Improvement in Process Safety Performance from the Monitoring and Evaluation of an Alarm Management System

The objective of this work is to present how from the monitoring and evaluation of an alarm management system, implemented from what is defined in the ISA 18.2 - 2016 standard, it has been possible to obtain important information to support proactive decision making that have allowed the improvement in the process safety performance for our company's industrial facilities.

The definition of the alarm management system is found within the company standard named "Guide for the management of process alarm systems" which was developed by professionals specialized in process safety, functional safety, instrumentation and control, and process engineering. This document was based on the ISA 18.2 – 2016 standard and applies to upstream, midstream and downstream facilities.

The monitoring and evaluation process of the alarm management system implemented in our company has the objective of allowing the verification that the design, rationalization, implementation, operation and maintenance of the alarms are adequate and allow to avoid the materialization of process safety events.

Also, it defines the evaluations that must be made to the management system in order to identify if corrective actions and/or advanced alarm techniques are required to achieve the performance objectives.

For this, teams have been created made up of personnel with knowledge in alarm management together with operations professionals. In the work sessions, the historical reports of the alarms that are generated in the period of time established specifically for the installation are reviewed. The evaluation has allowed to determine the alarms that require optimization and thus, it has been possible to reduce the over flow of alarms that are presented at certain times to the operator. This optimization has contemplated the possible elimination or inhibition of the alarm depending on the operating mode of the plant or equipment, or its conversion to an event. As a general rule, those alarms that are not real (that do not require an action by the operator) must be reported as events.

Other objectives that have been met from the evaluation and monitoring are to maintain the integrity of the alarm system and its management processes, based on the identification of needs for system improvement and the adjustment of the alarm philosophy or its definitions.

Finally, and as the main objective of this document, the monitoring and evaluations have made it possible to anticipate the occurrence of serious process safety incidents, as a result of the analysis of the defined metrics.

The metrics have been defined within two (2) categories:

• Static metrics: based on the configuration of the control system, associated with the operator (panel operator/console operator).

For example: Distribution of configured alarm priorities.

• Dynamic metrics: calculated from the database in real time of operation, where alarm activations, acknowledgments and other operator actions are captured.

For example: Average alarm rate and Frequent alarms (top 10).

According to this definition, the metrics that have been used the most for proactive decision-making focused on preventing the occurrence of process safety events are:

1. Distribution of configured alarm priorities: The metric is determined from the list of all alarms configured in the system excluding Fire & Gas class alarms. This metric determines how many alarms are critical, high, medium and low within the prioritization of the system, in order to bring them to what is suggested in the company's standard (based on ISA 18.2 version 2016).
2. Average Alarm Rate: Displays the average number of alarms that are announced each hour on the operator's control system.
3. Frequent alarms (Top 10): This metric reviews the most frequent announcements of alarms in the system, that is:

Inventory of recurrent critical priority alarms in the period.

Inventories of high, medium and low priority alarms in the period.

Inventory of recurring Fire&Gas alarms in the period.

These monitoring is a key piece of information for the operational analysis and alarm management meetings in order to see trends in real time and define action plans to eliminate the causes that led to the recurrent activation of these layers of protection.

During the presentation, we expect to show in detail real cases where monitoring and evaluation made it possible to show shortcomings in the process safety management of the installation before an incident occurred. Evidence of this from the analysis of the metrics generated actions such as changes in the technology of the process, adjustments in the maintenance plans, optimization of the structured rounds or changes in procedures and practices.

Thanks to this practice, it has been possible to act proactively and generate a direct impact on the safety of our processes.