@inproceedings{scholars17354,
            note = {cited By 2; Conference of 2022 Offshore Technology Conference Asia, OTCA 2022 ; Conference Date: 22 March 2022 Through 25 March 2022; Conference Code:187172},
            year = {2022},
             doi = {10.4043/31438-MS},
         journal = {Offshore Technology Conference Asia, OTCA 2022},
       publisher = {Offshore Technology Conference},
           title = {Well Pair Based Communication Strength Analysis for Water Injection Reservoir Surveillance Using Monte Carlo Simulation Coupled with Machine Learning Approach},
            isbn = {9781613998359},
          author = {Pratama, E. and Ridha, S. and Negash, B. M.},
        abstract = {With the increasing of water injection activities especially for marginal or stranded fields, the well pair analysis in routine water injection surveillance is crucial to understand the reservoir performance and identify opportunities to improve the ultimate oil recovery. This article aims to propose an alternative technique to evaluate the communication strength between injector - producer well pairs based on statistical and machine learning algorithms. The proposed technique is applied to an offshore water injection field located in the North Sea from open-source data. A novel formulation to quantify the communication strength coefficient for an injector - producer well pair was derived from the Spearman's rank correlation coefficient. The calculation is controlled with injection/production rates pattern for each well pair. Subsequently, multivariate parametric regression is performed to model the communication strength coefficient as a function of injector - producer spacing, injection pattern (dip angle), and reservoir permeability-thickness. Monte Carlo technique is then applied to simulate 100 cases prepared using the uniform probability distribution. Afterward, the communication strength for all the well pairs in the field is classified based on K-means clustering. To identify opportunities to improve the effectiveness of water injection operation, random forest and support vector machine algorithms are used to evaluate the effect of the reservoir and operational parameters on the communication strength of the injector - producer well pair. It is identified that the communication strength for all the well pairs in the field varying from limited, intermediate, and good communication. Good communication strength shows the correlation coefficient of more than 0.50 which indicates there is a good correlation between injection and production rates pattern. It is also observed that reservoir permeability-thickness is the most variable importance that affects the communication strength between injector and producer well pair. It is followed by the injector-producer spacing and reservoir dip angle. The optimum condition has been identified to formulate the screening criteria in order to obtain the good communication strength between injector and producer well pair. This result help in identifying the producer with limited communication strength with the existing injector and low production rate to be converted as the injector well. Unlike reservoir simulation which is a very expensive and time-consuming process, this work provides a quick and inexpensive alternative to evaluate the communication strength of injector-producer well pair from widely available measurements of production and injection rates at existing wells. Application of this novel workflow provides insight for better decision-making and can be a prudent complementary tool to quantify the effectiveness of the water injection operation and identify opportunities. Copyright {\^A}{\copyright} 2022, Offshore Technology Conference.},
             url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85143075232&doi=10.4043\%2f31438-MS&partnerID=40&md5=00feab64e01d4935a50d848c69677d55},
        keywords = {Decision making; Intelligent systems; K-means clustering; Learning algorithms; Monte Carlo methods; Offshore oil wells; Offshore technology; Petroleum reservoir engineering; Probability distributions; Support vector machines, Dip angle; Injection rates; Injector-producer spacings; Monte Carlo's simulation; Permeability-thickness; Production rates; Reservoir permeability; Reservoir surveillance; Strength analysis; Strength coefficients, Offshore oil well production}
}