Process Resilience Analysis Framework for Design and Operations

Process plants are complex socio-technical systems that degrade gradually and change with advancing technology. This research deals with exploring and answering questions related to the uncertainties involved in the process systems, and their complexity. It aims to systematically integrate resilience in process design and operations through three different phases of prediction, survival, and recovery using a novel framework called Process Resilience Analysis Framework (PRAF). The analysis relies on simulation, data-driven models and optimization approach employing the resilience metrics developed in this research. In particular, an integrated method incorporating aspects of process operations, equipment maintenance, and process safety is developed for the following three phases: •Prediction: to find the feasible operating region under changing conditions using Bayesian approach, global sensitivity analysis, and robust simulation methods, •Survival: to determine optimal operations and maintenance strategies using simulation, Bayesian regression analysis, and optimization, and •Recovery: to develop a strategy for emergency barriers in abnormal situations using dynamic simulation, Bayesian analysis, and optimization. Examples of a batch reactor, and cooling tower operations process unit are used to illustrate the application of PRAF. The results demonstrate that PRAF is successful in capturing the interactions between the process operability characteristics, maintenance, and safety policy. The prediction phase analysis leads to good dynamic response and stability of operations. The survival phase helps in the reduction of unplanned shutdown and downtime. The recovery phase results in in reduced severity of consequences, and response time and overall enhanced recovery. Overall, PRAF achieves flexibility, controllability and reliability of the system, supports more informed decision-making and profitable process systems