Efficient Reservoir Management: A Data-Driven Decision-Making Approach Using Integrated Streamline-Based Model

Abstract Understanding flow patterns between injectors and producers is key to better reservoir management. Streamline tracing serve as a powerful tool to map these patterns through quantitative analysis that estimates various parameters such as well efficiency and allocation factors for both injectors and producers. However, the true value lies in translating this data into actionable insights that aid in refining injection strategies, reducing water production, enhancing sweep efficiency and ultimately improving oil recovery. This paper introduces a comprehensive approach that integrates streamline tracing results with external databases containing a broad range of historical and up-to-date field data, including well completions, performance metrics, production trends, logging records, and operational diagnostics, to manage large reservoirs efficiently and effectively. Coupled with defined management strategies, this approach can be implemented through a versatile, user-friendly platform featuring robust visualization tools that provide an automatic field management plan for a large number of injecting and producing wells. The approach comprises two workflows with two main objectives: (1) optimizing water injection strategies and (2) enhancing oil production. The first workflow of the created model leveraged streamline tracing results by reassessing the injection efficiency, through weighted criteria that accounted for simulation flow rates, calculated percentages of water leakage and recycling, injection and production targets, as well as well potentials. Based on the results, the model instantly generated a ranked list of candidates to achieve various management goals: minimizing injection leakage, reducing water recycling and enhancing oil sweep efficiency. Accordingly, an action plan was established, such as assessing injector candidates that leak most of their injected water and producer candidates that recycle all received injection water for potential sidetracking. The second workflow involved a sequential analysis of field data, underpinned by reservoir management practices, to develop a well-specific action plan. Each well was categorized based on performance and guided through a strategically defined pathway to create the action plan. The model analyzed the data in a defined sequence and recommended an optimal plan for each well, including workover activities (i.e., sidetracking or running liners), temperate or annular surveys to confirm well integrity, acid stimulation, water shut-off, or ICD/ICV shifting, nitrogen kick-off; revised injection plans based on identified candidates; and choke size adjustments. This data-driven decision-making approach converts dynamic data into immediate, actionable insights that aid in managing large reservoirs, with thousands of injectors and producers. It can be implemented using simple software, such as Power BI, significantly accelerating traditional, time-intensive reservoir management tasks and bridging the gap between analyzing vast amounts of data and taking actions.