Safety analysis using an expert system in chemical processes

Abstract

A knowledge-based expert system for hazard and operability study (HAZOP) is developed. HAZOP study is regarded as one of the most systematic and logical qualitative hazard identification methodologies. But, it requires a multidisciplinary team and is very time-consuming and repetitious task in nature. By developing an computer-aided automation system, these drawbacks of HAZOP study can be overcome. Considerable manpower and time can be reduced and even past experiences of engineers and existing checklists can be stored for future use in the form of knowledge base. The developed knowledge-based HAZOP expert system has a frame-based knowledge structure for equipment failures and process properties, and rule networks for consequence reasoning which uses both forward and backward chaining. The system is open-ended and modular in structure to make it easy to implement wide process knowledge for future expansion. LPG storage and fractionation process has taken as example to test the applicability of the developed system as an automated HAZOP study system. The result shows that savings more than 50% of the required manpower and time for HAZOP studies can be achieved, and the system is very efficient and reliable, too.