Framework for Modern Learner-Centric Technical Knowledge Management and Learning System

The oil and gas industry stands at a pivotal juncture, where traditional methods of workforce development and technical knowledge management are increasingly inadequate. Operational environments are becoming more complex, regulated, and digitized, yet training approaches often remain outdated. A growing wave of retirements, commonly termed the "great crew change," threatens the retention of decades of tacit knowledge, placing operational safety and continuity at risk (Dua, 2025; Sumbal et al., 2023). The scale of this challenge is underscored by recent studies indicating that human errors are a contributing factor in as many as 70% of incidents in the industry, a statistic that highlights the urgent need for more effective training and knowledge management systems (Peter Wilson, 2023). Conventional training practices, classroom sessions, document-based procedures, and informal mentorship—have historically formed the backbone of workforce development. However, they lack personalization, scalability, and contextual relevance in today's high-risk, data-rich environments (Arruda et al., 2024). One-size-fits-all models fail to address individual competency gaps, diverse job roles, or the dynamic nature of industrial operations. Static training materials quickly become obsolete, heightening risks of procedural errors and safety incidents. While many industries—such as aviation, pharmaceuticals, and advanced manufacturing—have embraced adaptive learning, realtime digital support tools, and AI-enhanced knowledge systems, oil and gas have been comparatively slow to adopt such innovations. This lag is due to several factors: highly conservative safety cultures, stringent regulatory environments, legacy infrastructure, and fragmented digital ecosystems. Moreover, frontline oil and gas roles frequently involve hazardous conditions, remote locations, and variable workflows that complicate technology deployment and require carefully engineered user experiences (Wanasinghe et al., 2021).