A New System Function for Maximum Processable Flow in Process Plants and Application to Reliability Assessment

Abstract

This study presents a new system performance function for process plant reliability analysis, formulated to capture both structural topology and process sequencing constraints. Built on a modified maximum-flow framework and solved via linear programming, the proposed function efficiently quantifies the maximum feasible flow through a series of interconnected stages. It addresses limitations of existing models, such as fault trees and event trees, that often overlook flow continuity and topological dependencies. The system function is integrated into a reliability assessment framework, enabling the evaluation of system failure probability and reliability-based component importance measures. Application to two benchmark examples, including a gas supply plant with 57 nodes and 102 edges, demonstrates the effectiveness of the proposed system function and the validity of the resulting reliability assessment for risk-informed layout planning. A reconfiguration guided by component importance measures yields up to a 20% reduction in system failure probability, underscoring the importance of effective equipment and pipeline layout. The proposed framework offers a promising direction for reliability-based management of industrial process facilities.