Day 1 Tue, January 17, 2023最新文献

{"title":"Environomics Framework for Sustainable Business Practices: Industrial Case Studies on True Impact Reduction and Process Optimization Through AI","authors":"A. Suboyin, M. Eldred, J. Thatcher, Abdul Rehman, Ivan Gee, Hassaan Anjum","doi":"10.2118/214459-ms","DOIUrl":"https://doi.org/10.2118/214459-ms","url":null,"abstract":"\u0000 Artificial Intelligence (AI) has significant potential to optimize practices, processes, and energy consumption along with maximizing yield, quality, and uptime. This has substantial impact on putting organizations on the path to net-zero, as such optimizations can reduce greenhouse gas emissions by 20% with minimal capital investments. This comprehensive study presents proven industrial case studies that delivered economically strong strategies coupled with sustainability practice and providing strategic insights to identify, manage and/or attenuate the associated impacts.\u0000 Environomics presented in this study is a novel framework which deals with unifying economic strategies with sustainability practices (through artificial intelligence) for optimal business performance in terms of finances but also environmental impact. This is achieved through a track, trace, and optimize approach for resources (particularly emissions, energy, water, waste, materials,, and safety)\u0000 This was achieved through a combination of AI methods such as unsupervised machine learning, multi-variate optimization, and the implementation of similarity measures. A few of the inputs included well data (including production data, drilling data, completion data etc.), logistics/supply chain data (scheduling data, production inventory, mobilization data etc.), safety data (near-miss, observations, hazards, disciplines and insights etc.) with associated costs and emission data.\u0000 Multiple industrial case studies are presented where sustainability metrics are identified through validated AI models to optimize productivity while reducing emissions and inventory. For instance, well profiling can be used to identify historical parameters that have maximized production potential while optimizing for aspects such as cost or emissions. Furthermore, we can identify the optimal completion parameters for a new well which satisfies carbon targets, use well profiles to build an optimized drilling schedule that meets budget or production criteria while still achieving production targets and optimizing drilling rig routes. Thus, the approach can quickly (within run time) solve interrelated environomic challenges in the reservoir studies space and the field development space.\u0000 Further case studies indicate that the supply chain can have immense optimization impact on scope 3 aspects with results indicating 30-50% asset utilization improvement with respect to fleets (Vessel, Truck, Rigs). With respect to materials, a 10-20% reduction of material inventory levels all improved through AI. As the workforce are also part of the environment it has been observed that identifying unsafe behaviors within a large operation, also leads to enhanced sustainability behaviors. The models indicate potential of overall emission reduction ranging from 12-20%. This led to the comprehensive framework presented in this study to support sustainable practices that are also economically feasible and deployable. The r","PeriodicalId":393098,"journal":{"name":"Day 1 Tue, January 17, 2023","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115325837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leveraging AI for Inventory Management and Accurate Forecast – An Industrial Field Study","authors":"M. Eldred, J. Thatcher, Abdul Rehman, Ivan Gee, A. Suboyin","doi":"10.2118/214457-ms","DOIUrl":"https://doi.org/10.2118/214457-ms","url":null,"abstract":"\u0000 Accurately forecasting demand is one of the most undervalued and complex strategies that can significantly impact organizations bottom line. This industrial field study was co-conducted with Sumitomo Corporation's Tubular Division which primarily deals with high-grade Oil Country Tubular Goods (OCTG) globally. The presented solution demonstrates how with the right data set (drilling sequence data, stock data and consumption data), artificial intelligence can be used to build out a model that can quantify and predict future demand accurately thereby reducing cost, working capital and emissions.\u0000 Multiple multi-layered machine learning models were built to compare and analyze a wide variety of data inputs for bill of materials, operational/project schedules; This includes (a) ‘product movement data’ which describes the changes in demand and supply of a product, (b) ‘product specification data’ which describes the characteristics of a product, and (c) ‘activity specification data’ which describes the characteristics of an activity. The models follow the base temporal map design with different weighting on model inputs. With a temporal map, a sequence of monthly data values (called lags) is used to predict the next monthly value in the sequence. The lags are rolled so that there are six months of data for the model to predict on. All models also use boosted decision-tree-based ensemble machine learning algorithm.\u0000 It is critical to understand how product movement metrics (actual and safety stock levels, historical forecasts, and consumption patterns), product specification data (lead time, product grade, well function, well category, work center), and external factors (oil price, rig counts, national budget, production targets) can be utilized together to better understand future product demand. Using historical data acquired from drilling operations and supply chain over an eight-year period, multiple machine learning models were trained to predict one year of demand across the most consumed products. Across five years of predictions (2016 to 2019), the models were able to predict with 78% average accuracy for the top 10 products by volume which represents 75% of inventory volume. Across the same time-period, they were able to predict with 73% average accuracy on all 17 products which account for 80% percent of inventory volume. Further iterative updates with additional data led to improvement in results and the model where the model predicted with an improved accuracy of 83% on the top 17 products and an accuracy of 86% on the top 10 products.\u0000 Moreover, the data can also be used to generate dashboards featuring metrics on material uncertainty / velocity and expected differences between the internally predicted forecasts and actual sales. The results further indicate that, on average, and within a simulated environment (where shipping delays were not considered for instance,) the AI model can maintain a lower inventory than the originally planned","PeriodicalId":393098,"journal":{"name":"Day 1 Tue, January 17, 2023","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130112169","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Into the Unknown: Expert System Guides Energy Transition Strategy","authors":"P. Allan","doi":"10.2118/214458-ms","DOIUrl":"https://doi.org/10.2118/214458-ms","url":null,"abstract":"\u0000 Most E&P companies have publicly stated some form of ‘carbon reduction’ planning as they communicate their strategies to stakeholders. Internal efforts for reducing carbon generation might include reduced flaring, pipeline integrity improvements, or carbon sequestration for which E&P companies have the experience and skill sets to adequately evaluate. Other companies have committed to more extensive and fundamental changes to their business models – potentially necessitating a need to expand into historically ‘non-E&P’ energy sectors, such as wind, solar, or hydrogen businesses. The expertise to explore these types of strategic decisions can potentially be acquired through hiring or acquisitions but is often insufficient from within the ranks of typical E&P firms. This can make the initial exploration into these ‘possible’ alternatives risky and / or inadequately informed. As an aid to companies entering the renewables space, the following paper describes a portfolio modelling approach to assessing clean energy business alternatives. Renewable energy characteristics, including investment profiles, cost structures, and location specific efficiencies and returns (economics) are incorporated into a portfolio model as based on ‘expert guidance’ and publicly available data sets. This model makes it possible to capture the characteristics of the existing hydrocarbon business (production, cash flows, capital investments, etc.) and layer in ‘possible’ alternatives for wind, solar, or carbon offset investment alternatives. This modelling allows decision makers to begin exploring possible investments in these sectors without the requirements for large investments in new personnel, acquisitions, or other costly steps.\u0000 A simple portfolio model representing a conventional E&P organization has been developed and expanded to include possible sampling of renewable energy projects. This model provides a means of selecting from various investment options (either manually or utilizing a linear optimization routine) and assessing the performance characteristics across multiple metrics. The model includes operational and economic descriptions of renewable energy investment alternatives, including investments in onshore wind, offshore wind, solar photovoltaics, concentrated solar, and carbon offset and sequestration projects. The key drivers and assumptions for these investment alternatives are based on current industry trends and cost structures and are clearly noted and open for revision or customization to a company's specific location or existing business knowledge as needed.\u0000 This paper will demonstrate how these techniques can assist in positioning company decision makers for more informed entry or exploration of new business options as these opportunities evolve. The methods combine proven techniques in portfolio assessment (utilizing linear optimization and Monte Carlo simulation) with ‘expert’ guidance as to the characteristics of clean energy businesses. ","PeriodicalId":393098,"journal":{"name":"Day 1 Tue, January 17, 2023","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129973583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Decision Tree Ensembles for Automatic Identification of Lithology","authors":"M. Desouky, A. Alqubalee, Ahmed Gowida","doi":"10.2118/214460-ms","DOIUrl":"https://doi.org/10.2118/214460-ms","url":null,"abstract":"\u0000 Lithology types identification is one of the processes geoscientists rely on to understand the subsurface formations and better evaluate the quality of reservoirs and aquifers. However, direct lithological identification processes usually require more effort and time. Therefore, researchers developed several machine learning models based on well-logging data to avoid challenges associated with direct lithological identification and increase identification accuracy. Nevertheless, high uncertainty and low accuracy are commonly encountered issues due to the heterogeneous nature of lithology types. This work aims to employ decision tree ensemble techniques to predict the lithologies more accurately in time saving and cost-efficient manner, accounting for the uncertainty.\u0000 This study investigated the real-world well logs dataset from the public Athabasca Oil Sands Database to identify and extract the relevant features. Then, we conducted a thorough training using grid search to optimize the hyperparameters of the ensemble decision tree models. This paper evaluated two ensemble techniques: random forest (RF) and extreme gradient boosting (XGB). We picked metrics such as accuracy, precision, and recall to assess the developed models' performance using 5-fold cross-validation. Finally, we performed a chi-squared test to test our hypothesis of the identical performance of the developed models.\u0000 The XGB and RF models have 94% and 93% accuracy, respectively. Also, the extreme gradient boost model's weighted average recall and precision of 93% and 93% are only 5% and 4% higher than the RF model. In addition, the chi-squared test resulted in a p-value as low as 0.013, suggesting a low probability of difference in both models' performance. Classification of sand and coal formations is more straightforward than sandy shale and cemented sand. The dataset's low representation of sandy shale and cemented sand can be the reason behind their prediction errors. The developed models can classify the studied field lithologies with an overall accuracy of 94%. In addition, there is no statistically significant evidence of a difference in prediction performance between extreme gradient boost and random forest.","PeriodicalId":393098,"journal":{"name":"Day 1 Tue, January 17, 2023","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133573887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}