In the recurring need to optimize drilling operations and reduce costs, a full RTOC (Real Time Operations Center) solution was deployed as part of the organization structure. To bring accurate, automatic and relatable data capture from surface sensors, the RTOC introduced a digital twin approach to improve field to town collaboration. The paper will demonstrate the benefits brought to operations by the solution in terms of risk identification and lessons learnt. RTOC digital twin solution integrates standard physical models’ workflows for hydraulics, torque and drag with advanced solutions using machine-learning algorithms. Capitalizing on operations recognition algorithm, the solution identifies thresholds and calibrates parameters to automatically classify operations into "Rig States" and "Drill States". The algorithm is trained to identify operational sequences and can derive complex measurements like downhole weight-on-bit and torque that are in turn fed into different workflows. This holistic event-based torque and drag baseline determination is used to define hole cleaning roadmap with minimum manual inputs. RTOC receives, processes and publishes the real time data on through its platform for all drilling and completion operations. This continuous process has enabled drilling operations team to assess and intervene on a need basis thanks to the clear event identification it offers. Amongst the digital workflows, the hole cleaning roadmap, combines modelled and automatically identified torque and drag data points rendered and shared with the stakeholders to ensure the capture of deviations and framing of potential risks to acceptable levels through a common decision platform. The clear output of single identifiable drilling event (such as pick up, slack off and free rotating weight) provides constant fact-based data for an adequate protocol to run casings and liners and refine engineering designs. In turn it has enabled to break casing and liner run records in their different operating fields. The drilling efficiency roadmap rely on quantitative algorithm and reliable output of downhole weight-on-bit, downhole torque and mechanical specific energy with automatic calibration, without user intervention nor bottom-hole-assembly modelling, allowing to substitute actual downhole measurements. This has been a performance enhancer in the improvement of rate of penetration regardless of the availability of downhole sensors. This new approach based on modern data science and digital twin based on a robust method, provides with a consistent and clear outcome regardless of service providers involved in the direct operations. It was trained, tested and validated prior to deployment, on more than 80 wells. This has also made possible the introduction of other algorithmic developments for Realtime dynamic modelling.

This project used predictive analytics and machine learning-based modeling to detect drilling anomalies, namely stuck pipe events. Analysis focused on historical drilling data and real-time operational data to address the limitations of physics-based modeling. This project was designed to enable drilling crews to minimize downtime and non-productive time through real-time anomaly management. The solution used data science techniques to overcome data consistency/quality issues and flag drilling anomalies leading to a stuck pipe event. Predictive machine learning models were deployed across seven wells in different fields. The models analyzed both historical and real-time data across various data channels to identify anomalies (difficulties that impact non-productive time). The modeling approach mimicked the behavior of drillers using surface parameters. Small deviations from normal behavior were identified based on combinations of surface parameters, and automated machine learning was used to accelerate and optimize the modeling process. The output was a risk score that flags deviations in rig surface parameters. During the development phase, multiple data science approaches were attempted to monitor the overall health of the drilling process. They analyzed both historical and real-time data from torque, hole depth and deviation, standpipe pressure, and various other data channels. The models detected drilling anomalies with a harmonic model accuracy of 80% and produced valid alerts on 96% of stuck pipe and tight hole events. The average forewarning was two hours. This allowed personnel ample time to make corrections before stuck pipe events could occur. This also enabled the drilling operator to save the company upwards of millions of dollars in drilling costs and downtime. This project introduced novel data aggregation and deep learning-based normal behavior modeling methods. It demonstrates the benefits of adopting predictive analytics and machine learning in drilling operations. The approach enabled operators to mitigate data issues and demonstrate real-time, high-frequency and high-accuracy predictions. As a result, the operator was able to significantly reduce non-productive time.

Majority of organizations endeavor to reduce operating costs and improve operational efficiencies. The concept of Mechanical Specific Energy (MSE) has long been implemented in the industry to improve drilling performance. The Drilling Real time Operations Center (RTOC) has taken the concept of MSE beyond its traditional approach by developing a Drilling Performance Measure combining data science and statistics to benchmark drilling efficiency. To extract maximum value from the available database, a workflow was developed to construct a Drilling Efficiency Benchmarking Tool. The different steps will be described for performing the data ingestion, cleansing, selection (offset well selection), methodology of computing the statistical model for MSE baseline per Formations and visualization of the output (charts and logs), to compare the actual MSE with baseline and thereby measuring the performance efficiency. The offset wells analysis results show that the workflow can construct an MSE baseline using high frequency data in a meaningful way, which is then set as a target envelope and projected through the real-time platform for monitoring and intervention purposes. This implementation of real-time MSE benchmarking helps identify the area of potential improvement, optimize drilling parameters to ultimately improve ROP and minimize lost time. As an analytical tool, it highlights achievable performance for each field and provide insights to consider new Best Practices.

Big data analytics is the often complex process of examining large andvaried data sets to uncover information. The aim of this paper is to describe how Real TimeOperation Center structuring drilling data in an informative and systematic manner throughdigital solution that can help organizations make informed business decisions and leverage business value to deliver wells efficiently and effectively. Real Time Operation Center process of collecting largechunks of structured/unstructured data, segregating and analyzing it and discovering thepatterns and other useful business insights from it. The methods were based on structuringa detailed workflow, RACI, quality check list for every single process of the provision of real-timedrilling data and digitally transform into valuable information through robust auditableprocess, quality standards and sophisticated software. The paper will explain RTOC DataManagement System and how it helped the organization determining which data is relevantand can be analyzed to drive better business decisions in the future. The big data platform, in-house built-in software, andautomated dashboards have helped the company build the links between different assets,analyzing technical gaps, creating opportunities and moving away from manual data entry(e.g. Excel) which was causing data errors, disconnection between information and wastedworker hours due to inefficiency. These solutions leverage analytics and unlock the valuefrom data to enhance operational efficiency, drive performance and maximize profitability. As a result, the company has successfully delivered 160 wells in 2019 (6% higher than 2019 Business Plan and 10% higher than number of delivered wellsin 2018) more efficiently with 28.2 days per 10kft fornew wells (10% better than 2018), without compromising the well objectives and quality of the wells. Moreover, despite increasing complexity, the highest level ofconfidence on data analytics has permitted the company to go beyond their normaloperating envelop and set a major record for drilling the world's fifth longest well as amilestone in 2019.

Drilling, tripping and running casing represents approximately fifty percent of the total well time, where the connection time KPI is the common performance indicator for those operations. Therefore, enabling real-time monitoring on drilling weight to weight and tripping connection time KPI's will add significant value through well time saving. The objective of this paper is to discuss the detailed implementation of machine learning to automate the detection and computation of the KPI's in real-time. The existing method for drilling performance monitoring requires extensive human data interpretation to calibrate the parameters required in this process. To overcome the complexity and reduce the human interaction, the automated Rig state and Drill state activity level were implemented based on Machine Learning (ML). The algorithm learns from the previous connections, drilling stand or tripping conditions to define the thresholds necessary to determine the current rig operation. With automatic rig activity detection, statistics to monitor the performance can be done in a systematic way. As a result, consistency of computation allows to compare performance and to improve it. The automated process using Machine Learning (ML) delivered consistent and powerful real time KPI computation, this helped to eliminate any human interpretation. This enabled real-time performance analysis delivery to rig site operations team. The machine learning model results were compared with the existing performance engine output and the comparison showed accurate and identical rig state/drill state detection and KPI's computation. The initial potential time saving with the implementation of this methodology is estimated around 15%, this was achieved through performance improvement on drilling and tripping connection KPI's. Further potential time saving can be achieved by extending the concept to track casing and liner running performance monitoring and other relevant drilling activities. This project introduces novel Rig state detection and KPI computation based on automated machine leaning model, demonstrating the benefits through improvement in drilling performance. The approach allows operators to mitigate data issues related with human interpretation and demonstrate real-time, high frequency and high-accuracy KPI's to significantly improve the drilling performance.