DNV in collaboration with the international energy company Equinor, has introduced a new incident taxonomy that helps organizations improve how they classify, understand, and learn from incidents. The taxonomy, called Event Learning Taxonomy CLUE, uses neutral language, fewer categories, and a systems-based approach to strengthen learning and reduce bias.
Many organizations rely on incident reporting systems to identify risk and support learning. However, existing taxonomies often use deficit-based language that encourages hindsight bias and individual blame. This approach limits learning and directs attention away from the system conditions that shape everyday work.
Event Learning Taxonomy CLUE addresses these challenges by reframing how organizations assess and communicate incidents. It supports clearer reporting, more constructive discussions, and better-informed actions to prevent reoccurrence.
The taxonomy applies to all types of safety, quality, and HSE incidents across a range of industries including construction, energy, manufacturing and maritime. Organizations can also use it to capture insights from normal work, including successful outcomes. By supporting learning from both failures and successes, Event Learning Taxonomy CLUE helps organizations build a more complete understanding of risk and performance.
Prajeev Rasiah, Senior Vice President & Regional Director, Energy Systems, Northern Europe at DNV said 'Meaningful safety improvement depends on how well organizations learn, not just from failures but from everyday work. Event Learning Taxonomy CLUE helps organizations move beyond blame and focus on understanding the conditions that influence performance. That shift enables better decisions, stronger learning, and more resilient operations.'
Event Learning Taxonomy CLUE differs from traditional taxonomies in three ways. First, it uses neutral, descriptive language that reduces judgement and implicit blame, supporting system focused learning for more effective risk-reduction strategies. Second, Event Learning Taxonomy CLUE includes fewer factors, making it easy to use for people without specialist investigation training. Finally, it replaces the concept of 'cause' with 'contributing factor' and uses a single analytical layer, reflecting how incidents emerge from interacting conditions rather than linear cause-effect relationships.
DNV and Equinor developed Event Learning Taxonomy CLUE principles from Human and Organizational Performance (HOP) and Safety-II proactive safety approaches, which highlight learning, context, and system design, as well as actual work practices under varying conditions.
Per Henry Gonsholt, Vice President, Safety at Equinor said 'Strengthened learning is a key part of Equinor’s safety roadmap. For better learning, we need to tell better stories. CLUE (Event Learning Taxonomy) helps us do that. It not only modernizes incident analysis but also simplifies it. This will help us get better data, and ultimately better insight into areas that need improvement.'
With Event Learning Taxonomy CLUE, DNV and Equinor aim to help organizations move beyond blame focused models toward more systemic, constructive learning that improves safety, quality, and operational performance.
Source: DNV
