Oracle has announced the general availability of Oracle Construction and Engineering Advisor for Safety, an AI-powered predictive intelligence solution aimed at helping construction firms get ahead of jobsite accidents instead of just reacting to them after the fact.
The product is built on an industry-specific safety model that Oracle trained on data covering the equivalent of over 10,000 project-years, spanning diverse project types and locations. That is a notable distinction from older approaches, which required customers to supply large amounts of their own data just to get a working predictive model off the ground. Firms can plug in and start benefiting from predictions right away, regardless of how mature their existing safety program happens to be.
So what does it actually do? Each week, Advisor for Safety generates risk forecasts designed to surface the top 20 percent of projects most likely to account for 80 percent of safety incidents. The idea is that project managers can focus resources where risk is genuinely concentrated, rather than spreading attention thin across every job.
The system also provides prioritized corrective actions for high-risk projects, such as flagging a need for increased supervision in specific areas or zeroing in on hazardous zones. Whether those recommendations prove genuinely useful in the field or wind up being suggestions that experienced safety managers already know is something worth watching as real-world adoption grows.
Alongside Advisor for Safety, Oracle also introduced a new Observation capability for its Aconex and Primavera Unifier Accelerator platforms. The feature lets field teams capture structured safety data, including severity and frequency scoring, through a mobile device or web browser in a standardized format.
That structured data then feeds back into the predictive model, theoretically improving its accuracy over time as more observations come in. It is a sensible loop on paper, but the real question is whether field workers will actually log observations consistently enough for the model to benefit.
The solution pulls from multiple data streams: safety observations, incident reports, payroll data, and project schedules, sourced from Oracle Aconex, Oracle Primavera Unifier Accelerator, Oracle Fusion Cloud ERP, and third-party systems. Over time, firms can also fine-tune the model on their own proprietary data to sharpen predictions for their specific challenges or priority areas.
Oracle’s Josh Kanner, Senior Director of Analytics and AI for Oracle Construction and Engineering, put it plainly: “By predicting safety incidents and providing actionable insights, our customers can now focus on prevention rather than reaction.” Kanner also claimed the product has shown “customer reductions in incident rates by up to 50% or more and workers’ compensation costs by up to 75% in the first year.”
Those are big numbers, and Oracle does note they come with a footnote, so it is worth digging into the methodology before taking them at face value. That said, even more modest real-world improvements in construction safety would be genuinely meaningful, as the industry consistently ranks among the most dangerous for workers.
Mark Webster, Senior Vice President and General Manager of Oracle Infrastructure Industries, framed the broader goal this way: “organizations can immediately transition from reactive to predictive safety management, improving safety outcomes and reducing both human and financial costs associated with workplace injuries.”
Construction has historically been slow to adopt digital tools, so a solution that does not require a sophisticated internal data operation to get started is at least addressing a real barrier to entry. Advisor for Safety is now generally available as part of Oracle’s construction and engineering portfolio, and pricing details are available directly through Oracle.
