Alarm management: 6 hazards, 4 strategies

Alarm management in an IIoT world: Correlation and classification of industrial process control alarms before the operator sees them vital for safety, speed, and efficiency. Networking and advanced software tools help.

By Shashidhara Dongre and Manoviram Rath, L&T Technology Services February 8, 2018

An alarm management system communicates any hindrance in the smooth functioning of plant operations. Plant manufacturing remains a key element of the overall business ecosystem and business continuity is a direct function of the uptime of such plants. Large displays that show a number of alarms and information associated with it could be baffling to the operator. This could strain human alertness levels and hamper the operator’s capacity to deal with a potential situation.

General results indicate a typical plant loses more than 5% of its total capacity each year due to slowdowns and an approximately equal amount due to off-spec product, quality giveaway, and other lost opportunities, not considering any unscheduled unit outages that might occur. 

Smarter alarm classification

Correlation and classification of alarms before they are shown to the operator is vital. While information overload is a reality in the control room, only showing alerts on the screen is insufficient. It may be worthwhile to explore the business benefits brought about by the visual and aural indicators like sound buzzers and colored lights bulbs so as to capture the operators’ attention for dealing with the situation systematically. Companies also are exploring propositions around augmented reality (AR) and virtual reality (VR) to complement alarm management.

An alarm management solution that connects alarms from disparate multi-vendor systems should be integrated to enhance its efficiency. Integration helps classify the alarms as major or minor, which is an imperative part of the decision-making process. Data insights derived with integration are superior and this data can be used to deploy a corrective action plan. The information also helps the operator forecast the life of the equipment by studying its operation history to check for any product failure as well as fluctuating performance levels.

Factories with functional alarm management systems also report a sustained reduction in the variable maintenance and insurance cost. While the above might seem like an idealistic approach, the immediate business benefits are immense. 

Six alarm management hazards

Alarm management systems are one of the undervalued aspects of process automation and face a number of challenges, which could prove to be a hindrance to the process operators responding to a critical situation. Six prominent hazards are highlighted:

1. With the advent of low cost sensors and digital controls, the ease and low cost of adding alarms could lead to unchecked growth in the number of alarms installed at the plant. This leads to the problem of alarm floods, which is where the same fault triggers multiple alarms in a short span of time. This makes it difficult for operators to ascertain the underlying cause.

2. Multiple alarms show extensive amount of data to the operator. During an event when these alarms are triggered in a flood, it could lead to the operator missing critical signals in the chaos.

3. Without a disciplined alarm management program, the systems and momentum will be disrupted due to false triggers, which will prevent operators from correctly identify the critical alarms and signals.

4. Alarm systems that have not received the attention and resources that are warranted often encounter the issue of nuisance alarms, which are triggered when no abnormal condition exists or when no operator action is required.

5. Many alarms are assigned with the wrong priority, which makes it unimportant for the operator, and potentially meaningless. This can lead to the wrong choice of action when multiple alarms are triggered.

6. A number of plants still use the alarm management philosophy that was in practice when the plant was built. The systems are rarely revisited for revaluating the alarm limits and priorities, which leaves a question mark over the alarm system’s integrity and effectiveness. This adversely affects the speed and accuracy with which the operator can identify the alarms that require immediate action.

A plant needs a robust alarm management strategy that enhances operational efficiency and timely detection of failures. More importantly, the system needs to be in sync with modern requirements. A comprehensive alarm system provides actionable information to the operator and provides assistance in taking corrective action. 

Integrated alarms: 4 steps

Strategies are changing for deploying an efficient integrated alarm management system because of the Industrial Internet of Things (IIoT). The IIoT presents numerous opportunities for developing industrial applications that cater to broad ranging requirements like automated and pro-active monitoring, control, decentralized decision making and maintenance.

Evolving IIoT platforms can be used to enable machinery in industrial processes to be integrated with sensing, identification, processing, communication, analytics, and networking capabilities. This allows industrial objects to be monitored and controlled remotely across existing network infrastructure. Integrated alarms should follow these four steps. 

1. Alarm integration and correlation: Alarms in a plant typically come from disparate multivendor systems. These alarms should be integrated using IoT as it will enable the operator to quickly identify the underlying cause of the issue. Sensors can also be integrated over existing systems to augment the overall efficiency of the alarm management system. Another concept that greatly complements the modern alarm management systems is condition-based monitoring (CBM). For example, rotating machines that use current, temperature, and vibration sensors to generate alarms should be integrated with traditional alarms to complete a value chain to help ensure foolproof and intelligent alarm management. 

2. New-age AR and VR: AR and VR technologies can save the operator a significant amount of time in resolving an issue. Typically, issue resolution has two phases. The first is finding the root cause and the second is applying the fix and testing it. In most cases, finding the root cause takes up 80% of the time; fixing the issue takes 20%. AR- and VR-enabled technology can reduce the time consumed in finding the root cause of the alarms and help enable quick resolution of the issue. 

3. Cloud-based alarm management: A cloud-based alarm management application enables an organization to enhance the overall alarm management mechanism. It also provides the plant manager access to real-time meaningful alarms even when being mobile or away from the plants to help facilitate quick action. For example, large organizations with a number of sites having alarm management would benefit from a centrally managed, cloud-based solution. Administering and upgrading all these sites becomes much easier and cost effective. For smaller end users that don’t have the resources needed to deploy a fully functional alarm management solution, a cloud-based solution can again provide a cost-effective alternative, without the need to manage on-site IT infrastructure and other administrative requirements.

4. Integrated operator response guidance: It is likely an abnormal situation might arise where the operator is incapable of dealing with the situation efficiently. To overcome this challenge, one can integrate operator response guidance and abnormal situation response guidance into an alarm management strategy. It helps in improving operator effectiveness by providing the operator with the correct actions to take, depending on the situation.

The above strategies may not be always considered greenfield pursuits. These can be integrated with existing systems as well and be upgraded to modern requirements.

A typical scenario in large plants is encountering hidden values and, therefore, data that would otherwise pass off as masked abnormal or "near-miss" situations. Considering the critical nature and dependency of productivity from such large plants, such near-misses cannot be allowed to pass without action. Event context is needed for improvement. This calls for organizations to have offline alarm management systems in place to further complement the online digital mechanisms.

A pure offline exercise also is required. Having such a mechanism in place ensures insightful data is available to assist in avoiding accidents and disasters in the future. The data also can be linked to machine-learning algorithms and predictive analytics architecture to enhance the intelligence quotient. 

Effective alarms, fewer losses

Improperly configured alarms and poor alarm management strategy are sure to contribute to accidents. Plant downtime costs are often underestimated and in today’s environment where operating profit margins are strained, it necessitates outlining key concepts and governing rules for alarm strategy to reduce unscheduled plant shutdowns, refine performance, and mitigate the risk of incidents and excursions.

Shashidhara Dongre is head of delivery, plant engineering, and Manoviram Rath is discipline head instrumentation & control, both of L&T Technology Services, a CFE Media content partner. Edited by Mark T. Hoske, content manager, Control Engineering, CFE Media, mhoske@cfemedia.com.

KEYWORDS Alarm management, alarm strategies, alarm hazards

  • Alarm-management hazards can include an expanding number of sensor inputs.
  • Integrated alarm strategies include smarter correlation and augmented reality.
  • Effective alarms translated into better, safer operations.

CONSIDER THIS

How can smarter alarm strategies and technologies improve your plant overall effectiveness? 

ONLINE EXTRA

This online version contains more information than would fit in the Control Engineering February 2018 issue.