Using Data Analytics to Improve Process Safety
Increasingly, data from multiple sources are compiled into central storage systems for entire facilities and in some cases entire corporations. The low cost of digital memory and the increasing process speed of modern microprocessors and computers continue to make it easier to analyze, manipulate and merge very large databases. Big Data and Data Analytics have been touted as allowing staff at all levels to make real time, data-driven business decisions. These concepts can be used to improve process safety at facilities by proactively analyzing sources of data to identify opportunities for improvement in operating practices, identify and track near misses, and in real time detect incipient instrument and equipment failures that could lead to catastrophic incidents. The aviation industry is one example that is voluntarily implementing similar analytics. Over the last decade the aviation industry has implemented flight operations quality assurance (FOQA) or flight data monitoring (FDM) programs which collect, store, and analyze recorded flight data with the goal of increasing overall safety.
In this presentation we show examples of analytics Exponent performed on data from facilities. These analyses were used to analyze alarm frequencies during specific types of operations, identify near misses, and proactively detect failures of instrumentation and equipment in order to prevent or mitigate catastrophic incidents.