Alarm Management  

Our clients appreciate the fact that we are involved in projects across a wide spectrum of industries because
we bring insights from common challenges experienced by other industries that lead to innovative solutions
to their problems.  A good example is alarm management which is a consideration in the design of almost
every kind of critical system.  Although the details of alarm management may vary considerably between
technical domains, our approach to helping clients with alarm management is based on the same fundamental
concepts and principles.

Thanks to Hollywood and movies such as The China Syndrome (1979), most of us have a sense of the
adrenaline fueled drama of a control room during an emergency with alarm bells ringing and lights flashing.  
But helping a client develop a sound approach to alarm management goes well beyond thinking about the
rare moments of high drama.

For example, an operator could eventually become desensitized to a spurious alarm that is repeatedly triggered
by a faulty sensor and miss the significance of the alarm on the one day when there really is a need for urgent
action.  To reduce the distraction of nuisance alarms, operators often have a means of suppressing an alarm.  
However, an ad hoc approach to alarm suppression can have dire consequences – as illustrated by the
1997 crash of a Korean Air 747 jetliner in which one of the Minimum Safe Altitude Warning (MSAW) alarms of the
ground-based equipment used by air traffic controllers had been modified in such a way that it would no longer
alert controllers when an approaching aircraft was below the minimum safe altitude.

Our work on client projects has taken us into many different control rooms, such as the control room at CERN
a short distance from Geneva Switzerland.  CERN operators are responsible for eight different accelerators
including the remarkable Large Hadron Collider (LHC).  Less than 24 hours after participating in a meeting in the
CERN control room building we watched, along with thousands of people around the world, the
restart of the LHC in November 2009.
 
For some client projects we are asked to help with the development of an alarm management system. Other
clients have asked us to evaluate an existing approach to alarm management and provide recommendations
for improvement.  In either case, the starting point for this activity is usually to identify the purpose of alarms.  
In theory, every alarm should require a response by an operator.  In practice, however, we see some alarm
management systems for which the primary purpose of alarms is to maintain situational awareness.

The theory and best practices of alarm management are not just applicable to control rooms staffed by an
operator sitting in front of an array of high resolution graphical displays. For example, the same fundamental
concepts and principles are also applicable to the design of alarm management system for a futuristic
passenger vehicle which may rely on very different technology such as a “seat shaker” to get the attention
of a driver rather than ringing a chime or flashing a light on the dashboard.

The trend towards increasing automation and remote operation of critical systems brings an increasing
importance to alarm management. A physical tremor may have alerted Jack Lemmon’s character in the
China Syndrome to an abnormal condition in the nuclear power plant before any alarm was sounded.   
But if the operator of a critical system is hundreds or even thousands of miles away, he or she must rely
on the alarm management system.

Helping clients address alarm management challenges is interesting work for us because it allows us to
simultaneously draw from a variety of experiences and areas of expertise. First of all, we need to determine
the relationship between abnormal states of the systems and potential hazards. Secondly, we must
understand the technology for detection of abnormal states - including the potential limitations of this
technology.  Thirdly, we must take into account human factors to ensure that alarms are effectively
communicated to operators.

An excellent resource for alarm management theory and practice is Alarm Management for Process
Control by Douglas H. Rothenberg. We also depend on guidelines published by Engineering and Equipment
Materials Users' Association, EEMUA 191. However, the best resource is the time we have spent observing
operators at work in control rooms across a variety of industries.