The integration of the Predict Au tool allows safety managers to analyze over ten million data patterns to identify high-risk sites months in advance. This initiative represents a core component of the NextGen Growth 2030 strategy, a vision that places the physical and mental well-being of the workforce at the heart of industrial operations. Historically, safety in heavy industry relied on reactive measures, where improvements followed unfortunate incidents. Under the leadership of Jeffrey Giesse, Group Head of HSE, the approach has undergone a fundamental shift toward a proactive culture. By synthesizing digital innovation with human-centric policies, the organization seeks to neutralize potential hazards before they ever manifest as physical threats. This transformation is not merely a technical upgrade but a philosophical pivot, ensuring that every operational decision is informed by data-driven insights. Such an evolution is essential in managing a global footprint that spans thousands of sites and involves tens of thousands of workers daily.
Advanced Predictive Modeling and Risk Detection
Anticipating Hazards with Predict Au
The centerpiece of this technological arsenal is the Predict Au model, which operates with the precision of an advanced algorithmic filter designed to detect subtle indicators of risk. By examining over 1,500 distinct data points, the system evaluates site-specific variables such as the frequency of leadership presence in the field and the speed at which identified hazards are mitigated. This predictive capability allows safety professionals to identify potential high-risk locations between five and ten months before a serious incident is statistically likely to occur.
Drawing from the global iCare platform, the AI continuously monitors leading indicators, ensuring that the window for preventative intervention remains wide. This long-term foresight enables regional managers to allocate resources effectively, addressing systemic weaknesses rather than just surface-level symptoms. The result is a highly sophisticated early-warning system that empowers personnel to maintain a constant state of vigilance across diverse operations. This methodology ensures that safety is managed as a rigorous science rather than a variable subject to chance or human error.
Maintaining Transparency Through the Fitness Score
To further enhance accountability across its vast global portfolio, the organization introduced a comprehensive metric known as the Fitness Score. This unified Key Performance Indicator provides a real-time rating of a site’s HSE program effectiveness by integrating traditional lagging indicators with dynamic risk signals. One of the primary functions of this score is to eliminate the phenomenon of statistical masking, where a high national performance average might hide a single facility operating under dangerous conditions.
In this system, if the risk indicators at one local site spike significantly, the entire country’s overall score reflects that change immediately. This structure ensures that localized issues remain visible to the Executive Committee, preventing complacency at the highest levels of management. By demanding transparency, the Fitness Score forces a collective responsibility for every individual site, regardless of its size. This approach creates a culture where the health of the entire organization depends on the safety of its parts, ensuring that operational risks are never buried in aggregated data.
The Intersection of Machine Intelligence and Human Leadership
Empowering the Workforce with Interactive AI
While the technology behind these systems is incredibly complex, the philosophy remains firmly grounded in human agency. The organization adheres to a “human in the loop” principle, ensuring that AI functions as a supportive assistant rather than a replacement for human judgment and experience. A prime example is the integration of an AI-powered coach named Flamey within the Boots on Ground application. This tool makes safety learning interactive and engaging, moving away from traditional instruction manuals toward a dynamic digital interaction.
By gamifying safety protocols, the platform encourages workers to participate more actively in the identification of hazards. This bottom-up engagement is critical for maintaining a culture where safety is viewed as a shared responsibility. The AI provides immediate feedback and guidance, helping employees to refine their safety skills in real-time. This interactive approach ensures that safety knowledge is not just acquired during training sessions but is continuously reinforced. This synergy between machine intelligence and human leadership creates a more responsive and protective safety environment.
Validating Results and Looking Toward Predict 2.0
The efficacy of these advanced digital tools was rigorously validated by independent third-party experts specializing in machine learning. Their analysis confirmed a direct and measurable correlation between the leading indicators identified by the AI and actual safety outcomes in the field. Sites that actively responded to AI-driven alerts by strengthening their internal controls and field presence successfully avoided critical accidents. Since the implementation of the platform, the organization observed a 63% reduction in the Lost Time Injury Frequency Rate, demonstrating the tangible impact of a digital safety culture.
The organization finalized the development of Predict 2.0 to further automate and refine data analysis across all global operations. This initiative focused on minimizing the time safety professionals spent on administrative data entry, allowing human experts to dedicate their energy to on-site interventions. These efforts established a new benchmark for safety management in the manufacturing industries, offering a clear blueprint for future standards. Integrating AI with field leadership proved to be the most effective strategy for reducing risk in high-stakes environments, ensuring that every worker returned home safely.
