Modern Society followed, and it is still largely following, a linear process for natural resources utilization: raw material extraction, goods manufacturing, use/consumption, waste generation. Such a model is proven to be not sustainable because it cannot go forever considering the limited quantity of available resources on Earth but also because of waste and process by-product management. A multitude of diverse initiatives to change this process started at different levels and within several industries in the last years including the O&G sector. Most of these initiatives share the same principle of "regeneration": waste and materials represent, in this view, the "feedback loop" able to make the production process a circular process, instead of an open-end one. Eni promotes and supports different initiatives to implement the principles of a Circular Economy and the objective of this paper is to describe a process implemented within Eni aimed at reducing the footprint of the Oil and Gas Industry with reference to material usage. Surplus and damaged materials are no more treated as waste, but they are re-engineered or re-conditioned (if necessary) in order to be redeployed to other projects within Eni affiliates all around the world. Nowadays this process is well structured and formalized within Eni and it is extensively applied involving all worldwide affiliates, reducing the overall CO2 footprint. Results achieved within Eni, in the last few years, averages between 6,500 and 17,500 ton of steel of material redeployed among Eni's affiliates for a value ranging between 26 and 71 million USD. The overall average result is 23,000 CO2 equivalent ton not released per year and 242,000 GJ of energy saved (ref. to steel manufacturing estimated impact:1.9 ton di CO2/ton steel cast and 20 GJ/ton steel cast). Extending this process to involve material and equipment manufacturers, it is possible to improve the whole supply process reducing at the "source" the material storage needs, the material surplus and the produced wastes, including the CO2 emission produced in the transport phase. Initiatives like "Just in Time" delivery and material "Buy Back", mainly applied in Countries where Framework Agreements are already in place between Eni and its manufacturers and where manufacturers have their production sites and other facilities, are essential to achieve this target. Eventually, applying the approach to the whole supply chain and operations management will allow to reduce the "last mile" warehousing and transportation needs, including the dimension and capability of the operations fleet.

At the early stages of field life, the subsurface project team operates under lack of information. Due to the high uncertainties, decisions at the exploration and appraisal stages are often influenced by cognitive distortion that leads to overestimation or underestimation of hydrocarbon reserves and, as a result, to suboptimal investment decisions. World practice allows us to identify the most common causes of cognitive bias: the team focus on the most provable according to their view scenario and may ignore data that contradicts the chosen scenario, the opinions of the team members differ in the choice of the most likely scenario, the team members work with geological and geophysical (G&G) data performing separate tasks and may miss important connections between various sources of information. The consequences of these cognitive distortions cause an increase in risk capital, the duration of exploration activities, and the choice of suboptimal field developmentstrategy resulting in a decrease in the effectiveness of the exploration program and the project as a whole. To reduce such risks, it is possible to attract subject matter experts with extensive experience to support the project team. But the amount of experts is limited and this approach cannot be implemented for the entire portfolio of exploration projects. As result of a research project of Gazpromneft in a partnership with IBM Research, an innovative approach was developed for the objective integration of geological and geophysical data. The main idea of this approach is to support the geologist's decisions by an intelligent assistant working on the principles of the modern theory of knowledge engineering. Using the generalized expert knowledge, the intelligent assistant impartially integrates disparate geological information into a set of conceptual geological models (scenarios, objectively evaluates their probabilities, and helps to plan optimal exploration/appraisal activities.

This paper considers the problem of steady-state optimal resource allocation in an industrial symbiotic oil production network, or in general, a large-scale oil production system network, where different organizations share common resources. These allocation problems are typically solved in a distributed optimization framework, where the optimization problem is decomposed into smaller subproblems, a central coordinator is used to coordinate the different subproblems. However, the use of a central coordinator may introduce additional practical challenges, such as impartiality issues, or additional operating costs, which is undesirable even in the technological selection phase. To eliminate the need for a central coordinator, this paper proposes a consensus-based optimal resource allocation, where each subproblem or organization is locally optimized, and the coupling constraints are negotiated among the different organizations over a fixed communication network with limited information exchange. The proposed approach is applied to a large-scale subsea oil production system, where the different wells are operated by different organizations. The simulation results of the application show that the proposed approach can optimally allocate the shared resources.

Reservoir characterization requires accurate prediction of multiple petrophysical properties such as bulk density (or acoustic impedance), porosity, and permeability. However, it remains a big challenge in heterogeneous reservoirs due to significant diagenetic impacts including dissolution, dolomitization, cementation, and fracturing. Most well logs lack the resolution to obtain rock properties in detail in a heterogenous formation. Therefore, it is pertinent to integrate core images into the prediction workflow. This study presents a new approach to solve the problem of obtaining the high-resolution multiple petrophysical properties, by combining machine learning (ML) algorithms and computer vision (CV) techniques. The methodology can be used to automate the process of core data analysis with a minimum number of plugs, thus reducing human effort and cost and improving accuracy. The workflow consists of conditioning and extracting features from core images, correlating well logs and core analysis with those features to build ML models, and applying the models on new cores for petrophysical properties predictions. The core images are preprocessed and analyzed using color models and texture recognition, to extract image characteristics and core textures. The image features are then aggregated into a profile in depth, resampled and aligned with well logs and core analysis. The ML regression models, including classification and regression trees (CART) and deep neural network (DNN), are trained and validated from the filtered training samples of relevant features and target petrophysical properties. The models are then tested on a blind test dataset to evaluate the prediction performance, to predict target petrophysical properties of grain density, porosity and permeability. The profile of histograms of each target property are computed to analyze the data distribution. The feature vectors are extracted from CV analysis of core images and gamma ray logs. The importance of each feature is generated by CART model to individual target, which may be used to reduce model complexity of future model building. The model performances are evaluated and compared on each target. We achieved reasonably good correlation and accuracy on the models, for example, porosity R2=49.7% and RMSE=2.4 p.u., and logarithmic permeability R2=57.8% and RMSE=0.53. The field case demonstrates that inclusion of core image attributes can improve petrophysical regression in heterogenous reservoirs. It can be extended to a multi-well setting to generate vertical distribution of petrophysical properties which can be integrated into reservoir modeling and characterization. Machine leaning algorithms can help automate the workflow and be flexible to be adjusted to take various inputs for prediction.

This study investigated the ability to produce accurate multiphase flow profiles simulating the response of producing reservoirs, using Generative Deep Learning (GDL) methods. Historical production data from numerical simulators were used to train a GDL model that was then used to predict the output of new wells in unseen locations. This work describes a procedure in which data analysis techniques are used to gain insight into reservoir flow behavior at a field level based on existing historical data. The procedure includes clustering, dimensionality reduction, correlation, in addition to novel interpretation methodologies that synthesize the results from reservoir simulation output, characterizing flow conditions. The insight was then used to build and train a GDL algorithm that reproduces the multiphase reservoir behavior for unseen operational conditions with high accuracy. The trained algorithm can be used to further generate new predictions of the reservoir response under operational conditions for which we do not have previous examples in the training data set. We found that the GDL algorithm can be used as a robust multiphase flow simulator. In addition, we showed that the physics of flow can be captured and manipulated in the GDL latent space after training to reproduce different physical effects that did not exist in the original training data set. Applying the methodology to the problem of determining multiphase production rate from new producing wells in undrilled locations showed positive results. The methodology was tested successfully in predicting multiphase production under different scenarios including multiwell channelized and heterogeneous reservoirs. Comparison with other shallow supervised algorithms demonstrated improvements realized by the proposed methodology, compared to existing methods. The study developed a novel methodology to interpret both data and GDL algorithms, geared towards improving reservoir management. The method was able to predict the performance of new wells in previously undrilled locations without using a reservoir simulator.

A major Russian oil company is currently carrying through an ambitious program aimed at transforming corporate E&P business model. The new model involving product-based approach to exploration and production will require young professionals with new skills and mindsets beyond regular university curricula. To proactively satisfy this demand, the company joined forces with one of its partner universities to champion Engineers of the Future, a training initiative aimed at senior students about to graduate and join the company. Engineers of the Future offer a fresh perspective and approach to training young professionals, mixing conventional training with problem-based and game-based learning to deliver a unique combination of hard and soft skills required by company's new operating paradigm. Program graduates are expected to make a great addition to corporate product teams, enabling the company to achieve its challenging strategic goals.

After the oil price crashes in 2014 & 2020 several M&A deals ended up in legal debates because operators cancelled major projects or infills wells that were booked in the "probable" reserves only. This document challenges the compatibility between the deterministic incremental reserve assessment method (PRMS2018, chapter 4.2.1.3), and the concept of split condition (PRMS2018 chapter 2.2.0.3), which is not allowed for reserves booking under PRMS. With a few examples, we explain why the incremental method may be misleading investors, if used wrongly. Policies, stock market requirements, or simply the understanding of reserves guidelines may differ from one company to another. Many filers and auditors are still keen on using the deterministic incremental approach. This method consists in "defining discrete parts or segments of the accumulation that reflect high, best, and low confidence regarding the estimates of recoverable quantities under the defined development plan". In principle, this should give similar result to the widely accepted scenario method (PRMS2018, chapter 4.2.1.3) but in reality, major discrepancies are observed. Some reserve evaluation may also become misleading for banks, investors, or even for good asset management. I many cases, the estimation of recoverable volumes is reasonable, but it does not match the company CAPEX requirements, affecting corporate cash flow as well as potential Reserves Based Lending (RBL) requirements. In another case, the 1P case will be robust, but the 2P may be grossly overestimated, affecting M&A or share price. "Reserves guidelines are principle based" this has recently become a very fashionable statement in the context of SEC bookings. Similar discussions will also occur when reviewing PRMS reports. However, different interpretations for keywords such as "Project", "Spit condition", or "FID" should not prevent the evaluator to provide a reliable reserves estimation to investor or company management. This document questions the threshold where ethics disappears, and a Madoff scheme may become legal.

Trips and failures are common occurrences in the Electric Submersible Pump (ESP) systems. The random nature of these trips and failures will lead to low industry run-life and high workover costs for ESP companies and operators. To perform early detection and take corrective actions to handle the potential incidents, ESP operation data collected from downhole and surface sensors are used to perform diagnostics and prognostics to identify trips and failures. In this study, Principal Component Analysis (PCA) method serves as a pre-processing method to retain the most essential principal components to reevaluate the initial ESP system. For a single well system, the Squared Prediction Error (SPE) and Hotelling T-square statistic (T 2 ) equations are employed for numerical visualization in the new principal component space and therefore detection of the potential ESP trips or failures. For the whole well group, the score plot of three principal components provides a solution that enables to distinguish different clusters of stable operation, trip and failure regions, and diagnose the upcoming ESP trips and failures. In this way, the predictive model is bulit to continuously analyze the ESP operation and automatically perform health monitoring for any ESP system. This paper concludes that the predictive model has the potential to construct a real-time proactive surveillance system to identify dynamic anomalies and therefore predict developing trips or failures in the ESP system.

Interwell tracers are powerful reservoir surveillance tools that provide direct reservoir flow paths and dynamics, which, when integrated with near real-time production optimization, can greatly improve recovery factor, and return on investment, the so-called "Advanced Tracers System" (ATS). Applying full field ATS is attractive for resource-holders, especially for those with large waterflood operations. However, to scale up ATS to cover large fields with potentially tens to hundreds of injectors and producers, the required unique tracer variations ("barcodes") and materials and associated analysis may increase rapidly. Here, we explore different tracer injection schemes that can acquire the most information while using reduced numbers of tracers, thereby controlling costs in field operations. We tested the designs of various modified tracer injection schemes with reservoir simulations. Numerical experiments were performed on synthetic fields with multiple injector and producer wells in waterflooding patterns. Two tracer injection schemes were considered: In Scheme 1, all injectors were injected with unique tracers representing the most information-rich case. In Scheme 2, some injectors were injected with the same tracers ("recycling" the same barcodes), and some injectors received no tracer injection ("null" barcodes). Production and tracer breakthrough data was collected for history matching after waterflooding simulations on the synthetic fields. The ensemble smoother with multiple data assimilation with tracers algorithm was used for history matching. We calculated the root-mean-square errors (RMSE) between the reference data and the history matched production simulation data. To improve the statistics, 20 independent testing reference synthetic fields were constructed by randomizing the number and locations of high permeability zones crossing different injectors and producers. In all cases, the history matching algorithms largely reduced the RMSE thereby enhancing reservoir characterization. Analyzing the statistical significance with p -values among testing cases, first, as expected, the data mismatch is highly significantly lower after history matching than before history matching ( p < 0.001). Second, the data mismatch is even lower when history matching with tracers (both in Scheme 1 and 2) than without tracers ( p < 0.05), demonstrating clearly that tracers can provide extra information for the reservoir dynamics. Finally, and most importantly, history matching with tracers in Scheme 1 or in Scheme 2 result in statistically the same data mismatch ( p > 0.05), indicating the cost-saving "recycling" and "null" tracer barcodes can provide equally competent reservoir information. To the best of our knowledge, this is the first study that evaluated the history matching qualities deriving from different tracer injection schemes. We showed that through optimal designs of the tracer injections, we can acquire very similar information with reduced tracer materials and barcodes, thus reducing costs and field operational complexities. We believe this study facilitates the deployment of large-scale reservoir monitoring and optimization campaigns using tracers such as ATS.

Objectives/Scope Historically, facility outages, cyber-attacks, natural disasters, supply interruptions and other disruptions have caused significant impacts to the flow of crude oil and petroleum products. The impacts on assets, primary and secondary markets, and economic indices need to be quantified to address the effects of disruptions efficiently and effectively. Methods, Procedures, Process This paper details the design process that was used to develop an analytical tool to predict the impacts from disruptions, determine the cascading impacts on downstream markets and dependent assets and provide estimates for recovery potential. The entire petroleum supply chain, upstream, midstream, and downstream facilities, were included in the development of the modeling capability. To best capture the supply chain network and geographic dependencies, the area of interest was divided into finite regional markets based on supply and demand market dynamics and connectivity. Extensive review was conducted of historical disruptions to understand the impacts on infrastructure and the petroleum supply chain network interdependencies to support a holistic approach to disruption simulation. Algorithms were developed to determine the probability and severity of damage of disruptions at varying intensities. The modeling capability was designed to utilize a comprehensive up-to-date database and geospatial system. The database was designed considering public, private and proprietary data, the frequency of updates, validity of source, and value added. Results, Observations, Conclusions The petroleum disruption analytical modeling tool that was developed assesses disruptions to the supply chain network and predicts the duration and severity of impacts on facilities and the cascading effect on primary and secondary markets. The tool provides disruption results in tabular, graphical, and geospatial forms for individual assets, regions, and the nation as a whole. For the United States, the tool models 26 geographical regions delineated by refining sectors, mainly along the coastlines, and dependent demand markets. The crude oil and petroleum product supply and demand is balanced for each region using local supply, net imports, interregional connectivity, and the multi-modal transportation network. This paper demonstrates how the analysis tool provides business intelligence insights for the days of stocks available throughout the disruption duration and the total loss in products to the markets. An economic submodule was integrated with the tool that determines the impacts on the crude oil and gasoline price and the resulting impact on Gross Domestic Product (GDP). The model was benchmarked during Hurricane Laura 2020 and the predicted reduction in refining production was 85% of realized losses. Novel/Additive Information The innovative analytical tool simulates disruptions and provides predicted forecast of facility, market, and economic impacts for an extended period of time after the event occurs supporting response, recovery, and planning efforts. The model can be assimilated for any geographical or geopolitical region with consideration given to region specific disruptions. This paper provides case studies exemplifying use cases and model simulated results for petroleum supply disruptions.

It is common to have measured depth exceeding 20,000 ft for unconventional oil and gas wells. To ensure the pressure pulse can be detected on the surface, many MWD tools have been designed to generate mud pressure pulse with very large amplitude. While the large pressure pulse solved the problem of sending the measured information up to the surface, it creates significant impact on drilling system energy variation and downhole drilling dynamics. This paper focuses on understanding the effects using big data and drilling system modeling. When a commonly used MWD tool generates mud pulse sequence, it chokes the flow path at designed patterns. This creates mud flow variation in the mud motor below the MWD tool. It also generates axial force variations due to pressure changes, which affect WOB. These changes cause the motor and the bit to experience significant rpm variations. The combined rpm variation and WOB variation often excite more severe axial and lateral shock and vibration. These effects are quantified by thousands of high-frequency downhole datasets and advanced numerical modeling. In the high-frequency downhole datasets, some of them are obtained from BHAs with MWD tools generating large mud pressure pulse, and some of them are obtained from BHAs with MWD tools generating smaller mud pressure pulse or transmitting the measurements using electromagnetic signal. Statistics of rpm variation and axial and lateral shock and vibrations are compared. It clearly shows that the BHAs utilizing large mud pressure pulse experience more severe torsional, axial, and lateral vibrations. When looking into specific datasets, it showed that mud pressure pulse could cause the motor to lose more than half of its rpm during the flow choking phase. Typical datasets indicate that mud pressure pulse correlates to severe high-frequency torsional oscillation (HFTO) in motorized rotary steerable BHA. An advanced transient drilling dynamics model was built to simulate the whole drilling system subjecting to mud pressure pulse incurred loading conditions. It was found that large-magnitude mud pressure pulse induced more stick/slip and axial and lateral vibrations as recorded in downhole high-frequency data. The increased rotational, axial, and lateral vibrations correspond to more loading variations in the mud motor components and PDC cutters on the drill bit. These variations could cause accelerated damage to the drill bit and downhole tools. In summary, large mud pressure pulse utilized by some MWD tools introduces significant rpm variation and shock and vibration, which is quantified by big data and further demonstrated by drilling system modeling. The information could help make decisions on BHA design and tool selection to achieve improved drilling performance and reduce the risk of premature tool failure.

The post COVID-19 era will undoubtedly present paradigm shifts in operational planning and execution and advanced automation will become an important factor. However, drilling automation without directional drilling ( Cayeux 2020 ) capability will exclude the use of automation in a vast number of fields where precise placement of the wellbore has shifted from a luxury to a necessity. This is important in unconventional plays where automation can make a step change in operational outcomes ( Chmela 2020 ). However, most efforts in automating directional drilling are using bespoke rigs ( Slagmulder 2016 ) and bespoke bottom hole assembly (BHA) that limit operational options. The goal is in designing systems that enable directional drilling automation ( Chatar 2018 ) with existing BHAs. This paper will look at three challenges that were identified and overcome to deploy a vendor agnostic system for automating the directional drilling (DD) process. The three challenges identified here are as follows: Using any mud motor including low-cost motors in a closed loop Integration with an existing measurement and logging while drilling (MLWD) system Ability to roll out automation systems on any operations with existing rigs The system is a modification of an operator’s autonomous drilling system ( Rassenfoss 2011 ), designed to use existing rigs, BHAs and have minimum footprint on the rigs for operational use. The system will have a dedicated connection to the rig’s programmable logic controller (PLC) via common industrial protocols including Modbus, EthernetIP or Profinet, a physical connection the MLWD receiver and a brain box with a cloud connection to aggregate, process data and send commands to the rig PLC to execute directional commands. A vendor agnostic system will increase adoption of automated technologies and further drive improvements in operational and business performance.

Knowledge of well connectivity in a reservoir is crucial, especially for early-stage field development and water injection management. However, traditional interference tests can often take several weeks or even longer depending on the distance between wells and the hydraulic diffusivity of the reservoir. Therefore, instead of physically shutting in production wells, we can take advantage of deep learning methods to perform virtual interference tests. In this study, we first used the historical field data to train the deep learning model, a modified Long- and Short-term Time-series network (LSTNet). This model combines the Convolution Neural Network (CNN) to extract short-term local dependency patterns, the Recurrent Neural Network (RNN) to discover long-term patterns for time series trends, and a traditional autoregressive model to alleviate the scale insensitive problem. To address the time-lag issue in signal propagation, we employed a skip-recurrent structure that extends the existing RNN structure by connecting a current state with a previous state when the flow rate signal from an adjacent well starts to impact the observation well. In addition, we found that wells connected to the same manifold usually have similar liquid production patterns, which can lead to false causation of subsurface pressure communication. Thus we enhanced the model performance by using external feature differences to remove the surface connection in the data, thereby reducing input similarity. This enhancement can also amplify the weak signal and thus distinguish input signals. To examine the deep learning model, we used the datasets generated from Norne Field with two different geological settings: sealing and nonsealing cases. The production wells are placed at two sides of the fault to test the false-negative prediction. With these improvements and with parameter tuning, the modified LSTNet model could successfully indicate the well connectivity for the nonsealing cases and reveal the sealing structures in the sealing cases based on the historical data. The deep learning method we employed in this work can predict well pressure without using hand-crafted features, which are usually formed based on flow patterns and geological settings. Thus, this method should be applicable to general cases and more intuitive. Furthermore, this virtual interference test with a deep learning framework can avoid production loss.

This paper aims to present user experience survey results of innovative software assessment technologies available in the market aimed in evaluating risks of industrial-related ergonomic hazards. The scope covers industrial ergonomics softwares currently available for purchase where time-limited free trial is offered, particularly those that utilize 3D Motion Capture Assessment which relies on kinematic inputs aided by non-invasive computer technology and artificial intelligence, and makes use of pre-determined weightings based on biomechanical risk factors. In light of the inclusion criterion, six industrial ergonomics softwares were considered. User field trials were conducted during January to September 2020 among 10 Occupational Health Subject Matter Experts (OH SME) coming from seven oil and gas Group Companies. Each OH SME attended a product demonstration by the shortlisted software vendor, participated in software trial at their respective workplaces, and provided feedback on the software's usability by filling out a survey questionnaire. OH SME responses were then collected for further qualitative analyses. Three of the eligible softwares relied on photo snapshot capturing work activity where subsequent analysis is done through competent professional judgment of qualitative risk. Another three were dependent on 3D Motion Capture Assessment where upper and lower limb motions of employees are digitally captured, recorded, and analyzed. Two of the softwares utilized sensors attached to different parts of employee's body, while one relied on Android/Smartphone snapshot of work activity and analyzed by the software's algorithm. Analyses of OH SME feedback revealed majority of them (n = 7) preferred using 3D Motion Capture Assessment over professional judgment of qualitative risk as an effective tool in evaluation of industrial work-related ergonomic risks. 3D Motion Capture Assessment provided accurate measurements of employee joint postures and postural angles. The tool ensured consistency in risk scoring for a particular industrial-related work activity as the calculation is standardized. The tool's algorithm is aligned with globally accepted assessment tools in evaluating ergonomic risks which enhances its validity. OH SMEs have expressed concerns on use of Android/Smartphone in Critical Infrastructure and Coastal Protection Authority facilities, training time needed in learning the software, and repetitive use of motion sensors among different employees which may lead to personal hygiene issues. 3D Motion Capture Assessment is a novel ergonomics software tool that can be used in real-time and accurate evaluation of ergonomic risks arising from industrial work-related activities. It can replace observational assessment of a work activity that may be prone to professional judgment errors. However, more validation and reliability studies need to be done in future as well as determining association between ergonomics risk scores obtained from the software and prevalence of work-related musculoskeletal disorders.

A decision-centric approach to projects creates confidence, improves value, and shortens time to revenue. A straight-forward objective based approach to managing project decisions is presented in the form of four primary questions. Those questions are: Does the issue/threat/opportunity make a material difference to the project? (Materiality) Can anything be done to affect the outcome? (Influence) Can you afford to do anything about it? (Value) What if you are wrong? (Confidence) Materiality – An issue/threat/opportunity must make a material difference to a project decision to be worth receiving attention. The concept of materiality will vary in size and consequence from project to project, so it is important to maintain a decision focus. Understanding the variability in the project with respect to decision thresholds can provide an indication of materiality. We must also ask how different our current assessment of the project, or its environment could get before we would like to change our decision. Affective ability – Accepted risk-management options of avoidance, mitigation, transfer, and acceptance present the decision options within this category. In considering the consequences the options, decision tools such as Indifference Assessment and Pain and Regret Assessment. Avoidance, Mitigation, or Transfer – while most projects can benefit through risk reduction, such effort must make economic sense. Risk reduction paths must add value to the project through added upside or elimination of at least a portion of downside threat. The value of these efforts is aided by use of tools such as Value-of-Information, Value-of-Control, and Value-of-learning. Being wrong – Making a regretful decision is always a possibility but the source of the "wrongness" and its likely impact is often overlooked by teams. In project planning and execution, decision-makers are often presented with a plethora of issues, threats, and opportunities. From development planning through implementation significant time and resource waste can be cut by prioritizing effort to the issues that matter. Understanding the issues in the context of materiality and then what to do, if anything, about an issue, becomes key to maximizing success. This approach cuts waste and focuses the attention on what matters. Decision Intelligence not only increases the probability of making the best decisions, but it also prioritizes work to those items that matter either for value or decision path. While most of the decision tools referenced are well documented in the literature, placing them into the context of the Four Question Approach allows teams and management to focus more closely on efficiently mitigating issues, shortening workflow, and creating significantly higher decision confidence. This novel approach works well in all phases of project planning through project management implementation.

The focus of this work is on developing a cognitive tool that predicts the most frequent HSE hazards with the highest potential severity levels. The tool identifies these risks using a natural language processing algorithm on HSE leading and lagging indicator reports submitted to an oilfield services company’s global HSE reporting system. The purpose of the tool is to prioritize proactive actions and provide focus to raise workforce awareness. A natural language processing algorithm was developed to identify priority HSE risks based on potential severity levels and frequency of occurrence. The algorithm uses vectorization, compression, and clustering methods to categorize the risks by potential severity and frequency using a formulated risk index methodology. In the pilot study, a user interface was developed to configure the frequency and the number of the prioritized HSE risks that are to be communicated from the tool to those employees who opted to receive the information in a given location. From this pilot study using data reported in the company’s online HSE reporting system, the algorithm successfully identified five priority HSE risks across different hazard categories based on the risk index. Using a high volume of reporting data, the risk index factored multiple coefficients such as severity levels, frequency and cluster tightness to prioritize the HSE risks. The observations at each stage of the developed algorithm are as follows: In the data cleaning stage, all stop words (such as a, and, the) were removed, followed by tokenization to divide text in the HSE reports into tokens and remove punctuation. In the vectorization stage, many vectors were formed using the Term Frequency - Inverse Document Frequency (TF-IDF) method. In the compression stage, an autoencoder removed the noise from the input data. In the agglomerative clustering stage, HSE reports with similar words were grouped into clusters and the number of clusters generated per category were in the range of three to five. The novelty of this approach is its ability to prioritize a location’s HSE risks using an algorithm containing natural language processing techniques. This cognitive tool treats reported HSE information as data to identify and flag priority HSE risks factoring in the frequency of similar reports and their associated severity levels. The proof of concept has demonstrated the potential ability of the tool. The next stage would be to test predictive capabilities for injury prevention.

Gas production from unconventional shale reservoirs is known for rapid declines. Intermittent shut-in production constitutes a technique typically applied to low-production wells during late life stages to maintain economic rates. This technique involves a cyclic process of shutting in the well temporarily to allow it to build up pressure and subsequently switching the well to production. Operators often manage hundreds of wells on intermittent shut-in production; these wells, however, incur different shut-in and production cycle times, thus requiring a complicated management approach. Because every well has a unique production behavior and reservoir characteristics, searching for optimum operational conditions individually is not only technically challenging, but also operationally time-consuming and labor- intensive. Our goal was to use active learning analytic, a type of machine learning deployed on an edge computing platform, to autonomously control and optimize these unconventional gas wells. The field trial results show increased production, reduced liquid loading, decreased manual intervention, and reduced carbon footprint. Our solution utilizes an edge computing platform to deploy the analytic on the wellhead without requiring a stable internet connection. A computing device at the edge connects to controllers on site, processes data, sets system control parameters, and enables automation for operations deploying an optimization algorithm. Active learning algorithms are valuable for use in the optimization of systems that are not mathematically definable. These algorithms are also proven to learn the relationship between the inputs and outputs and use prior knowledge to intelligently search for the optimum settings within the defined operating limits. The low latency of edge computing allows for high-frequency data collection in seconds and a rapid control of the wells. The edge device continuously monitors production and initiates re- optimization as needed when operational conditions change. We developed an analytic that autonomously controls the intermittent production technique where a well is shut-in based on a specified minimum gas production rate and opened when the pressure builds up to the specified target during the shut-in period. The analytic actively learns and measures the ways in which the specified parameters improve production rates. Additionally, the analytic continuously monitors production data and identifies any well liquid loading events. When liquid loading occurs in the wellbore as observed from the production pattern, the analytic automatically shuts in the well to build up pressure and minimizes additional liquid formation. In the field trial, we deployed the edge analytic to monitor gas production and the specified well shut-in and open conditions for 10 different wells in the Haynesville Shale Play. Analyzing each well in the context of approximately 30 intermittent production cycles (shut- in/open), the analytic successfully mapped the surface response, identified the optimal setting for well shut-in/open conditions, and continuously updated the surface response. Overall, the analytic improved production by 4% and reduced the liquid loading occurrences and manual well unloading events by 94%, resulting in an average reduction of approximately 600 tons of CO 2 equivalent per well per year. In summary the active learning analytic was developed and deployed on an edge computing platform to 1) optimize intermittent shut-in by searching for the optimum settings that yield the most gas production; 2) automate the optimization process; and 3) monitor the liquid formation for potential loading events. In this paper, we present a use case for an algorithm adapted for the optimization of a dynamic system such as hydrocarbon production from a well.

Fiber-optic transmission has been applied in oil and gas industry over the years. Compared with other methods applied in the industry, fiber-optic transmission has the advantages of low loss, long-distance, high-capacity and robust to the electromagnetic interference. The ability to provide reliable transmission systems in the harsh environments like high temperatures is the key driver for the continued use of fiber-optic communication for in-well applications. We design a fiber-optic communication system under high temperatures for well logging applications. It consists of high-temperature laser diode, high-temperature photodetector with photoelectric detection circuit, drive control circuit, and field-programmable gate array (FPGA) as the communication chip. This system ensures that data can be transmitted at a rate of 15 Mbps at temperatures up to 155°C. The FPGA board makes the system to control data transmission flexibly and enable the serial communication between the photoelectric module and the host computer. Additionally, the number of fibers used in fiber-optic communication in logging will be reduced to only a single fiber for transmitting and receiving. A series of experiments on the performance and effects of fiber-optic communication at different temperatures was carried out. Data transceiver tests and eye diagram tests are presented. The experimental results demonstrated that this fiber-optic communication system is capable of working steadily over a long period of time in harsh environments around 155°C to realize broadband and remote transmission of logging information. This system provides a way that allows optical information to transmit in a high-temperature environment. It can be applied to well logging and fiber-optic sensing (e.g., real-time environmental parameters transmission, fiber-optic well monitoring) for developing real-time, high-data-rate, bidirectional fiber-optic communication in the future.

To contribute to the construction of young professionals with an integral profile in which the competitive spirit is promoted through the resolution of a technological challenge applied to the Oil & Gas industry, under a scenario that demands a high degree of commitment and with the accompaniment of professionals with great experience in the industry. In accordance with the data acquired and the program's target population, an appropriate methodology has been established for the development and strengthening of technical and soft skills. This methodology consists of four main phases: 1. Challenge construction 2. Best Teams selection 3. Challenge Resolution and 4. Results presentation. During the challenge resolution phase, participants have multiple technical and soft skills training sessions where experienced professionals share their knowledge and experiences related to the challenge theme in an environment of generational knowledge transfer. During the last 4 years, around 120 participants from different cities of the country have managed to potentiate their soft skills and strengthen their knowledge in areas of the Oil & Gas industry such as unconventional, enhanced recovery, heavy oil, huff & puff, combustion had tests, wettability, waterflooding, among others and materialize it with the construction of an innovative technological tool. The best of each version was rewarded with their first work experience in the Oil & Gas industry where they were able to put into practice what they had learned during the program. Likewise, the continuous acquisition of valuable information about variables at the end of the program such as the knowledge areas of greatest interest for students and young professionals, the level of knowledge related to oil engineering and programming, the handling of a second language, and the skills acquired during the university training stage, has allowed strengthening the methodology of the"SPE en Sinergia" program, adjusting it to the current needs of the industry and placing the participants outside their comfort zone. The development of the program and the results obtained are based on the identification of the strengths and weaknesses currently displayed by students in their final semesters and young professionals in careers related to petroleum engineering in Colombia and that in one way or another have become part of the reasons why their link to the professional world has become more difficult. This has enabled a way to identify opportunities for improvement and establish lines of action to promote the professional fast-track development of the new generations in a joint effort between industry and academia.

In an era of reduced profit margin and high market uncertainty, more than ever it is important to meet operational excellence as a key factor for business sustainability. This is common to most technical applications, but it is particularly true for the drilling operations, where considerable investments and associated risks are involved. During last four years, as part of its digital transformation process, Eni has equipped itself with several digital tools for the diagnosis and the monitoring of drilling and completion operations. Goals and reached benefits can be summarized in risk reduction, operational efficiency and performance optimization. Based on a wide case history started in 2019, a Digital Drilling Package was developed for operations support, from the design to the construction phase. Three main tools are now available to be applied to the most complex wells, either stand-alone or in parallel, covering drilling operations non-productive time (NPT) prediction, performance advanced analytics and real time simulations. This last simulation tool was deployed for the first time in late 2020 on some wells and is now being included in the engineering and operation workflows. Attacking operational NPT and invisible lost time with the aim to increase safety and to reach the technical limit is not only a matter of processing tools. It requires a deep integration with headquarter (HQ), geographical units and field locations, with the definition of a strong data management infrastructure. This paper describes Eni's experience both on-site and in office, showing how the portability and integration of big data systems, suitable data lake architectures and human factor synergies can create effectiveness at all levels. An Africa Offshore field case history is reported to show how predictive and data analytics modelling and tools interact. In addition, the way in which these tools have been managed to support optimum decision-making processes is highlighted. Next development steps will target an even higher level of integration of all available digital tools to have a single diagnostic approach based on univocal dashboards and in-house data server infrastructures.