It is very difficult to calculate the probability of failure on demand (or the risk reduction factor) for an operator’s response to alarm because it is dependent on many different things. The same operator can perform differently from one day to the next depending on their physical condition, amount of sleep, number of days worked in a row, and the level of stress in their personal life. Performance is also driven by issues such as nuisance alarms, alarm floods, poor human-machine interface (HMI) design, and insufficient operator training. A recent article published in Chemical Engineering Progress (CEP) discusses how human factors best practices can be applied to alarm management in order to change operator behavior and improve operator response.
There are many great takeaways from this paper. One that “sticks” with me concerns operators ignoring nuisance alarms, which is commonplace in control rooms today. When nuisance alarms contribute to an incident, a typical corrective response is to add more alarms (to back up the ones the operator is ignoring) or discipline the operator. These actions only make the situation worse.
When one looks at the situation through the lens of human factors, we gain a different understanding of how to correct the situation. Nuisance alarms represent conditions where the alarm is not true, or when no action is needed on the part of the operator. Since operators are busy people and have a lot of data to process through their mental models, the ability to discount data that is not important in order to focus on critical data, is a useful and desirable skill. Consequently, the ignoring of nuisance alarms is NORMAL human behavior and the best way to stop operators from ignoring nuisance alarms is to eliminate the nuisance alarms…
As stated by Mica Endsley, a leading human factors expert, “A person’s reluctance to respond immediately to a system that is known to have many false alarms is actually quite rational. Responding takes time and attention away from other ongoing tasks perceived as important” (Ref: “Designing for Situation Awareness: An approach to User Centered Design”).[/vc_column_text][/vc_column][/vc_row][:zh][vc_row][vc_column][vc_column_text]No! They are not Inherently Safe!
A collaborative robot is intended to work “collaboratively” with a person. i.e. share a common workspace. It is force and speed limited by design to minimize any potential hazard. Collaborative robots fit the application where the task cannot be easily or cost effectively automated. They are easy to deploy, program and repurpose. Collaborative robots are new to everyone including the standards agencies.
A hazard and risk assessment is required that assesses the robot and the environment that it is deployed in. Just as any other robot, things such as collisions, speed, type of end effector and worksite need to be evaluated. Collaborative robots have their own sorts of collisions and hazards. They may not be as severe, but they still exist.
This all comes down to risk and the amount of risk that you are willing to accept! The diagram below shows the high-level steps for doing a Hazard and Risk Assessment. When following the steps, if you assess the risk and find it to be acceptable (your companies acceptable risk norms) then you are done. No need to add any risk reduction.
The next best approach is to determine if protective measures other than a Safety Function can reduce the risk to an acceptable level. If not, then you must assign a SIL and implement a safety function that will provide the required risk reduction.
exida can effectively train your team to perform machine hazard and risk assessments to identify all possible hazards and estimate the risk for each hazard. Specifically, exida coaches you through the process of evaluating the risk, developing and implementing risk reduction options. exida can also educate your team in multiple approaches to SIL target selection. These are just some of the things exida does to ensure you are on the right path![/vc_column_text][/vc_column][/vc_row] [:]