Written by Garrett Harvey | August 7th, 2023
As the owner, manager, or operator of a building or group of buildings, you are probably always looking for ways to make your facilities run more smoothly. Hopefully you utilize some kind of supervisory control system, often called a building automation system (BAS) or building management system (BMS), which provides tools for AST- alarming, scheduling, and trending. While data trending is an important troubleshooting and analytical tool, and intelligent scheduling is an effective energy-saving practice, alarming is often the key BAS function that impacts the day-to-day operation and maintenance of building equipment. For example, facility operators and maintenance technicians may rely on alarms to identify issues and prioritize repairs, while facility managers may use alarm metrics to gauge the performance of building equipment.
However, poor alarm management can nullify these benefits and even make it more difficult to control and maintain a building. Anyone familiar with facility alarms is likely all too familiar with the common problem:
That’s right, there are just too many alarms- or more specifically, too many alarms with little to no useful information. These nuisance alarms and the resulting alarm fatigue cause actionable and even critical alarms to go unnoticed, and they force facility operators to spend more time deciphering alarms than identifying and addressing their root cause. The solution to this problem is not disabling alarms, simply reducing the number of alarm notifications, but instead better alarm management. From our experience working with facility managers and operators, Essex has identified several key methods to achieve better alarm management:
Let’s take a look at each of these methods to see how they not only improve ineffective alarm infrastructure, but are necessary to get the most value out of your facility alarms.
Perhaps the most straightforward way to improve facility alarms is - no surprise - better alarm programming. Your facilities are likely composed of complex systems with many interconnected parts, meaning the state of one component can affect the state of many others. Without hierarchical alarm suppression, an alarm from a single upstream component can cause an overwhelming number of alarms (i.e. an alarm flood) in downstream components. For example, consider a facility with 5 air-handling units and 100 terminal units. The AHUs are served chilled water, and if the chiller fails, the AHUs cannot maintain their cooling setpoint even with the cooling valves fully open. Likewise, the terminal units will likely be unable to cool their spaces. Without alarm suppression, it’s possible that all 100 terminal units, each of the five AHUs, and the chiller will trigger at least one alarm, for a total of 106 alarms in the span of an hour:
However, with alarm suppression, only the first alarm is triggered - the chiller - which reduces the number of alarm notifications and helps to pinpoint the root cause of the problem. The suppressed alarms can be added to a report to be reviewed later, if desired.
Another important consideration is alarm configurations - the high and low limits and delay times that define the alarm conditions. The reality of the built environment is, not all systems have the same design. In the case of HVAC systems, equipment may require different setpoints and age at different rates. As a result, the same alarm should require different settings for different systems. An intelligent alarm system should make it easy to apply the same alarm to multiple equipment, but also tailor the configuration of each alarm to fit the characteristics of specific equipment. One way to streamline this process is through the use of metadata tags to store alarm settings. Software applications that enable the batch modification of tags make alarm configuration straightforward and ensure fine-tuned control over alarm programming.
Proper alarm programming will only go so far to simplify alarm management. The next major step to improve any alarm system is to introduce better alarm organization. Control systems with alarm functionality commonly include alarm prioritization, to assign relative importance to different alarms. Alarm priorities in and of themselves are a great tool to organize alarms and help operators focus on those most critical, but just like any other tool, alarm prioritization is only effective when used properly. Overwhelmed facility managers and operators are often inconsistent in their application of alarm priorities, or hasty in creating and modifying alarm priorities without considering how easily they can be managed. For this reason, Essex has found the best way to use alarm prioritization is to plan and develop an alarm standard that clearly defines what priorities will be used, to which alarms they will be assigned, and the operator response that the alarm requires. As an example, the EEMUA suggests a standard with three priority levels, specifying that the two highest priorities should make up no more than 20% of a facility’s total alarm count [1]. As one might expect, for a standard to be effective, all facility managers and operators must commit to it; that means the best standard is simple to put into practice. With that said, more complicated alarm priority structures can be devised and implemented well if facility operators are equipped with the right tools to manage them. Once again, metadata tags can be used to define alarm priorities, modify them, and apply them throughout a facility all at once. With a comprehensive alarm standard and powerful management tools, facility operators can trust alarms to organize themselves, so they can focus on fault resolution.
In addition to defining alarm priorities, an alarm standard can even define alarm programming for consistent alarm behavior across all facilities. The alarm standard could then be given to designers or referenced directly in project specifications, to ensure new controls work adheres to the established alarm protocol. With regard to any operational standard, the importance of a regular review and revision process cannot be overstated; an alarm standard must be evaluated regularly to determine if it still meets the needs of your facilities. When developing an alarm standard, remember the age-old KISS principle:
Facility alarms are meant to assist operators by identifying system abnormalities, not burden them with notifications or lead them down rabbit holes. So far we have addressed ways that alarms or the overarching alarm system can be improved to achieve this goal, but none of them really matter without better alarm utilization - helping facility managers and operators to use alarms, or more importantly, alarm information, more effectively. The first step in using alarms better is, as one might guess, getting useful data from them. In most systems, the alarm notification will include the basic information necessary to investigate the issue - the timestamp of the alarm, the equipment/space from which the alarm originated, the alarm priority level, whether or not someone has acknowledged or silenced the alarm - but after correcting the individual issue, you can go a step further to evaluate groups of similar issues:
A view like the one above is created by aggregating data across all alarms, and since few people likely have the time or patience to perform such a task, this process should be automated - perhaps within the BAS itself, but more efficiently with an analytics platform that can be seamlessly integrated with the BAS. The alarm metrics calculated by such an analytics platform can be used to measure equipment and building performance as well as track operator responses to alarms. For example, the process industry has, since 1998, generally accepted one alarm every ten minutes as a reasonable alarm rate that gives operators enough time to respond [2]. In 2009 the ISA published an alarm standard stating 300 alarms per day, or about two every ten minutes, as the rate at which human operators begin to be overwhelmed [3]. For understaffed teams, these numbers drop dramatically. Every facility is different, and you can start working with your facility team to improve once metrics like these are tracked and a baseline established.
It’s worth noting that these alarm improvements are usually easiest to implement in a centralized alarm system. Facility operators often have to deal with a variety of control systems from an assortment of vendors - sometimes even within the same building. Each system has its own type of alarms that may be difficult to reconcile with others. As a result, operators get notifications from all directions and have to log into multiple interfaces to troubleshoot and review alarms. Instead, all facility alarms, as well as other BAS data, can be integrated to a single, enterprise-wide hub enabling comprehensive, high-level monitoring and control.
The three alarm improvements we’ve discussed - better programming, better organization, and better utilization - can radically enhance the effectiveness and ease of alarm management at your facilities. Essex can help you at every step of the way, from developing and benchmarking an alarm standard, to evaluating and modifying your current alarm programming, to creating and automating alarm metric reports sent directly to the right inboxes. Apply some of these ideas to your facility alarms, and contact us to learn in what other ways your facilities can be better with Essex.
References:
[1] EEMUA Publication 191: Alarm systems - a guide to design, management and procurement, Third Ed.
[2] M.L. Bransby & J. Jenkinson, CRR 166/1998: The Management of Alarm Systems
[3] ANSI/ISA-18.2-2016: Management of Alarm Systems for the Process Industries
🙂