As the automation world continues to embrace Industry 4.0 technologies and smart manufacturing, SCADA systems are becoming more complex. As complexity increases, there is a greater likelihood that a process or machine will not meet specifications. Because of this, modern systems need to be constantly monitored as they run, which is why having a well-maintained alarm system is so important.
The right alarm settings will not only ensure that your system is running efficiently, but also prevent problems that could lead to downtime or worse. While every facility is truly different, we've gathered some simple best practices to help optimize any SCADA alarm system.
1. Properly Define Your Alerts
Every alert system starts with a basic question: what is an alert? The only defining characteristic of an alert is that it should always indicate an abnormal event or situation and never simply confirm that a process is running normally. Think of it this way: an alert should not be ignored. If you don't need to worry about it, then it's not an alert.
2. Priority alarms
Not all alarms are created equal! Some may warn of problems that could damage the entire facility and require immediate resolution, while others are less time sensitive. We recommend prioritizing alarms into 5 categories: 0-Diagnostic, 1-Low, 2-Medium, 3-High, and 4-Serious. By categorizing alarms, operators can quickly diagnose the severity of each event and take appropriate action.
3. Putting an alarm on hold
On hold is the middle ground between acknowledging or disarming an alarm and keeping it active. Think of it as pressing the snooze button. Operators can still view on-hold alerts as needed, but they are temporarily hidden from the main view so operators can focus on more relevant things.
4. On-call roster
Sending notifications to off-duty operators is never the best way to solve a problem. Synchronizing alarm notifications with operator schedules ensures faster response and reduces confusion when new operators take over.
5. Alarm Pipeline
Alarm pipelines provide a way to configure notification logic. Depending on how you set up the pipeline, you can determine who receives notifications and when, rather than sending alerts to everyone. Here is an example: An alarm goes off and notifies the shift operator only 3 times, with a 30 second delay between each notification. If no one acknowledges the alert, the pipeline escalates to notify the administrators (as a group or a specific individual). Once the alert is acknowledged, the pipeline stops. Essentially, the pipeline creates a customized way to increase the effectiveness of alert responses.
6. Staying in touch
There are several ways to send alert notifications. Two-way e-mail is a convenient, unobtrusive way to receive and acknowledge alerts. However, in more urgent situations, voice messaging may be the most appropriate option. For operators using mobile devices, they may prefer SMS (aka text messaging) because it is instant and can be easily acknowledged while working on a noisy factory floor.