Quantifying Operational Effectiveness of Autonomous Drilling Platforms: A Systematic Approach to Defining Measures of Performance

Abstract An effective Model of Measurement of Effectiveness (MOE) has been proposed for Autonomous Drilling Platforms (ADP) that use a directional wellpath trajectory of drilling. The objective is to standardize the evaluation of Artificial Intelligence (AI) effectiveness employed in directional drilling. The framework, called i-AUTO, provides guidance for assessing ADP operations based on five aspects of mission utility: 1) Industry standards, 2) Adaptability to unforeseen scenarios, 3) Uncompromised performance in any condition, 4) Prioritizing insights over information, and 5) Optimal level of autonomy. i-AUTO connects drilling engineering with defense strategies to systematically evaluate ADPs. To clarify the prioritization of insight over information using ADPs, the authors used the concept of the Center of Gravity (CoG) (Clausewitz, 1997), which represents the sources of strength and balance in an operation (Echevarría, 2003). The authors identified several key sources that align with this concept: a) Data Integration and Analysis, b) Operational Efficiency and Automation, c) Integration of Cyber-Physical Sensors, and d) Decision-Making based on Insights and Tendency Patterns. By synthesizing these elements, the authors provided a comprehensive understanding of prioritizing insight over information, drawing parallels to the concept of the CoG and what should be considered essential in the operation of autonomous drilling platforms. This framework introduces a novel blueprint for operationalizing ADPs. It integrates technical and operational considerations and standardizes the definitions for reliability, repeatability, and consistency. According to the prior information, the purpose of this text is to discuss the possibility of creating a sustainable operational model that enables the smooth integration of ADP's operational concepts and human-machine teaming to drilling operations. So, the authors purpose is to explain how MOE, from military artificial intelligence (AI) programs, can be integrated into ADPs and provide a benchmark to evaluate these platforms performance1.