Application of machine learning to assess the value of information in polymer flooding

In this work, we provide a more consistent alternative for performing value of information (VOI) analyses to address sequential decision problems in reservoir management and generate insights on the process decision-making. These are often solved modeled as stochastic dynamic programs, but once state space becomes large complex, traditional techniques, such policy iteration backward induction, quickly become computationally demanding intractable. To resolve these issues utilize fewer computational resources, instead make use viable called approximate programming (ADP), which is powerful solution technique that can handle large-scale discover near-optimal intractable making. We compare test performance several machine learning techniques lie within domain ADP determine optimal time beginning polymer flooding development plan. The approach utilized here takes into account both effect obtained before made might be support future decisions while significantly improving timing decision, thereby leading significant increase economic performance.