Safety Instrumented System Design Philosophy Paradigm Shift to Achieve Safe Operations of Interconnected Operating Sites

Abstract

Petrochemical industry and other process manufacturing facilities require major infrastructure investment with dangerous operations that are susceptible to great risks such fire, explosion, and or un-orchestrated process upsets. The Safety Instrumented System (SIS) ensures process operation is managed by a Basic Process Control System (BPCS). SIS is deployed in the local process automation zone of the processing facility without interconnection to a remote monitoring and operations facility. This resembles one of the major key challenge and limitations of SIS systems. There are historical major incidents in the Oil and Gas industry that could have been avoided if the SIS system performance is proactively known and acted autonomously upon abnormal conditions. This paper examines a set of major Petrochemical industry process related incidents with primary focus on identifying sensor network and system weaknesses. The detailed analysis of earlier incidents revealed the need for SIS design enhancements in the networking, system architecture, data flow interworking model, and unexistence of centralized data processing with execute and autonomous decision-making rights. The use of different communication mediums fiber, wireless, and VSAT introduces new capabilities that can be utilized to achieve required data delivery for process safety related actions. Moreover, the emerging Industrial Internet of Things (IIoT) Technologies introduce a new automation layer that increase proactive decision making. This paper introduces a new concept in inter-linking process operations that have multiple, distributed, and remote operational zones. The paper concluded with best practices that enhance the current design model, efficiency and operational reliability. The intent is to bridge an existing gap that well identified and prevent escalation of hazardous events with reduced time response (miliseconds to seconds) as compared with current design philosophy that relies in non-autonomous decision making processes (human intervention) that take longer times (minutes to hours) to detect and react after the fact, lagging indication and lacking the proactive approach.