drilling systems. Addressing these gaps presents opportunities for both academic research and industrial innovation, aligned with the broader objectives of Mining 4.0. 3. RESEARCH GAP AND NEED FOR INNOVATION Despite substantial progress in the digital transformation of mining operations, current implementations of drilling digitalization remain heterogeneous in scope, maturity, and impact. Recent research and industrial deployments highlight several areas where further advancements are required to fully realize the potential of Digital Mining or Mining 4.0, particularly in the context of large-scale production drilling. 3.1 Scalability of Interoperable Architectures There is limited literature on frameworks that seamlessly scale across entire mining fleets with heterogeneous equipment and legacy systems. A persistent challenge in drilling digitalization arises from the fragmentation of control systems and proprietary communication protocols across multi-vendor equipment fleets. Interoperability remains limited, inhibiting unified data access and cross-platform analytics. 3.2 Operational Integration of AI Feedback Loops Contemporary digitalization efforts in drilling operations are predominantly centered on descriptive analytics, emphasizing historical performance review rather than advancing toward predictive or prescriptive intelligence capable of forecasting outcomes and guiding corrective actions. As discussed by Wang et al., the integration of predictive machine learning approaches enables the estimation of key drilling performance variables, including rate of penetration and tool degradation patterns. Such predictive capability supports more proactive operational strategies, improves planning accuracy, and enhances decision-making processes by shifting the analytical focus from retrospective assessment to forward-looking optimization [6]. 3.3 Integration of Geological Context and Real-Time Drilling Feedback A recurring limitation in current systems is the inadequate integration of geological information with drilling telemetry. Real-time lithology recognition and hardness estimation remain largely manual or based on delayed post-processing. Recent works demonstrate that combining drilling telemetry with downhole sensors and geophysical logs improves real-time lithology classification using ensemble learning techniques. These integrated models enable dynamic adjustments to drilling parameters, reduce nonproductive time, and improve fragmentation outcomes. 3.4 Human-Machine Coordination Human cognitive processes to avoid automation complacency and maintain situational awareness. In drilling contexts, this translates to interactive visualization, explainable AI outputs, and intuitive interfaces that guide operators without obscuring underlying uncertainties.
RkJQdWJsaXNoZXIy MTM0Mzk2