program,”


Griffin said. “This information is used to find patterns around ac- cidents and injuries and develop the appropriate risk mitigation solutions.” Predictive analytics on a work-


site is used as follows: data around a contractor’s worksite history


and workplace safety program is fed into the modeling tool and compared against a database comprised of statistics from other projects. The models used by Predictive Solu-


tions, for example, include a dataset of more than 100 million observations from more than 1,500 worksites, and were de- veloped in collaboration with algorithm experts from Carnegie Mellon University. From this robust modeling, a predic-


tion of where a safety incident is most like- ly to occur on the job site is then provided to the general contractor. In conjunction with its risk manager or insurance provid- er, the general contractor can then develop a strategic action plan to help reduce the likelihood of future incidents. “Building a culture of safety at con-


struction sites requires a proactive ap- proach, and predictive analytics makes being proactive that much easier,” Lam- bert said.


Leading Indicators Lead The Way This new paradigm of data-driven safe-


ty is a distinct shift from the construction industry’s more traditional view of safety that focuses on past failures using regula- tory citations, litigation and loss ratios. “The problem with these traditional


metrics is that they are lagging; it’s too late to prevent an incident,” Lambert said. He suggested a more effective approach


is using leading indicators—the unsafe behaviors, conditions and near-miss inci- dents that can influence future construc- tion safety performance. “Lagging indicators reveal


there is a


problem, while leading indicators help identify the source of the problem and guide how to create effective solutions,” Lambert said. “Leading indicators are meant to be active metrics, something that prompts a proactive response to the pro- cess it measures.” Typically, data collection for a predic-


tive model concentrates on leading indica- tors like:  Near Miss Reporting  Project Management Team Safety Process Involvement


 Worker Observation Process  Stop Work Authority  Auditing Program  Pre-Task Planning  Housekeeping Program  Owner’s Project Manager (PM) Participates in Worker Orientation


 Foreman Feedback Meetings with Owner’s PM


 Owner Performs Safety Walk Through  Pre-Task Planning for Vendor Activities


 Vendor Design for Safety


The Burden of Safety Moves to the Boardroom By using these leading indicators and


advanced predictive analytics, some com- panies have reduced their injury rates by more than 90 percent, according to Schul- tz.


“The ROI on implementing a predictive


analytics solution for general contractors can include a reduced incident rate, re- duction in workers comp/insurance pay- ments, OSHA fine avoidance, a more en- gaged workforce and improved sales and marketing differentiation,” he said. Such success stems from more than


data mining, though, Lambert said. The data must then be incorporated into a ro- bust safety management program, typical- ly with the help of the general contractor’s risk manager or insurance provider. “General contractors have to go beyond


merely looking at the data and actually do something with it—whether it’s training,


California Constructor


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24