Multi-level Failure, Causality and Hazard Insights via Knowledge Based Systems
Abstract
Over many decades there has been a significant development of knowledge-based, intelligent design tools and their use in the design of process systems. Amongst such tools are “intelligent” piping and instrumentation (P&IDs) design environments, coupled to life cycle design environments. These tools can provide opportunities for the development of new, more efficient and re-usable approaches to hazard identification and diagnostic systems. They leverage modern information technology characteristics of such design environments. These considerations are part of a growing trend in industrial digitalization, as reflected in such initiatives as Industry 4.0 in Europe and driven by the Industrial Internet of Things (IIoT). Within this larger industrial digitalization picture, this work discusses the principles, developments and application of a hazard identification methodology (BLHAZID) that exploits structured representations of the design in the form of ISO15926 data standards. The hazard identification methodology is based in knowledge representations of failure modes of equipment types that are found in many process designs and how those failures subsequently affect the system states and other components. The underlying causal models can be used at various levels of aggregation, model fidelity and component inclusion detail. The aggregation can span across the most detailed view at the smallest component level through subsystem level to plant level perspectives. The ability to represent and then display failure causation and implications at different levels of granularity allows deeper insight into system failures, and the potential for real-time diagnostic deployment. The importance of failure and subsequent propagation prevention through the use of safety instrumented systems and other barrier devices is possible. Outcomes can be visualized in informative ways. The presentation will discuss these intelligent information technology approaches via some a case study, highlighting the advantages and challenges such approaches bring to hazard identification as well as highlighting other application areas such as real-time diagnosis, corporate knowledge capture of failures, operator training and accident investigation.
Description
PresentationSubject
Knowledge Based SystemsCollections
Citation
Németh, Erzsébet; Cameron, Ian (2018). Multi-level Failure, Causality and Hazard Insights via Knowledge Based Systems. Mary Kay O'Connor Process Safety Center; Texas &M University. Libraries. Available electronically from https : / /hdl .handle .net /1969 .1 /193435.