Propose an alternative method to generate alarm management KPI utilizing IMS in offshore platforms

Abstract

Weaknesses regarding alarm management practices have been recognized as partial causes of plant incidents with tragic results, for example the Texas City Oil Refinery explosion in 2005. Necessity in alarm regulation was recognized by industry bodies and norms were created (EEMUA 119, ANSI/ISA 18.2) in order to address this. These norms define a complete alarm management lifecycle, including the definition of an alarm performance monitoring process with the supervision of alarm’s key performance indicators (KPI) Alarm KPI generation in oil instalations is a complex technological task and generally includes the integration of alarms coming from different sources in the process plant. The Information Management Systems (IMS) as defined in NORSOK I-002 as the natural gathering engine for process data in an offshore installation. The utilization of IMS in Alarm Management Purposes is included in this NORSOK standard. The original driver of this investigation was that Offshore Platforms in the Norwegian Continental Shelf do not utilize IMS for Alarm Management KPI generation, despite their intense utilization of IMS for other purposes. Technical reasons for this practice are not known. In order to evaluate the current capacity of IMS in alarm KPI generation, a minimum required set of KPIs was established although different standards show variations between KPIs. The data structure gathered inside IMS in related to alarms (alarms metadata) was compared against the established set of KPIs. The main study finding conclude that data contained whiting the IMS structure is currently not enough for generate the priority- related KPIs, mainly because alarm priority is not included in current IMS. In general terms; IMS data structure is not defined by NORSOK and even when included in some industry related regulations (as Statoil TR2258) Alarm Priority is not considered to be part of Alarm metadata. The above finding points to an easy solution and raises questions regarding lack of “data priority” in the IMS data structure. This comes from a lack of regulation in IMS Implementation, mainly because of NORSOK standards.