WingBlog

Up until the mid to late 1970s, alarms were an important tool used to enhance a process operator’s situation awareness.  The steps to provide the operator with a single alarm included:

1. Purchase and add a sensing switch to detect an alarm condition.
2. Run conduit and cable back to the control room.
3. Purchase an enunciator module to add to the control panel annunciation system.
4. Connect all of the components.

This was not a task to be taken lightly and you only did it for alarms that really meant something.  The physical space for adding the alarms was also very limited.  As a result, each new alarm was evaluated and justified.

With the advent of the computer based control systems with layered applications, every tag in the system is a software construct.  Alarms are now free.  Also, every tag in the system has the capability of twelve to sixteen alarms To compound the issue, the young engineer, working for the engineering consultant or controls manufacturer doing your current controls upgrade, wants to make sure you get “your money’s worth” and will turn them all on.

As a result, the number of alarms per operator has gone up exponentially as shown in the graph.  Now, with the addition of fieldbus technology, we stand on the brink of an accelerating problem. Is it important? In the early 1990s, several industrial accidents occurred with alarm systems being identified as contributory factors.  One example is the Health and Safety Executive (HSE) report describing the Milford Haven Refinery explosion.  The report identifies: Twenty-six people were injured and £48million was lost. There were too many alarms and they were poorly prioritized. The alarm system did not effectively help the operators to understand the situation. The operators had to deal with 275 alarms in the last eleven minutes before the explosion occurred. The Abnormal Situation Management (ASM) Consortium was created and started studying the problem in 1994 and in 1999 the Engineering Equipment and Materials Users Association (EEMUA) issued its Publication 191 on the subject.

The following standards have been issued by two prominent professional societies: the American Petroleum Institute (API RP1168) and the Instrument Society of America (ISA 18.2).

In the United States, a frequently overlooked fact is that OSHA (the US Occupational Safety and Health Administration) specifically outlines it requirements for alarm systems in CFR 29, regulation 1910.119, Process Safety Management of Highly Hazardous Chemicals.  These regulations outline the requirements for accurate documentation of critical alarms and the operator training needed to understand and respond to the alarms.

However, for many companies, the question of importance boils down to how we justify it.  The ASM  has estimated that poor alarm management is one of the leading causes of unplanned downtime, contributing to over $20B in lost production every year in the United States alone. In worst case scenarios, when things go really bad, we have a major industrial incident such as the one at BP’s Texas City refinery resulting in the deaths of 15 and the injury of over 170 others.

How do we address this issue?

Almost all facilities recognize that significant alarm management improvements are needed.  They are aware that the issue needs to be thoroughly investigated and corrected.  This issue is becoming a constant topic of articles in trade journals and at technical society meetings.

Over the next couple of weeks, I will be addressing this issue by discussing the following seven steps to successful alarm management.

1. Developing and adopting an Alarm Philosophy
2. Collecting data and benchmarking your system
3. Identifying and eliminating “Bad Actor” alarms
4. Rationalizing and documenting your system
5. Implementing automated alarm auditing and Management of Change (MOC)
6. Implementing real time alarm management
7. Maintaining your system as your process changes

The approach that I will be outlining has been successfully used, modified and refined over the last several years.

P.J. Plauger said, “My definition of an expert in any field is a person who knows enough about what’s really going on to be scared.”  After a decade of working with alarm systems in different industries, I have seen enough to be scared.