Automated control and safety systems are prevalent in modern chemical plants, as they help plants return to normal operating conditions when abnormal events occur. The data associated with these systems contain a wealth of information about near-miss occurrences. Frequent statistical analyses of the information can help identify problems and prevent accidents and expensive shutdowns. Such analyses are referred to as dynamic risk analyses.
Predictive maintenance is an important evolution in the effective and efficient management of industrial chemical processes. Often predictive maintenance is focused upon individual equipment within a process (e.g., compressors). Lately, in addition, proactive risk management is recognized as a key concept for the evolution to the next level of safety, operability, and reliability performance in chemical operations involving plant-wide analyses toward prescriptive maintenance. Dynamic risk analysis refers to a methodology that utilizes various tools to collectively help to achieve proactive risk management. Hence, its application has been gaining significant importance in the chemical and process industries.
This chapter discusses dynamic risk analysis of alarm data. It provides a general overview of what these analyses are, how they can be used in chemical processing to improve safety, and challenges that must be addressed over the next 5–10 years. It also highlights current research in this area and offers perspectives on methodologies most likely to succeed.
Keywords: risk, safety, bayesian analysis, process dynamics, alarms, data compaction, steam-methane reforming