Webinar: Alarm Management - Yokogawa Control System

welcome to the webinar alarm management getting the most out of your yokogawa control system my name is todd stoffer and i will be presenting to you so if you have questions along the way please enter them into the questions box and we&#39;ll take a look at them at the end and if we have some time to review them and answer them live we will otherwise we&#39;ll get back to you afterwards so as i mentioned my name is todd stauffer the director of alarm management here at exeter which means that&#39;s what i focus on spend pretty much most all of my time engaged in the activities of alarm alarm management a lot of them follow from the isa 18.2 standard which i&#39;ve had the privilege of helping to write and shepherd through the process since 2005 i also wrote or participated in writing a couple of the isa technical reports and my latest role is as a co-chair for a new standard on safety controls alarms and interlocks bringing together the worlds of functional safety and alarm management so if this is your first time thank you for joining a little bit of background information about exeter we were formed a little over 20 years ago by two gentlemen that wanted to make the world a safer place they actually left their jobs to start a new company called exeter that would focus on functional safety and then several years later they expanded the scope of the company to also include cyber security and alarm management so other areas that affect risk in a process plant exeter is a leader in the certification of products and systems and people by products i mean safety devices so if you&#39;ve ever put together a safety instrumented system and worked with sill rated equipment chances are it was exeter that did those sill verifications or validations also have a program for certification of people so that they can acquire the knowledge that they need and demonstrate the competence that they need to perform a functional safety lifecycle tasks and speaking of the functional safety life cycle we have a suite of tools called excelentia that actually provides safety system design and engineering capabilities and guides you through the whole entire functional safety life cycle one module in that suite is called sill alarm which is an alarm rationalization tool and that&#39;s something that we&#39;ll talk about today so in terms of the agenda we&#39;re going to look at alarm management functionality things like prioritization how do we do rationalization for a yokogawa system etc alarm shelving we&#39;re also going to look at some of the key tools from yokogawa and others that would help deliver this alarm management functionality so as our guiding light we use the standards and guidelines that you see here a mua 191 first created in 1999 and actually proposed the concept of alarm shelving which we&#39;ll talk about quite a bit and then the isa 18.2 and iec 62682 alarm management standards which are pretty much identical to each other the isa 18.2 standard was first published in 2009 was updated in 2016 and the iec standard was first published in 2014 and is in the process of being updated right now hopefully for release later this year so both standards i say and iec define the concept of an alarm management life cycle which is basically the workflow process to follow for alarm management to create a sustainable program over time so let&#39;s look at how some of those tools that we mentioned earlier fit into this life cycle so for alarm rationalization which we&#39;re going to talk about there are a couple of different tools that can be used we&#39;re going to focus on using the sew alarm tool and the exit quantum amd tool from yokogawa the cams software also has some capability for rationalization as well it really comes into play though when we look at the presentation of alarm information to the operators as well as invoking alarm suppression functions and advanced alarm functionality so we&#39;ll talk about what this cams software can provide from that point of view there&#39;s another tool from yokogawa called exopilot that we&#39;ll look at how that can be brought into the picture to achieve some advanced alarming for the monitoring and assessment stage of the life cycle you can use either exa quantum ara again from yokogawa exa plug another product from yokogawa where of course there there&#39;s other third-party pieces of software as well and then from a management of change point of view you can use the sill alarm tool the exaquantum amd tool and cams as well so those are the key tools that we&#39;re going to look at let&#39;s just go through them relatively quickly to kind of define what they do so first and probably most important for a yokogawa system is the cams software cam stands for consolidated alarm management system so this is a piece of software that sits on top of the dcs within the hmi and actually controls the presentation of alarm and message information to the operator so it actually takes the alarm information that comes up from the controllers and presents or adds additional information to it in some cases actually enhancing it for presentation to the operator this software also provides the capability to do advanced alarming things like alarm shelving alarm suppression significant filtering is built in much of which we&#39;re going to spend some time talking about so that&#39;s the cams software exaquantum ara is a tool for performance monitoring and assessment you can see some of the screen snaps there and these are some of the standard canned reports or capabilities that it has for generating performance reports also in the xaquanum family is a tool called amd which is something that acts as a master alarm database so you load your rationalized alarm configuration into this database and then you can use it for management of change auditing enforcement etc to help maintain the system integrity of your alarm system then there&#39;s the exa plug software which is for event analysis so it can analyze alarms and present reports of what&#39;s happening that you can use to drive your alarm management program it also does analysis of operator actions and events that occur so you can see correlations between those and alarm event activity it&#39;s a it&#39;s a lower budget tool compared to the exaqu so depending upon what your price point versus functionality needs are this may be something that&#39;s useful to go for or the the ara provides i think more capability and is a slightly higher cost one of the other tools that we&#39;ll talk about is the exapilot software which is used for procedural automation so it basically provides a way for operators to create their their own semi-automatic sequences instead of doing things via manual operation it also can be used for some advanced alarming capability suppression of groups of alarms we&#39;re going to show how it can be used for static suppression and dynamic suppression and then last but not least is the sill alarm tool from exeter which will guide us through the rationalization process it&#39;ll also act as the master alarm database it can define alarm response procedures it can do advanced alarming design and also has capabilities for doing auditing and enforcement as well so one of the things that you&#39;ll see or probably think about is we have different tools here that kind of do the same thing and a little bit overlap certainly that&#39;s the good news and bad news good news is you have different options to choose from bad news is you have to figure out what&#39;s the best fit for your situation now in terms of how we&#39;re going to talk about alarm management we typically follow a seven-step process that looks like this that starts with some initial benchmarking and the creation of an alarm philosophy document and then you get into continuous improvement mode and that&#39;s what we&#39;re going to spend most of our time talking about today is how the the tools and functionality help us continuously improve our yokogawa system to make the alarm system performance acceptable so we&#39;ll go through the rationalization phase we&#39;ll talk about advanced alarm design we&#39;ll talk about implementation and building of alarm response procedures and then look at the capabilities for performance monitoring and assessment that is kind of the backbone of a continuous improvement loop here we&#39;ll also then look at some of the audit capabilities for comparing the alarm system configuration that&#39;s actually running in the dcs versus what you expect or want coming out of rationalization so a little bit of architecture overview you can see here that alarms are actually generated from function blocks that exist in the controllers they that information gets propagated up through cams gets processed further in cams additional information can be added some information can be changed for example if you were just to configure the priority of alarms in these function blocks you would find that every alarm from a given tag has the same priority it&#39;s not the best approach nowadays that we know that we want to analyze each alarm condition for its uniqueness and understanding what it&#39;s trying to prevent so in most cases it makes sense to provide the flexibility to have alarms that can have different priorities for each alarm condition so that gets done by cams up here before it&#39;s presented to the operator so the yokogawa control system natively provides high medium and low as process alarm priorities when the information goes through cams it maps into high medium and low but then it adds a fourth alarm priority called critical that you can use for your extra special most important alarms now we want to talk about how to prioritize alarms what are options how can we do that for a yokogawa system so priority is pretty important that&#39;s what operators should be able to use to determine which alarm do they respond to first so when you have multiple alarms present at the same time the idea is priority guides them to the one that is more important from a business point of view and the way we typically evaluate or assign priority is based on evaluating how bad the consequences would be if the operator didn&#39;t respond and how much time is available to respond combination of those two dictates the priority so here is a typical example of the way the majority of of people that i&#39;ve run into prioritize alarms you have a top portion which allows you to assess how bad the consequences would be this typically comes from a risk matrix used for a pha or a has up and then down below here we&#39;re factoring in the operator response time the urgency for them to respond and it&#39;s the combination of the two really significant consequence short response time that dictates the priority of the alarm this is the most common method of prioritization now cams does support some rationalization capabilities and some and some prioritization capabilities what it supports is actually prioritization based on the purpose of the alarm meaning safety environmental or financial and the worst case consequence is it large very large medium etc so you can see here based on that combination you would end up with a priority it doesn&#39;t include or factor in the response time though so if you have a matrix like we just looked at which a lot of people do you would not really be able to use this prioritization wizard with that matrix that you&#39;ve defined so that&#39;s one of the reasons why we&#39;re going to assume that we use sill alarm as the rationalization tool and master alarm database for this discussion so here&#39;s an example of how we would set up so alarm to define our prioritization scheme based on the information that would be in your alarm philosophy defining the assignment of priorities based on the combination of how bad the consequence is and the response time so now let&#39;s look at the alarm rationalization process so to start things off we need to load the alarm configuration into our rationalization tool to essentially create our master alarm database we can do that for an existing system by extracting from cams or we could do it for a greenfield system from an excel spreadsheet so cams actually has a very nice export and import interface so in this case we can select which pieces of alarm configuration we want to export and then we can give it a name in the export file choosing names that line up with the terminology that we have in so alarm so it makes it very convenient for us to do the mapping to be able to take and export out of cams and be able to load it right into sill alarm and once that&#39;s done you can see here we&#39;ve got all of the alarms from one typical pid loop within the cell alarm database now we can proceed with the rationalization the first step is typically to evaluate the validity of the alarm and the alarm objective analysis by looking at what is the consequence that would occur if the operator doesn&#39;t respond what are the likely causes what are the corrective actions now you might think who who&#39;s able to define that information who knows that well rationalization really requires somebody who is very familiar with the process to provide this information and that typically resides in the brains of your senior operators your most experienced operators so they are key to a successful rationalization it also means that rationalization becomes not just an activity to improve your alarm system performance but it can also become a knowledge capture process because we can essentially document what your most senior experienced operator would do when that alarm occurs and then we can make that information available to all of the operators hopefully through the hmi in an alarm help document either a document like you see here which is a report coming out of sill alarm or as we&#39;ll see in a little bit within the cams interface but a key thing coming out of rationalization is to be able to extract that great information that you&#39;ve captured from your senior operators and make it available so every operator can see that when they get an alarm they can look and see what that senior operator would do if we&#39;ve once we&#39;ve defined the consequence now we&#39;re ready to prioritize that alarm so here we have it predefined and so alarm to match up with your priority matrix as defined in your philosophy so we&#39;ve got the imported priority essentially what the priority in the control system is originally and now we do some consequence ranking we evaluate how bad the consequence would be in each of our impact areas and we record and document that and then we also estimate how much time does the operator have to respond and based on this information it pulls out of the priority matrix what the recommended alarm priority should be and it maintains all this information it records it so we&#39;re not just recording the worst case consequence we&#39;re actually recording the consequence assessment in each of the impact areas so if something changes later on where you think that you may have made a misinterpretation or a mistake or you&#39;ve got some new information you can go back and look at what was done to be able to update it accordingly a lot of tools or if you&#39;re just using an excel spreadsheet to do this uh you may not fully document that rationale for setting the priority in which case it becomes kind of challenging to recreate it or modify it if you need to so we get the alarm priority coming out of there we can also override that priority and set it manually so we could set it based on the philosophy or we could set it to a non-alarm type notification like an alert or a prompt or message something that&#39;s going to use the alarm system for presentation but that notification is not as important as an alarm now in cams as i mentioned there is some capability to do rationalization and to document the information here&#39;s where we would define or go through the process of setting what the alarm priority would be by selecting a purpose in this case safety environmental financial and a consequence how bad is it and as i as we showed previously based on the combination of the purpose and the significance of consequence that would give you what the priority is no inclusion of response time though next we would want to define the alarm classification which is a way to group alarms that have common sets of requirements for things like testing or training or moc this allows you to show or keep your eyes on alarms that are more important or the most important alarms in your system so it could be alarms that are there for environmental purposes or to prevent environmental incident could be for safety purposes for personnel safety or alarms that are listed in a layer protection analysis those are good examples of classifications so during rationalization we want to identify what or assign what classification is appropriate for safety purposes and functional safety we want to go a step further and actually document the functional safety usage of each alarm so if it has a role as a safeguard in a hazop or as a protection layer in a layer protection analysis we want to document that so then it becomes easy to find in our entire database of alarms which ones are being used for functional safety purposes and then we also can create a cross-reference by documenting what lopa study it comes from or documenting what has upper pha the alarm was identified in this allows us to make sure we&#39;re taking these alarms seriously and not making design changes or disabling alarms coming out of rationalization where those alarms are actually counted on to provide a safety risk reduction now in the yokogawa cam system alarm classification or alarm class is not defined or used out of the box but what we&#39;ll see or what we&#39;ll talk about quite a bit is that you can define your own fields in cams so we could set up a field to be called alarm class or alarm classification and then that could be used to present that information and to hold that information now another thing that i wanted to mention about silarm when we document the alarm limit we can document the consequence threshold what we&#39;re trying to prevent from occurring we can document any process dynamic information that we know the rationale for setting the alarm limit and we can define our own user parameters that are unique to a certain entity or a certain project or certain site to document additional information that&#39;s relevant to the design of the alarm so if we&#39;re trying to set the alarm limit for a alarm let&#39;s say for a tank level we want to know whether we&#39;ve defined the safe operating limits where is there an operating envelope that we want to stay within that information can be documented in so alarm so we can define those as extra fields that we want to capture and put information in now part of the reason why i&#39;m focusing on this is we can then push that into the cam system if we decide that some of this information is useful and you&#39;d like to present it to the operators we can create these fields or any combination thereof define them in cams and then import this information coming out of the sill alarm rationalization process so the last part of rationalization that we&#39;ll talk about is defining some advanced alarming and in this case the requirement to allow an alarm to be shelved or not we&#39;ll talk a lot more about that shortly so when you&#39;re all done rationalization this is the information that you should have documented so part of what we want to talk about now is how do we get this information into the cam system and how do we present it to the operator because a lot of this is very useful good information to help them know how to respond to each alarm so when we are done rationalization we can export the information out of so alarm and then use the reverse of the original process and import the information back into cams so if we&#39;ve defined some additional fields and cams to bring in the extra rationalization information that we think is valuable we can make that process all work through this interface and the correspondence and the mapping of here&#39;s what that field is called in cell arm and here&#39;s the field that we want that information to be put in in cams so now let&#39;s talk about how we can make the most of that alarm rationalization information and actually create alarm response procedures so an alarm response procedure you could think of that as a as a job aid for the operator what kind of information should it include here you can see what&#39;s recommended so it&#39;s information like what are the potential causes of the alarm what&#39;s the consequence if you don&#39;t respond what is the operator action and what we really mean here is the actual description of what the operator should do so consequence we&#39;re not interested in is the consequence large small severe not severe we&#39;re actually interested in the description of what the consequence would be loss of containment uh potential damage to equipment that kind of stuff so let&#39;s look at how those alarm response procedures could be presented to an operator through a cam system there is a guidance tab provided for alarm messages which can be used to present information including the call up of a file like the file that we saw that could be created out of sill alarm so that is one possibility for providing alarm response or alarm help information to an operator but what we&#39;re going to focus on is actually a different uh implementation method and that involves changing the information that we see here and filling it in with more information and more information that&#39;s got useful details that the operator would benefit from so if we look at some of the fields here we have consequence we have time to respond we have purpose they sound pretty useful but if we look at the information that&#39;s in there it&#39;s really just from a drop down list so safety large urgent that&#39;s not really going to provide much benefit to the operator in terms of actionable advice so what we want to do is add some fields and cams and load those fields with information coming from rationalization so and then we want to present them to the operator so we would replace this window with something that looks like this and it&#39;s got some additional and different information in it so now we have the time available to respond and it&#39;s actually a time measurement that will make sense to the operator we&#39;ve got guidance or corrective action that actually tells them what they need to do we&#39;ve got a consequence that tells them what piece of equipment or part of the process is going to be impacted so much more actionable intelligence it can also then include some of that other alarm database information so if we know that there&#39;s a safe operating limit or there&#39;s an operating performance boundary we can define that and show that within this window as well so a lot of different information that we can pull out of rationalization and then present to the operator here so as soon as the new alarm occurs the operator can click to this screen and then be able to see this information as part of their response process hopefully making the response quicker more consistent and more effective now let&#39;s talk about alarm suppression so suppression is a way to present or prevent the alarm from being presented to the operator when the base alarm condition is present so basically means the operator doesn&#39;t need to see it for one of many different reasons the first type of suppression is called design suppression that&#39;s where the control system is able to determine when the suppression should occur automatically and then automatically suppress the alarms that should be suppressed of course you need to configure it to do that but once you&#39;ve done that that will all happen behind the scenes and those alarms are under control of the logic the second type of suppression is called alarm shelving that&#39;s where the operator can temporarily suppress an alarm let&#39;s say an alarm that&#39;s behaving like a nuisance alarm they can temporarily suppress it so it gets out of their view and it is not a distraction anymore so that&#39;s what shelving is for it&#39;s different from disabling the alarm the alarm will come back after a period of time and then the third type of suppression is called out of service this is where the alarm would be suppressed typically for maintenance of the underlying hardware now shelving is actually a required function in the isa and iec standards so it&#39;s now required functionality of a dcs now there are different ways of suppressing alarms in a yokogawa system some that are done directly in the control system control logic some that are done through cams some that are done through the hmi we&#39;re gonna focus primarily on the options for cams so now let&#39;s talk about alarm shelving functionality as provided by cams and i call this one of the best kept secrets yokogawa is actually one of the first dcs suppliers to provide this functionality was actually introduced in 2006 so before isa 18.2 was even released so it was built in based on the information and guidelines in the amua document but it&#39;s been been around for a while part of the reason why we&#39;re going to go through this is there are many customers of yokogawa or that have yokugawa systems that are not aware of the capability of shelving or perhaps haven&#39;t really dove into it to decide how they could take advantage of it so that&#39;s part of what we want to talk about here so when you are trying to figure out how to use shelving you want to consider things like should we allow every alarm to be shelved or only alarms of certain priority or certain classification is there a limit for how long alarms should be able to be shelved is there an authorization process can the operators do it themselves do supervisors need to do it so there&#39;s a lot of things to think about in order to make it an effective system for your operation and the way you choose to do it could be different from one site to the next one of the the ways this is defined then is by setting up different groupings so different alarm shelves so essentially we define different purposes for different shelves and then the operator when an alarm needs to be suppressed when it needs to be shelved they essentially drag and drop the alarm to whichever of these shelves they have access to and whichever one they think is most appropriate given the different characteristics of each one so we&#39;re going to talk about how to set them up in a second now the maximum amount of time that an alarm can be shelved in a cam system is for 20 24 hours and one of the other key elements related to cams and shelving is to understand that there&#39;s actually different forms of shelving that are supported one shot shelving continuous and automatic shelving we&#39;re going to talk about what that means so when you define a shelve you declare whether it&#39;s going to be a one-shot shelving type or a continuous shelving and those behave differently and kind of have different purposes so a one shot shelving means the alarm is suppressed and it will stay suppressed for the period of time that you select unless the alarm clears while it&#39;s on the shelf then it will come back to be able to be triggered if that condition should arise so it doesn&#39;t necessarily stay shelved for the whole entire time that you&#39;ve selected it will unshelve earlier if the alarm clears now when would that be useful that would be very useful for stale alarms where the alarm is active for a long period of time but as soon as it clears you would like it to be back able to generate or to enunciate again so rather than waiting for the end of the shelving period if it clears you want it to go back to an enabled alarm ready to trigger if necessary now that&#39;s different from the alarms that you would use continuous shelving for like a chattering alarm so continuous shelving the alarm stays on the shelf for the period of time that you have selected until the time is up or the operator has chosen to manually unshelve the alarm now why that&#39;s important is you can imagine if it&#39;s a chattering alarm if you had to reshelve it each time it came off each time it cleared then you&#39;d be shelving the chattering alarm each time you acknowledged the alarm or each time you silenced the horn so you&#39;d be adding another extra step another annoyance it would make the situation actually worse so that&#39;s why this different type of shelving is ideal for a chattering alarm situation and then the final type that&#39;s supported in cams is automatic shelving which can be triggered when you detect the scenario that would lead to an alarm flood or if you&#39;ve got a piece of equipment that you&#39;re gonna temporarily you want to temporarily suppress all of the alarms from that equipment those are both forms of automatic shelving that&#39;s supported in cams so how do we actually configure it then so here&#39;s where we build those different shelves or different groupings different parking lots for alarms to be placed so what we define for each shelf give it a name a description the type of shelving so is it a one shot continuous or automatic then we define the type of shelving so is it being specified for a certain time period like 30 minutes or eight hours or do you want it to expire at a specified time like the end of the shift that&#39;s what that is really useful for so you can decide when you&#39;re creating these shelves or defining them whether you want it to be for a fixed period of time or at a specific time to elapse then you set the maximum amount of shelving time and you also get to define who has access to the shelve to be able to put alarms on it to be able to shelve alarms and put them there so there could be certain shelves that are for the operator and then certain shelves that are maybe for a supervisor or for maintenance and that can be controlled by access so we want to define the different shelves and label them so that the operator understands what they&#39;re used for because when they see a nuisance alarm that needs to be shelved they&#39;re going to drag and drop the alarm to whichever shelf they think is most appropriate and for which they have access to now behind the scenes in the engineering side of cams you can define whether an alarm should be able to be shelved or not so each alarm has a separate enable disable parameter for shelving so they&#39;re all set to defaults this information can be set externally from an excel spreadsheet or from so alarm for example so after you&#39;re done rationalization depending upon what your rules are for alarm shelving could be based on the priority could be based on the classification you could set in the master alarm database whether you want shelving to be enabled or not based on priority and then when we export that information and import it back into cams it would update the shelving parameter for all of the alarms as appropriate we also get to define what information appears in the shell display information like when was the alarm shelved so the operator can look at that to know how close they&#39;re getting to the time when the alarm will come off the shelf so then the way the operator actually interacts with it is we&#39;ve got some new alarms down here if they are nuisance alarms or alarms that the operator feels should be shelved they drag and drop it to whichever of the appropriate shelves best meets their needs now when the alarm shelving time period has elapsed and the alarm is no longer shelved you get an indication of that with this little uh clock so this indicates that we&#39;ve uh expired in terms of our shelving and then here&#39;s the individual alarms that are done being shelved and then the operator would need to drag them back into the alarm summary so let&#39;s talk about examples of how you might use shelving how you might define it on a project these are just examples to give you something to think about so for this particular scenario we want to provide continuous shelving functionality to the operator for low and medium priority alarms so we&#39;ve decided that high and critical priority alarms cannot be shelved but the operator will have access to shelve low and medium priority for different durations so depending upon what they think is most appropriate they could put the alarm in any one of these different shelves and as i mentioned it&#39;s designed in this case to have the operator be able to do it so they have the authorization now we have another shelf which is uh 24 hours so the maximum amount of time and in this case it&#39;s a it&#39;s a one shot shelving so be good for uh stale alarms and in this case it&#39;s set up so that only the supervisory access or maintenance can use it so if we thought we wanted to make it so that operators can&#39;t shelve an alarm for 24 hours um if you&#39;re not comfortable with that then you could set it up this way where somebody with higher authority has the ability to do that so this is one example of how we can do it another example in this case we&#39;re providing the option for the operator to shelve an alarm for 30 minutes maximum and they get to pick whether it&#39;s continuous shelving or one shot so you&#39;d have to train them on the difference and when to use each but you see they have the authorization to be able to shelve it and then we have other different shelves set up where the alarm will come off the shelf at the end of the shift so we&#39;ve not specified a time duration but actually a specific time for it to become unshelved and in this case we&#39;ve also set it up as an example where only the supervisor has access to put the alarm in that shelf so hopefully this gives you some ideas on the granularity that you have in defining shelving to make it a useful safe and worthwhile uh function in your system now to make sure that shelving is being used appropriately we do want to review every so often performance reports here&#39;s this is an example of the exaquantum ar ara showing us some statistics about suppressed alarms so that we can look at it and make sure that alarms aren&#39;t being just continually suppressed without fixing the underlying situation because that&#39;s really what we want to have happen shelving is a way to remove a nuisance alarm from the operator&#39;s view but we want to still make sure that that alarm gets addressed and that that portion of the process is not forgotten so for more information there was actually a ebook that was created by yokogawa on alarm shelving that goes through a lot of the information that we just talked about i had the privilege of helping to create that so if you&#39;re interested in information about it or how to get a copy of it shoot me an email we also did a webinar a little while back that was on the topic of shelving and how shelving works on different control systems not just yokogawa so if you&#39;re interested in learning more about shelving in particular i&#39;d encourage you to check out that recorded webinar so now let&#39;s shift gears and talk about advanced alarming and alarm suppression so the cam system supports eclipsing which is a function that&#39;s designed to help automatically hide one alarm condition from a tag when it&#39;s not as important as another alarm condition so what do i mean by that so let&#39;s say you have both a high and a high high alarm active at the same time well probably one of them is more important than the other so you could hide one of them and all the operator needs to see or to know about is that the more important one is now active so would that be the high high alarm that&#39;s typically the way eclipsing works but you would actually want it to be perhaps the highest priority alarm should be the one that stays and the lower priority one gets eclipsed i&#39;m not exactly sure how that function would work so if you were going to try and use eclipsing you&#39;d have to make sure that it really works in the way that you want because there may be situations depending upon how you define alarms and do rationalization where your high alarm is a higher priority than your high high in which case you&#39;d want to make sure that eclipsing reflects that as well and doesn&#39;t end up kind of reversing what you&#39;re trying to do there so now let&#39;s talk about some design suppression of groupings of alarms so in cell arm we can define what&#39;s called static suppression which basically means there&#39;s a group of alarms that we need to suppress because we are there&#39;s a a planned transition into that mode so this is for scenarios where you purposely don&#39;t use a piece of equipment because it&#39;s not needed you purposely are taking a piece of equipment offline to do maintenance so it&#39;s a plan transition so you define which alarms you want to suppress and then that would get implemented in your control system now how do we implement static suppression in a yokogawa system well there&#39;s a couple of different ways you could do within the control logic and face plates custom face plates in cams that&#39;s one approach or you could use the exa pilot software which has the ability it&#39;s you can see this interface kind of designed specifically for static suppression of a group of alarms like all of the alarms that would be part of a reactor unit or whatever so it is designed to be able to handle that case the information that you see here actually maps to the sill alarm definition of static suppression as well so we could take the information out of sill alarm either as a report or as an excel spreadsheet and load it into the exopilot software to hopefully implement the static suppression there in addition then the second type of suppression we&#39;re going to look at is dynamic suppression the difference here is that we are planning for an unplanned mood transition as if a piece of equipment tripped unexpectedly so for example a compressor trip because the way we would activate the suppression here needs to work differently than the static suppression where we plan to transition into that mode so in this case within so alarm we&#39;re going to define what the suppression module is used for set up some state detection logic and define the alarms that should be suppressed so here&#39;s an example of how we define the state detection logic how we indicate which alarm should be suppressed including display of whether the alarm is being used as a safeguard or an ipl along with its classification and its priority so that you can make sure that it&#39;s okay that it&#39;s safe to suppress the alarm so we define all this in sole arm and then we can extract that in a report like you see here that can then be used for configuration within the control system and in this case in the yokogawa system so again if we wanted to do suppression in the control system itself in the control logic and in the cams hmi this would be an example of how that could be done it can also then use the exapilot tool as an alternative option where again we have this infrastructure for suppression of a group of alarms where we&#39;ve defined what alarms are in that group we&#39;ve defined what the transition scenarios are the maximum suppression time all again information that could be defined in soul arm so there&#39;s a couple other ways that design suppression can occur in a cam system here we can define a a suppression grouping within cams itself to automatically suppress a set of alarms there&#39;s also a function called load shedding which is i guess this would be the automatic alarm shelving which essentially says if the system detects a flood of alarms like more than 10 alarms in 10 seconds or whatever then it&#39;s gonna evoke a filter where it would for suppress or hide the low and medium priority alarms showing then only high priority alarms to help the operator focus on what are the most critical alarms at that point in time so it could be potentially useful to help the operator respond to abnormal situations so now we&#39;re going to talk about how to evaluate performance there are key performance indicators and metrics defined in the isa 18.2 and iec standards related to overall alarm rate and then certain categories of nuisance alarms and evaluation of how well the system is configured so primarily what we&#39;re showing here is the exaquantum ara tool with its dashboard to show you the alarm performance over time reports to show you for each operating position the alarm enunciation rate as well as the priority distribution judged against the amua 191 categories or criteria an evaluation again of how well the alarm system is performing for the different operating positions identification of alarm floods along with how long the flood lasted for and how many alarms resulted during that flood identification of chattering alarms alarms that are nuisance alarms that would benefit from being reviewed during rationalization and and updated to eliminate that chattering behavior stale alarms another type of nuisance alarm and bad actors so these last couple of reports give you some good input for performance improvement as part of the continuous improvement methodology that i mentioned earlier they would identify specific alarms that are performing poorly that could really benefit from rationalization it&#39;ll also automatically calculate your enunciated priority distribution so you can check to make sure that you don&#39;t have too high of a percentage of critical and high priority alarms because that would compromise the operator&#39;s ability to understand the relative importance of one alarm versus another and one of the advanced capabilities of this software is to be able to analyze alarms alarm history and look for correlation look for redundant alarms what i mean here is by looking at every time this alarm occurs there&#39;s a pattern that this other alarm occurs shortly thereafter so it&#39;ll go through and analyze that and do the calculations and tell you the correlation ratio so 100 percent means hey every time this one alarm occurs we get this other one as well well that&#39;s a a mini flood if you want to say it that way so going through rationalization we would look at these alarms and say we don&#39;t need all of them to indicate this one problem so can we eliminate some of them to reduce the alarm load on the operator do we really need all these so what we&#39;re finding is that one singular event occurs and it leads to a whole grouping of alarms a whole pattern of alarms that kind of go off together so this does the analysis of that for you and tells you what alarms have that correlation that you could go through rationalization and fix and then the last thing that we want to look at is auditing and enforcement and what we mean here is comparison of the master alarm database versus what&#39;s actually running in the control system so this is what so alarm can do so from a updated import we can compare to what&#39;s in the master alarm database already and look for changes identify for each alarm what are the different parameters that have changed what were they originally what&#39;s the proposed change or what&#39;s the different value and then it allows you to go through and look at each change and disposition it as accept reject or enforce it also allows you to do a even more formal audit where you assign someone to go do that review you document what the what you found as part of that audit review and then that gets entered into the um report that gets created with these audit results so this is what sill alarm can do from an auditing and enforcement point of view if you&#39;re using amd as your master alarm database and that can be loaded from silarm if you&#39;re using amd then you can use that to detect differences and it&#39;ll give you a report that shows you what&#39;s different it can also be used to do enforcement where it&#39;s automatically resetting the value in the control system to what you want it to be coming out of rationalization or has that set up as a manual process where you review what&#39;s desired to be enforced and agree to it so very rigorous auditing and enforcement protocol so now kind of getting to the end and summarizing what we&#39;ve talked about for a yokogawa system to get the most out of your alarming capability you definitely want to start with the cam software and look at its capabilities which are quite significant and evaluate the different functionality and decide what portions of it you want to use i would definitely encourage you if you if you&#39;ve not for example if you&#39;ve not implemented shelving to take a look and look into the the cams manuals and understand more about what you can do with it because it is really powerful and certainly if you have an issue with nuisance alarms or operators ignoring alarms you may find that adding shelving capability helps them to be able to manage some of those alarm situations on their own i&#39;d also encourage you to consider either doing alarm rationalization and then creating alarm response procedures from it potentially in cams as alarm help or if you&#39;ve already done rationalization to see whether you can load that information into cams on your own to create that reference information for the operator to see for each alarm what should they do when it occurs and of course if you&#39;re using so alarm as the rationalization tool a master alarm database then as part of the import into cams that information would be built there are a lot of capabilities within the yokoga kawa family of products for alarm suppression many of them built into cams itself some of them provided by auxiliary software like exopilot so you&#39;d need to evaluate what is best for your particular situation and we talked about some of the reports that can be created from the ara software the exaquantum ara software and certainly performance monitoring is a key part of a continuous improvement process it&#39;s really what drives the activities of alarm rationalization and improvement is getting some good reports that allow you to see how your performance is changing over time and helps identify those alarms that are performing badly that need to be fixed the most and go through the rationalization process so on that note we are going to wrap up and i&#39;m going to say thank you very much for attending and also mention some of the upcoming training classes that we have going on our flagship alarm management class the alarm fundamentals for the practitioner we have a online version of that class an online live version of that class coming up the end of april it&#39;s three half days and when you&#39;re done you have the option to sit for the certificate exam to demonstrate what you&#39;ve learned from the class we also have other courses available coming up not just for alarm management but also for functional safety and cyber security and then a lot of our courses are available online on demand so allowing you to take them whenever you want it could be two o&#39;clock in the morning could be five o&#39;clock in the morning whatever time makes best sense within your schedule you can take those courses and gain that knowledge at your convenience so i hope that you have enjoyed this webinar and learned something from it if you have and you&#39;re interested in other alarm management topics or webinars you can look at the exeter website it&#39;ll take you to the the list of what&#39;s been created and what&#39;s available or you can go to exeter&#39;s youtube channel where there are different collections created video collections that have the webinars from past presentations in different areas so like alarm management functional safety cyber security for alarm management for example there&#39;s quite the collection i think there&#39;s 32 videos there now so good way to continue to enhance your knowledge of alarm management and if you have any questions or comments please feel free to shoot me an email also always interested in ideas on new topics for webinars and areas that people would like to learn more about when it comes to alarm management so let&#39;s just do a quick look at the questions that we have we are over time so i think we&#39;re gonna defer um answering these questions live and what we&#39;ll do is we&#39;ll take a look at them offline and then um get back to you with the answers so maybe we&#39;ll just keep it open for another minute here so that if you have a question that you want to register for us to look at and respond to you can enter that into the question box so i don&#39;t see any more questions being added so with that we&#39;ll say again thank you very much for your attendance and have a good rest of the day and i hope you uh will come back and listen to other future webinars from exeter

welcome to the webinar alarm management getting the most out of your yokogawa control system my name is todd stoffer and i will be presenting to you so if you have questions along the way please enter them into the questions box and we&amp;#39;ll take a look at them at the end and if we have some time to review them and answer them live we will otherwise we&amp;#39;ll get back to you afterwards so as i mentioned my name is todd stauffer the director of alarm management here at exeter which means that&amp;#39;s what i focus on spend pretty much most all of my time engaged in the activities of alarm alarm management a lot of them follow from the isa 18.2 standard which i&amp;#39;ve had the privilege of helping to write and shepherd through the process since 2005 i also wrote or participated in writing a couple of the isa technical reports and my latest role is as a co-chair for a new standard on safety controls alarms and interlocks bringing together the worlds of functional safety and alarm management so if this is your first time thank you for joining a little bit of background information about exeter we were formed a little over 20 years ago by two gentlemen that wanted to make the world a safer place they actually left their jobs to start a new company called exeter that would focus on functional safety and then several years later they expanded the scope of the company to also include cyber security and alarm management so other areas that affect risk in a process plant exeter is a leader in the certification of products and systems and people by products i mean safety devices so if you&amp;#39;ve ever put together a safety instrumented system and worked with sill rated equipment chances are it was exeter that did those sill verifications or validations also have a program for certification of people so that they can acquire the knowledge that they need and demonstrate the competence that they need to perform a functional safety lifecycle tasks and speaking of the functional safety life cycle we have a suite of tools called excelentia that actually provides safety system design and engineering capabilities and guides you through the whole entire functional safety life cycle one module in that suite is called sill alarm which is an alarm rationalization tool and that&amp;#39;s something that we&amp;#39;ll talk about today so in terms of the agenda we&amp;#39;re going to look at alarm management functionality things like prioritization how do we do rationalization for a yokogawa system etc alarm shelving we&amp;#39;re also going to look at some of the key tools from yokogawa and others that would help deliver this alarm management functionality so as our guiding light we use the standards and guidelines that you see here a mua 191 first created in 1999 and actually proposed the concept of alarm shelving which we&amp;#39;ll talk about quite a bit and then the isa 18.2 and iec 62682 alarm management standards which are pretty much identical to each other the isa 18.2 standard was first published in 2009 was updated in 2016 and the iec standard was first published in 2014 and is in the process of being updated right now hopefully for release later this year so both standards i say and iec define the concept of an alarm management life cycle which is basically the workflow process to follow for alarm management to create a sustainable program over time so let&amp;#39;s look at how some of those tools that we mentioned earlier fit into this life cycle so for alarm rationalization which we&amp;#39;re going to talk about there are a couple of different tools that can be used we&amp;#39;re going to focus on using the sew alarm tool and the exit quantum amd tool from yokogawa the cams software also has some capability for rationalization as well it really comes into play though when we look at the presentation of alarm information to the operators as well as invoking alarm suppression functions and advanced alarm functionality so we&amp;#39;ll talk about what this cams software can provide from that point of view there&amp;#39;s another tool from yokogawa called exopilot that we&amp;#39;ll look at how that can be brought into the picture to achieve some advanced alarming for the monitoring and assessment stage of the life cycle you can use either exa quantum ara again from yokogawa exa plug another product from yokogawa where of course there there&amp;#39;s other third-party pieces of software as well and then from a management of change point of view you can use the sill alarm tool the exaquantum amd tool and cams as well so those are the key tools that we&amp;#39;re going to look at let&amp;#39;s just go through them relatively quickly to kind of define what they do so first and probably most important for a yokogawa system is the cams software cam stands for consolidated alarm management system so this is a piece of software that sits on top of the dcs within the hmi and actually controls the presentation of alarm and message information to the operator so it actually takes the alarm information that comes up from the controllers and presents or adds additional information to it in some cases actually enhancing it for presentation to the operator this software also provides the capability to do advanced alarming things like alarm shelving alarm suppression significant filtering is built in much of which we&amp;#39;re going to spend some time talking about so that&amp;#39;s the cams software exaquantum ara is a tool for performance monitoring and assessment you can see some of the screen snaps there and these are some of the standard canned reports or capabilities that it has for generating performance reports also in the xaquanum family is a tool called amd which is something that acts as a master alarm database so you load your rationalized alarm configuration into this database and then you can use it for management of change auditing enforcement etc to help maintain the system integrity of your alarm system then there&amp;#39;s the exa plug software which is for event analysis so it can analyze alarms and present reports of what&amp;#39;s happening that you can use to drive your alarm management program it also does analysis of operator actions and events that occur so you can see correlations between those and alarm event activity it&amp;#39;s a it&amp;#39;s a lower budget tool compared to the exaqu so depending upon what your price point versus functionality needs are this may be something that&amp;#39;s useful to go for or the the ara provides i think more capability and is a slightly higher cost one of the other tools that we&amp;#39;ll talk about is the exapilot software which is used for procedural automation so it basically provides a way for operators to create their their own semi-automatic sequences instead of doing things via manual operation it also can be used for some advanced alarming capability suppression of groups of alarms we&amp;#39;re going to show how it can be used for static suppression and dynamic suppression and then last but not least is the sill alarm tool from exeter which will guide us through the rationalization process it&amp;#39;ll also act as the master alarm database it can define alarm response procedures it can do advanced alarming design and also has capabilities for doing auditing and enforcement as well so one of the things that you&amp;#39;ll see or probably think about is we have different tools here that kind of do the same thing and a little bit overlap certainly that&amp;#39;s the good news and bad news good news is you have different options to choose from bad news is you have to figure out what&amp;#39;s the best fit for your situation now in terms of how we&amp;#39;re going to talk about alarm management we typically follow a seven-step process that looks like this that starts with some initial benchmarking and the creation of an alarm philosophy document and then you get into continuous improvement mode and that&amp;#39;s what we&amp;#39;re going to spend most of our time talking about today is how the the tools and functionality help us continuously improve our yokogawa system to make the alarm system performance acceptable so we&amp;#39;ll go through the rationalization phase we&amp;#39;ll talk about advanced alarm design we&amp;#39;ll talk about implementation and building of alarm response procedures and then look at the capabilities for performance monitoring and assessment that is kind of the backbone of a continuous improvement loop here we&amp;#39;ll also then look at some of the audit capabilities for comparing the alarm system configuration that&amp;#39;s actually running in the dcs versus what you expect or want coming out of rationalization so a little bit of architecture overview you can see here that alarms are actually generated from function blocks that exist in the controllers they that information gets propagated up through cams gets processed further in cams additional information can be added some information can be changed for example if you were just to configure the priority of alarms in these function blocks you would find that every alarm from a given tag has the same priority it&amp;#39;s not the best approach nowadays that we know that we want to analyze each alarm condition for its uniqueness and understanding what it&amp;#39;s trying to prevent so in most cases it makes sense to provide the flexibility to have alarms that can have different priorities for each alarm condition so that gets done by cams up here before it&amp;#39;s presented to the operator so the yokogawa control system natively provides high medium and low as process alarm priorities when the information goes through cams it maps into high medium and low but then it adds a fourth alarm priority called critical that you can use for your extra special most important alarms now we want to talk about how to prioritize alarms what are options how can we do that for a yokogawa system so priority is pretty important that&amp;#39;s what operators should be able to use to determine which alarm do they respond to first so when you have multiple alarms present at the same time the idea is priority guides them to the one that is more important from a business point of view and the way we typically evaluate or assign priority is based on evaluating how bad the consequences would be if the operator didn&amp;#39;t respond and how much time is available to respond combination of those two dictates the priority so here is a typical example of the way the majority of of people that i&amp;#39;ve run into prioritize alarms you have a top portion which allows you to assess how bad the consequences would be this typically comes from a risk matrix used for a pha or a has up and then down below here we&amp;#39;re factoring in the operator response time the urgency for them to respond and it&amp;#39;s the combination of the two really significant consequence short response time that dictates the priority of the alarm this is the most common method of prioritization now cams does support some rationalization capabilities and some and some prioritization capabilities what it supports is actually prioritization based on the purpose of the alarm meaning safety environmental or financial and the worst case consequence is it large very large medium etc so you can see here based on that combination you would end up with a priority it doesn&amp;#39;t include or factor in the response time though so if you have a matrix like we just looked at which a lot of people do you would not really be able to use this prioritization wizard with that matrix that you&amp;#39;ve defined so that&amp;#39;s one of the reasons why we&amp;#39;re going to assume that we use sill alarm as the rationalization tool and master alarm database for this discussion so here&amp;#39;s an example of how we would set up so alarm to define our prioritization scheme based on the information that would be in your alarm philosophy defining the assignment of priorities based on the combination of how bad the consequence is and the response time so now let&amp;#39;s look at the alarm rationalization process so to start things off we need to load the alarm configuration into our rationalization tool to essentially create our master alarm database we can do that for an existing system by extracting from cams or we could do it for a greenfield system from an excel spreadsheet so cams actually has a very nice export and import interface so in this case we can select which pieces of alarm configuration we want to export and then we can give it a name in the export file choosing names that line up with the terminology that we have in so alarm so it makes it very convenient for us to do the mapping to be able to take and export out of cams and be able to load it right into sill alarm and once that&amp;#39;s done you can see here we&amp;#39;ve got all of the alarms from one typical pid loop within the cell alarm database now we can proceed with the rationalization the first step is typically to evaluate the validity of the alarm and the alarm objective analysis by looking at what is the consequence that would occur if the operator doesn&amp;#39;t respond what are the likely causes what are the corrective actions now you might think who who&amp;#39;s able to define that information who knows that well rationalization really requires somebody who is very familiar with the process to provide this information and that typically resides in the brains of your senior operators your most experienced operators so they are key to a successful rationalization it also means that rationalization becomes not just an activity to improve your alarm system performance but it can also become a knowledge capture process because we can essentially document what your most senior experienced operator would do when that alarm occurs and then we can make that information available to all of the operators hopefully through the hmi in an alarm help document either a document like you see here which is a report coming out of sill alarm or as we&amp;#39;ll see in a little bit within the cams interface but a key thing coming out of rationalization is to be able to extract that great information that you&amp;#39;ve captured from your senior operators and make it available so every operator can see that when they get an alarm they can look and see what that senior operator would do if we&amp;#39;ve once we&amp;#39;ve defined the consequence now we&amp;#39;re ready to prioritize that alarm so here we have it predefined and so alarm to match up with your priority matrix as defined in your philosophy so we&amp;#39;ve got the imported priority essentially what the priority in the control system is originally and now we do some consequence ranking we evaluate how bad the consequence would be in each of our impact areas and we record and document that and then we also estimate how much time does the operator have to respond and based on this information it pulls out of the priority matrix what the recommended alarm priority should be and it maintains all this information it records it so we&amp;#39;re not just recording the worst case consequence we&amp;#39;re actually recording the consequence assessment in each of the impact areas so if something changes later on where you think that you may have made a misinterpretation or a mistake or you&amp;#39;ve got some new information you can go back and look at what was done to be able to update it accordingly a lot of tools or if you&amp;#39;re just using an excel spreadsheet to do this uh you may not fully document that rationale for setting the priority in which case it becomes kind of challenging to recreate it or modify it if you need to so we get the alarm priority coming out of there we can also override that priority and set it manually so we could set it based on the philosophy or we could set it to a non-alarm type notification like an alert or a prompt or message something that&amp;#39;s going to use the alarm system for presentation but that notification is not as important as an alarm now in cams as i mentioned there is some capability to do rationalization and to document the information here&amp;#39;s where we would define or go through the process of setting what the alarm priority would be by selecting a purpose in this case safety environmental financial and a consequence how bad is it and as i as we showed previously based on the combination of the purpose and the significance of consequence that would give you what the priority is no inclusion of response time though next we would want to define the alarm classification which is a way to group alarms that have common sets of requirements for things like testing or training or moc this allows you to show or keep your eyes on alarms that are more important or the most important alarms in your system so it could be alarms that are there for environmental purposes or to prevent environmental incident could be for safety purposes for personnel safety or alarms that are listed in a layer protection analysis those are good examples of classifications so during rationalization we want to identify what or assign what classification is appropriate for safety purposes and functional safety we want to go a step further and actually document the functional safety usage of each alarm so if it has a role as a safeguard in a hazop or as a protection layer in a layer protection analysis we want to document that so then it becomes easy to find in our entire database of alarms which ones are being used for functional safety purposes and then we also can create a cross-reference by documenting what lopa study it comes from or documenting what has upper pha the alarm was identified in this allows us to make sure we&amp;#39;re taking these alarms seriously and not making design changes or disabling alarms coming out of rationalization where those alarms are actually counted on to provide a safety risk reduction now in the yokogawa cam system alarm classification or alarm class is not defined or used out of the box but what we&amp;#39;ll see or what we&amp;#39;ll talk about quite a bit is that you can define your own fields in cams so we could set up a field to be called alarm class or alarm classification and then that could be used to present that information and to hold that information now another thing that i wanted to mention about silarm when we document the alarm limit we can document the consequence threshold what we&amp;#39;re trying to prevent from occurring we can document any process dynamic information that we know the rationale for setting the alarm limit and we can define our own user parameters that are unique to a certain entity or a certain project or certain site to document additional information that&amp;#39;s relevant to the design of the alarm so if we&amp;#39;re trying to set the alarm limit for a alarm let&amp;#39;s say for a tank level we want to know whether we&amp;#39;ve defined the safe operating limits where is there an operating envelope that we want to stay within that information can be documented in so alarm so we can define those as extra fields that we want to capture and put information in now part of the reason why i&amp;#39;m focusing on this is we can then push that into the cam system if we decide that some of this information is useful and you&amp;#39;d like to present it to the operators we can create these fields or any combination thereof define them in cams and then import this information coming out of the sill alarm rationalization process so the last part of rationalization that we&amp;#39;ll talk about is defining some advanced alarming and in this case the requirement to allow an alarm to be shelved or not we&amp;#39;ll talk a lot more about that shortly so when you&amp;#39;re all done rationalization this is the information that you should have documented so part of what we want to talk about now is how do we get this information into the cam system and how do we present it to the operator because a lot of this is very useful good information to help them know how to respond to each alarm so when we are done rationalization we can export the information out of so alarm and then use the reverse of the original process and import the information back into cams so if we&amp;#39;ve defined some additional fields and cams to bring in the extra rationalization information that we think is valuable we can make that process all work through this interface and the correspondence and the mapping of here&amp;#39;s what that field is called in cell arm and here&amp;#39;s the field that we want that information to be put in in cams so now let&amp;#39;s talk about how we can make the most of that alarm rationalization information and actually create alarm response procedures so an alarm response procedure you could think of that as a as a job aid for the operator what kind of information should it include here you can see what&amp;#39;s recommended so it&amp;#39;s information like what are the potential causes of the alarm what&amp;#39;s the consequence if you don&amp;#39;t respond what is the operator action and what we really mean here is the actual description of what the operator should do so consequence we&amp;#39;re not interested in is the consequence large small severe not severe we&amp;#39;re actually interested in the description of what the consequence would be loss of containment uh potential damage to equipment that kind of stuff so let&amp;#39;s look at how those alarm response procedures could be presented to an operator through a cam system there is a guidance tab provided for alarm messages which can be used to present information including the call up of a file like the file that we saw that could be created out of sill alarm so that is one possibility for providing alarm response or alarm help information to an operator but what we&amp;#39;re going to focus on is actually a different uh implementation method and that involves changing the information that we see here and filling it in with more information and more information that&amp;#39;s got useful details that the operator would benefit from so if we look at some of the fields here we have consequence we have time to respond we have purpose they sound pretty useful but if we look at the information that&amp;#39;s in there it&amp;#39;s really just from a drop down list so safety large urgent that&amp;#39;s not really going to provide much benefit to the operator in terms of actionable advice so what we want to do is add some fields and cams and load those fields with information coming from rationalization so and then we want to present them to the operator so we would replace this window with something that looks like this and it&amp;#39;s got some additional and different information in it so now we have the time available to respond and it&amp;#39;s actually a time measurement that will make sense to the operator we&amp;#39;ve got guidance or corrective action that actually tells them what they need to do we&amp;#39;ve got a consequence that tells them what piece of equipment or part of the process is going to be impacted so much more actionable intelligence it can also then include some of that other alarm database information so if we know that there&amp;#39;s a safe operating limit or there&amp;#39;s an operating performance boundary we can define that and show that within this window as well so a lot of different information that we can pull out of rationalization and then present to the operator here so as soon as the new alarm occurs the operator can click to this screen and then be able to see this information as part of their response process hopefully making the response quicker more consistent and more effective now let&amp;#39;s talk about alarm suppression so suppression is a way to present or prevent the alarm from being presented to the operator when the base alarm condition is present so basically means the operator doesn&amp;#39;t need to see it for one of many different reasons the first type of suppression is called design suppression that&amp;#39;s where the control system is able to determine when the suppression should occur automatically and then automatically suppress the alarms that should be suppressed of course you need to configure it to do that but once you&amp;#39;ve done that that will all happen behind the scenes and those alarms are under control of the logic the second type of suppression is called alarm shelving that&amp;#39;s where the operator can temporarily suppress an alarm let&amp;#39;s say an alarm that&amp;#39;s behaving like a nuisance alarm they can temporarily suppress it so it gets out of their view and it is not a distraction anymore so that&amp;#39;s what shelving is for it&amp;#39;s different from disabling the alarm the alarm will come back after a period of time and then the third type of suppression is called out of service this is where the alarm would be suppressed typically for maintenance of the underlying hardware now shelving is actually a required function in the isa and iec standards so it&amp;#39;s now required functionality of a dcs now there are different ways of suppressing alarms in a yokogawa system some that are done directly in the control system control logic some that are done through cams some that are done through the hmi we&amp;#39;re gonna focus primarily on the options for cams so now let&amp;#39;s talk about alarm shelving functionality as provided by cams and i call this one of the best kept secrets yokogawa is actually one of the first dcs suppliers to provide this functionality was actually introduced in 2006 so before isa 18.2 was even released so it was built in based on the information and guidelines in the amua document but it&amp;#39;s been been around for a while part of the reason why we&amp;#39;re going to go through this is there are many customers of yokogawa or that have yokugawa systems that are not aware of the capability of shelving or perhaps haven&amp;#39;t really dove into it to decide how they could take advantage of it so that&amp;#39;s part of what we want to talk about here so when you are trying to figure out how to use shelving you want to consider things like should we allow every alarm to be shelved or only alarms of certain priority or certain classification is there a limit for how long alarms should be able to be shelved is there an authorization process can the operators do it themselves do supervisors need to do it so there&amp;#39;s a lot of things to think about in order to make it an effective system for your operation and the way you choose to do it could be different from one site to the next one of the the ways this is defined then is by setting up different groupings so different alarm shelves so essentially we define different purposes for different shelves and then the operator when an alarm needs to be suppressed when it needs to be shelved they essentially drag and drop the alarm to whichever of these shelves they have access to and whichever one they think is most appropriate given the different characteristics of each one so we&amp;#39;re going to talk about how to set them up in a second now the maximum amount of time that an alarm can be shelved in a cam system is for 20 24 hours and one of the other key elements related to cams and shelving is to understand that there&amp;#39;s actually different forms of shelving that are supported one shot shelving continuous and automatic shelving we&amp;#39;re going to talk about what that means so when you define a shelve you declare whether it&amp;#39;s going to be a one-shot shelving type or a continuous shelving and those behave differently and kind of have different purposes so a one shot shelving means the alarm is suppressed and it will stay suppressed for the period of time that you select unless the alarm clears while it&amp;#39;s on the shelf then it will come back to be able to be triggered if that condition should arise so it doesn&amp;#39;t necessarily stay shelved for the whole entire time that you&amp;#39;ve selected it will unshelve earlier if the alarm clears now when would that be useful that would be very useful for stale alarms where the alarm is active for a long period of time but as soon as it clears you would like it to be back able to generate or to enunciate again so rather than waiting for the end of the shelving period if it clears you want it to go back to an enabled alarm ready to trigger if necessary now that&amp;#39;s different from the alarms that you would use continuous shelving for like a chattering alarm so continuous shelving the alarm stays on the shelf for the period of time that you have selected until the time is up or the operator has chosen to manually unshelve the alarm now why that&amp;#39;s important is you can imagine if it&amp;#39;s a chattering alarm if you had to reshelve it each time it came off each time it cleared then you&amp;#39;d be shelving the chattering alarm each time you acknowledged the alarm or each time you silenced the horn so you&amp;#39;d be adding another extra step another annoyance it would make the situation actually worse so that&amp;#39;s why this different type of shelving is ideal for a chattering alarm situation and then the final type that&amp;#39;s supported in cams is automatic shelving which can be triggered when you detect the scenario that would lead to an alarm flood or if you&amp;#39;ve got a piece of equipment that you&amp;#39;re gonna temporarily you want to temporarily suppress all of the alarms from that equipment those are both forms of automatic shelving that&amp;#39;s supported in cams so how do we actually configure it then so here&amp;#39;s where we build those different shelves or different groupings different parking lots for alarms to be placed so what we define for each shelf give it a name a description the type of shelving so is it a one shot continuous or automatic then we define the type of shelving so is it being specified for a certain time period like 30 minutes or eight hours or do you want it to expire at a specified time like the end of the shift that&amp;#39;s what that is really useful for so you can decide when you&amp;#39;re creating these shelves or defining them whether you want it to be for a fixed period of time or at a specific time to elapse then you set the maximum amount of shelving time and you also get to define who has access to the shelve to be able to put alarms on it to be able to shelve alarms and put them there so there could be certain shelves that are for the operator and then certain shelves that are maybe for a supervisor or for maintenance and that can be controlled by access so we want to define the different shelves and label them so that the operator understands what they&amp;#39;re used for because when they see a nuisance alarm that needs to be shelved they&amp;#39;re going to drag and drop the alarm to whichever shelf they think is most appropriate and for which they have access to now behind the scenes in the engineering side of cams you can define whether an alarm should be able to be shelved or not so each alarm has a separate enable disable parameter for shelving so they&amp;#39;re all set to defaults this information can be set externally from an excel spreadsheet or from so alarm for example so after you&amp;#39;re done rationalization depending upon what your rules are for alarm shelving could be based on the priority could be based on the classification you could set in the master alarm database whether you want shelving to be enabled or not based on priority and then when we export that information and import it back into cams it would update the shelving parameter for all of the alarms as appropriate we also get to define what information appears in the shell display information like when was the alarm shelved so the operator can look at that to know how close they&amp;#39;re getting to the time when the alarm will come off the shelf so then the way the operator actually interacts with it is we&amp;#39;ve got some new alarms down here if they are nuisance alarms or alarms that the operator feels should be shelved they drag and drop it to whichever of the appropriate shelves best meets their needs now when the alarm shelving time period has elapsed and the alarm is no longer shelved you get an indication of that with this little uh clock so this indicates that we&amp;#39;ve uh expired in terms of our shelving and then here&amp;#39;s the individual alarms that are done being shelved and then the operator would need to drag them back into the alarm summary so let&amp;#39;s talk about examples of how you might use shelving how you might define it on a project these are just examples to give you something to think about so for this particular scenario we want to provide continuous shelving functionality to the operator for low and medium priority alarms so we&amp;#39;ve decided that high and critical priority alarms cannot be shelved but the operator will have access to shelve low and medium priority for different durations so depending upon what they think is most appropriate they could put the alarm in any one of these different shelves and as i mentioned it&amp;#39;s designed in this case to have the operator be able to do it so they have the authorization now we have another shelf which is uh 24 hours so the maximum amount of time and in this case it&amp;#39;s a it&amp;#39;s a one shot shelving so be good for uh stale alarms and in this case it&amp;#39;s set up so that only the supervisory access or maintenance can use it so if we thought we wanted to make it so that operators can&amp;#39;t shelve an alarm for 24 hours um if you&amp;#39;re not comfortable with that then you could set it up this way where somebody with higher authority has the ability to do that so this is one example of how we can do it another example in this case we&amp;#39;re providing the option for the operator to shelve an alarm for 30 minutes maximum and they get to pick whether it&amp;#39;s continuous shelving or one shot so you&amp;#39;d have to train them on the difference and when to use each but you see they have the authorization to be able to shelve it and then we have other different shelves set up where the alarm will come off the shelf at the end of the shift so we&amp;#39;ve not specified a time duration but actually a specific time for it to become unshelved and in this case we&amp;#39;ve also set it up as an example where only the supervisor has access to put the alarm in that shelf so hopefully this gives you some ideas on the granularity that you have in defining shelving to make it a useful safe and worthwhile uh function in your system now to make sure that shelving is being used appropriately we do want to review every so often performance reports here&amp;#39;s this is an example of the exaquantum ar ara showing us some statistics about suppressed alarms so that we can look at it and make sure that alarms aren&amp;#39;t being just continually suppressed without fixing the underlying situation because that&amp;#39;s really what we want to have happen shelving is a way to remove a nuisance alarm from the operator&amp;#39;s view but we want to still make sure that that alarm gets addressed and that that portion of the process is not forgotten so for more information there was actually a ebook that was created by yokogawa on alarm shelving that goes through a lot of the information that we just talked about i had the privilege of helping to create that so if you&amp;#39;re interested in information about it or how to get a copy of it shoot me an email we also did a webinar a little while back that was on the topic of shelving and how shelving works on different control systems not just yokogawa so if you&amp;#39;re interested in learning more about shelving in particular i&amp;#39;d encourage you to check out that recorded webinar so now let&amp;#39;s shift gears and talk about advanced alarming and alarm suppression so the cam system supports eclipsing which is a function that&amp;#39;s designed to help automatically hide one alarm condition from a tag when it&amp;#39;s not as important as another alarm condition so what do i mean by that so let&amp;#39;s say you have both a high and a high high alarm active at the same time well probably one of them is more important than the other so you could hide one of them and all the operator needs to see or to know about is that the more important one is now active so would that be the high high alarm that&amp;#39;s typically the way eclipsing works but you would actually want it to be perhaps the highest priority alarm should be the one that stays and the lower priority one gets eclipsed i&amp;#39;m not exactly sure how that function would work so if you were going to try and use eclipsing you&amp;#39;d have to make sure that it really works in the way that you want because there may be situations depending upon how you define alarms and do rationalization where your high alarm is a higher priority than your high high in which case you&amp;#39;d want to make sure that eclipsing reflects that as well and doesn&amp;#39;t end up kind of reversing what you&amp;#39;re trying to do there so now let&amp;#39;s talk about some design suppression of groupings of alarms so in cell arm we can define what&amp;#39;s called static suppression which basically means there&amp;#39;s a group of alarms that we need to suppress because we are there&amp;#39;s a a planned transition into that mode so this is for scenarios where you purposely don&amp;#39;t use a piece of equipment because it&amp;#39;s not needed you purposely are taking a piece of equipment offline to do maintenance so it&amp;#39;s a plan transition so you define which alarms you want to suppress and then that would get implemented in your control system now how do we implement static suppression in a yokogawa system well there&amp;#39;s a couple of different ways you could do within the control logic and face plates custom face plates in cams that&amp;#39;s one approach or you could use the exa pilot software which has the ability it&amp;#39;s you can see this interface kind of designed specifically for static suppression of a group of alarms like all of the alarms that would be part of a reactor unit or whatever so it is designed to be able to handle that case the information that you see here actually maps to the sill alarm definition of static suppression as well so we could take the information out of sill alarm either as a report or as an excel spreadsheet and load it into the exopilot software to hopefully implement the static suppression there in addition then the second type of suppression we&amp;#39;re going to look at is dynamic suppression the difference here is that we are planning for an unplanned mood transition as if a piece of equipment tripped unexpectedly so for example a compressor trip because the way we would activate the suppression here needs to work differently than the static suppression where we plan to transition into that mode so in this case within so alarm we&amp;#39;re going to define what the suppression module is used for set up some state detection logic and define the alarms that should be suppressed so here&amp;#39;s an example of how we define the state detection logic how we indicate which alarm should be suppressed including display of whether the alarm is being used as a safeguard or an ipl along with its classification and its priority so that you can make sure that it&amp;#39;s okay that it&amp;#39;s safe to suppress the alarm so we define all this in sole arm and then we can extract that in a report like you see here that can then be used for configuration within the control system and in this case in the yokogawa system so again if we wanted to do suppression in the control system itself in the control logic and in the cams hmi this would be an example of how that could be done it can also then use the exapilot tool as an alternative option where again we have this infrastructure for suppression of a group of alarms where we&amp;#39;ve defined what alarms are in that group we&amp;#39;ve defined what the transition scenarios are the maximum suppression time all again information that could be defined in soul arm so there&amp;#39;s a couple other ways that design suppression can occur in a cam system here we can define a a suppression grouping within cams itself to automatically suppress a set of alarms there&amp;#39;s also a function called load shedding which is i guess this would be the automatic alarm shelving which essentially says if the system detects a flood of alarms like more than 10 alarms in 10 seconds or whatever then it&amp;#39;s gonna evoke a filter where it would for suppress or hide the low and medium priority alarms showing then only high priority alarms to help the operator focus on what are the most critical alarms at that point in time so it could be potentially useful to help the operator respond to abnormal situations so now we&amp;#39;re going to talk about how to evaluate performance there are key performance indicators and metrics defined in the isa 18.2 and iec standards related to overall alarm rate and then certain categories of nuisance alarms and evaluation of how well the system is configured so primarily what we&amp;#39;re showing here is the exaquantum ara tool with its dashboard to show you the alarm performance over time reports to show you for each operating position the alarm enunciation rate as well as the priority distribution judged against the amua 191 categories or criteria an evaluation again of how well the alarm system is performing for the different operating positions identification of alarm floods along with how long the flood lasted for and how many alarms resulted during that flood identification of chattering alarms alarms that are nuisance alarms that would benefit from being reviewed during rationalization and and updated to eliminate that chattering behavior stale alarms another type of nuisance alarm and bad actors so these last couple of reports give you some good input for performance improvement as part of the continuous improvement methodology that i mentioned earlier they would identify specific alarms that are performing poorly that could really benefit from rationalization it&amp;#39;ll also automatically calculate your enunciated priority distribution so you can check to make sure that you don&amp;#39;t have too high of a percentage of critical and high priority alarms because that would compromise the operator&amp;#39;s ability to understand the relative importance of one alarm versus another and one of the advanced capabilities of this software is to be able to analyze alarms alarm history and look for correlation look for redundant alarms what i mean here is by looking at every time this alarm occurs there&amp;#39;s a pattern that this other alarm occurs shortly thereafter so it&amp;#39;ll go through and analyze that and do the calculations and tell you the correlation ratio so 100 percent means hey every time this one alarm occurs we get this other one as well well that&amp;#39;s a a mini flood if you want to say it that way so going through rationalization we would look at these alarms and say we don&amp;#39;t need all of them to indicate this one problem so can we eliminate some of them to reduce the alarm load on the operator do we really need all these so what we&amp;#39;re finding is that one singular event occurs and it leads to a whole grouping of alarms a whole pattern of alarms that kind of go off together so this does the analysis of that for you and tells you what alarms have that correlation that you could go through rationalization and fix and then the last thing that we want to look at is auditing and enforcement and what we mean here is comparison of the master alarm database versus what&amp;#39;s actually running in the control system so this is what so alarm can do so from a updated import we can compare to what&amp;#39;s in the master alarm database already and look for changes identify for each alarm what are the different parameters that have changed what were they originally what&amp;#39;s the proposed change or what&amp;#39;s the different value and then it allows you to go through and look at each change and disposition it as accept reject or enforce it also allows you to do a even more formal audit where you assign someone to go do that review you document what the what you found as part of that audit review and then that gets entered into the um report that gets created with these audit results so this is what sill alarm can do from an auditing and enforcement point of view if you&amp;#39;re using amd as your master alarm database and that can be loaded from silarm if you&amp;#39;re using amd then you can use that to detect differences and it&amp;#39;ll give you a report that shows you what&amp;#39;s different it can also be used to do enforcement where it&amp;#39;s automatically resetting the value in the control system to what you want it to be coming out of rationalization or has that set up as a manual process where you review what&amp;#39;s desired to be enforced and agree to it so very rigorous auditing and enforcement protocol so now kind of getting to the end and summarizing what we&amp;#39;ve talked about for a yokogawa system to get the most out of your alarming capability you definitely want to start with the cam software and look at its capabilities which are quite significant and evaluate the different functionality and decide what portions of it you want to use i would definitely encourage you if you if you&amp;#39;ve not for example if you&amp;#39;ve not implemented shelving to take a look and look into the the cams manuals and understand more about what you can do with it because it is really powerful and certainly if you have an issue with nuisance alarms or operators ignoring alarms you may find that adding shelving capability helps them to be able to manage some of those alarm situations on their own i&amp;#39;d also encourage you to consider either doing alarm rationalization and then creating alarm response procedures from it potentially in cams as alarm help or if you&amp;#39;ve already done rationalization to see whether you can load that information into cams on your own to create that reference information for the operator to see for each alarm what should they do when it occurs and of course if you&amp;#39;re using so alarm as the rationalization tool a master alarm database then as part of the import into cams that information would be built there are a lot of capabilities within the yokoga kawa family of products for alarm suppression many of them built into cams itself some of them provided by auxiliary software like exopilot so you&amp;#39;d need to evaluate what is best for your particular situation and we talked about some of the reports that can be created from the ara software the exaquantum ara software and certainly performance monitoring is a key part of a continuous improvement process it&amp;#39;s really what drives the activities of alarm rationalization and improvement is getting some good reports that allow you to see how your performance is changing over time and helps identify those alarms that are performing badly that need to be fixed the most and go through the rationalization process so on that note we are going to wrap up and i&amp;#39;m going to say thank you very much for attending and also mention some of the upcoming training classes that we have going on our flagship alarm management class the alarm fundamentals for the practitioner we have a online version of that class an online live version of that class coming up the end of april it&amp;#39;s three half days and when you&amp;#39;re done you have the option to sit for the certificate exam to demonstrate what you&amp;#39;ve learned from the class we also have other courses available coming up not just for alarm management but also for functional safety and cyber security and then a lot of our courses are available online on demand so allowing you to take them whenever you want it could be two o&amp;#39;clock in the morning could be five o&amp;#39;clock in the morning whatever time makes best sense within your schedule you can take those courses and gain that knowledge at your convenience so i hope that you have enjoyed this webinar and learned something from it if you have and you&amp;#39;re interested in other alarm management topics or webinars you can look at the exeter website it&amp;#39;ll take you to the the list of what&amp;#39;s been created and what&amp;#39;s available or you can go to exeter&amp;#39;s youtube channel where there are different collections created video collections that have the webinars from past presentations in different areas so like alarm management functional safety cyber security for alarm management for example there&amp;#39;s quite the collection i think there&amp;#39;s 32 videos there now so good way to continue to enhance your knowledge of alarm management and if you have any questions or comments please feel free to shoot me an email also always interested in ideas on new topics for webinars and areas that people would like to learn more about when it comes to alarm management so let&amp;#39;s just do a quick look at the questions that we have we are over time so i think we&amp;#39;re gonna defer um answering these questions live and what we&amp;#39;ll do is we&amp;#39;ll take a look at them offline and then um get back to you with the answers so maybe we&amp;#39;ll just keep it open for another minute here so that if you have a question that you want to register for us to look at and respond to you can enter that into the question box so i don&amp;#39;t see any more questions being added so with that we&amp;#39;ll say again thank you very much for your attendance and have a good rest of the day and i hope you uh will come back and listen to other future webinars from exeter

Transcript for:Webinar: Alarm Management - Yokogawa Control System

Transcript for:
Webinar: Alarm Management - Yokogawa Control System