This article gives simple and practical guidance to managers, supervisors and operators on how-to recognize and deal with typical human-factor problems involving alarm systems in the Oil & Gas, Power Plants, Refineries, chemical and allied industries. It aims to prevent future accidents.
Here is a step-by-step approach for improving alarm handling; as with managing any other risk: firstly, identify any problems; secondly, plan what to do; and thirdly, eliminate or control them.
Take some measurements to find out:
Alarm rate targets: the long-term average alarm rate during normal operation should be no more than one every ten minutes; and no more than ten displayed in the first ten minutes following a major plant upset.
For example there is a tendency for hazard and operability studies (HAZOPs) to generate actions which result in a lot of ‘quick fix’ alarms being installed.
High operator reliability requires:
Can you demonstrate you have achieved this level of reliability, eg in safety reports and risk assessments, do you make unreasonable claims for the likelihood of operators responding correctly to alarms?
An effective alarm system should ‘direct the operator’s attention towards plant conditions requiring timely assessment or action and so should:
The team should include the right technical, operational and safety representatives. Relate identified problems back to the overall plant risk assessments.
Decide which ones present the biggest risks, and produce a timed action plan to deal with them. Identify and agree the necessary resources but be careful not to underestimate the effort involved.
A quick first-pass review may cover perhaps 50 alarms per shift but a thorough review and redesign may take more than 1 shift per alarm.
One company reviewed their alarm system after an incident and found it had poorly prioritized and designed alarms resulting in high alarm rates. They set up a project to review existing systems. It was run by a steering committee with a senior management ‘champion’. A multidisciplinary team, including operators, carried out the work.
They identified best practice and rolled out an action programme to:
The steering committee continued to review progress and strategy.
Well-designed simulators and simulator training can be very effective if properly integrated into the training program m e. Ensure that training is sufficiently realistic for both normal and abnormal conditions.
Improving alarm handling is not a one-off project. You now need to manage it in a systematic way as part of your normal safety or quality assurance management system to maintain control and ownership.
For example: Draw up a site/company alarm strategy and standard. The strategy should include a clear definition and purpose for all site alarms, a commitment to suitable training and reviews, and to ergonomic design. The standard should include a mechanism for regular review and alarm change-control, definitions of responsibilities, operator training requirements, etc.
Checking: This includes formal audits and reviews, consultation with safety representatives, informal feedback from operators and supervisors. For example, the original measurements already carried out (see Step 1) can be repeated to measure progress and to see if performance is now reasonable. The alarm steering committee (or its equivalent) has an ongoing role in this process.
Design of new alarm systems: Much of what we have written here is about improving an existing alarm system. It is better to design the alarm system right in the first place. Your alarm standard should be used to set suitable specifications for purchasing alarm system equipment.
Better alarm handling can have a significant effect on the safety of your business (the cost of not improving alarm handling can literally be your business in some cases).
An improved alarm system can bring tighter quality control, improved fault diagnosis and more effective plant management by operators.
A number of quick and relatively easy technical solutions are available which can bring immediate benefits. Medium and longer-term programmes can bring greater benefits still.
Note : Contains public sector information licensed under Open Government Licence v3.0.
Learn the difference between CompactLogix and ControlLogix PLCs with more I/O's from the Rockwell Automation…
Learn the electrical cabinet air conditioner cooling methods and the panel enclosure heat dissipation maintenance.
The control & instrumentation team in the detail design engineering company provides the package system…
In this article, you will learn the PLC ladder diagram example using memory bits and…
In this article, you will learn the PLC project examples for EcoStruxure Machine Expert software…
The weighing scales are found in many industries, almost every plant may have a weigh…
View Comments
Definition of the term “alarm” ?
The concept of alarms (French for “a l’arme” which means “spring to arms”) is very old and originates from the military concept where a guard warns his fellows in case of an attack. In process control, alarms have a very long tradition as well (e.g., a whistle indicating that water is boiling) and have evolved over time with older plants were using panels with bulbs and bells to alert the operators to large screen displays full of information.
However, in today’s control room we often encounter the situation where the operators are flooded with too many alarms. Hence making it difficult to decide where to focus their actions.
Historically, the configuration of the Panel Boards has been expensive, because each alarm required extra hardware and wiring. Therefore the designers of a control room put considerable thought in the decision of whether an alarm was meaningful or not. This form of implicit alarm management often resulted in good quality operator interfaces.
With the introduction of computerized control systems like the current generation of Supervisor Control and Data Acquisition (SCADA) systems, the cost associated with alarm configuration has been significantly reduced. Therefore, it is easy to configure several alarms on each and every point.
Engineers tend to hesitate before removing an alarm when unsure if the alarm is relevant. Modern field devices are able to generate a multitude of diagnostic messages, many of them issued as alarms. This typically results in huge amounts of configured alarms. As a consequence, during operation, operators receive huge amounts of alarms, even during normal operation.
Today it is very common to receive more than 2,000 alarms a day per operator. Most of those alarms are nuisance alarms that provide no value for the operators.
Definition of the term “alarm” according to ISA-18.2 ?
The definition means that an alarm must indicate a problem, not a normal process condition. The target audience of an alarm should be operators, not engineers, maintenance technicians, or managers.
Today’s reality in many control rooms is different: most of the occurring alarms have little or no value for the operators.
Great
Tnx