Streamline your document signing process with airSlate SignNow's pipeline SCADA alarm management for Marketing

How to create outlook signature

hello everyone welcome to this webinar today um so today we are going to talk about uh how some of the ideas in data science field or data analytics field and how we can actually apply them in the alarm management so before i start everything i do need to stress that this particular session of the webinar is also recorded for future uses and today the day is july the 27th of 2022 so um if you realize that that's not the right date that means that you actually listening to a recorded session and the reason why i bring this up is because um i do sometimes receive emails or questions that seems a little bit frustrated that i didn't answer the questions during the question session during the session um and mainly because i wasn't actually uh live in that session that session was recorded so really appreciate you guys if you could send me emails when you have any questions i'll try to answer as much as i can during the during the whole webinar session but if this recorded there's really nothing much i can do other than actually recently receiving questions from your emails so so um a little bit back a little bit of background about today's topic um i i don't know how you guys are familiar with the idea of alarm management so one of the core idea i call i would say the core portion of alarm management is actually about managing alarm data's alarm attribute and also the isa standard and also the iec center mandates you to upkeep a particular alarm database and there's not much of a guidance impact in the isa standard when it comes to how exactly we can we can try to navigate and maneuver around how exactly we can upkeep that particular database um so therefore while studying uh for some for some topics in data science and there's analytics i realized that we can actually basically apply some of the philosophies within the data science or fundamentals within data science and apply that in the alarm management and it interestingly they actually fit very very well in the context of alarm management and this is going to be one of the series of webinars i plan to do to focus on how exactly we can bridge the discipline of data science and alarm management for the for your alarm system for your control system and today we will mainly be focusing on the root cause of alarms so root cause is actually also a very big topic in data science mostly whenever we talk about data science we are catering to businesses so the root cause would mostly be about because of what kind of financial problem or marketing problem that the business is right now experiencing um but the root cause whenever we talk about root cause in the law management we are actually talking about the cause of the alarms and specifically what actually started the propagation of the process variables and in the end resulting in the alarms and today we're going to look at one of the methodology and also how the data scientists or data analytic analysts handle them using the philosophy of five whites in order to basically derive out and also uh document the root cause of everything so um a little bit about how exactly you can ask questions so you can type out your questions um within the go to meeting tool there is a question tool at the right side of the screen so you can use this mechanism to type up any questions but during the presentations um i usually won't be able to look at questions during the session so what i would do is at the very end of the webinar we have a q a session and i will look at the questions and answer as much as i can um and alternatively as um before i started i i told you guys you can also send me emails uh to ask questions as well and i will give you my email at the end of the webinar all right let's start digging into it so a little bit of introduction about me myself and also about exeter so i'm paul my name is paul chan and i'm i'm actually a transfer from the hong kong office of exeter so the hong kong office is part of the exeter asia pacific office so i was transferred about four years ago to the headquarter here in the us i started with safety with hands-on and also selection workshop i was scribing for etcetera in one of those hands-off and still selection workshop in seoul in south korea for three months and later they realized that um i could actually be an engineer so that they basically hired me because as you can see my background was it was not actually in engineering but instead i was raised a biochemist and then later actually i had a master of art in english because i was hoping to teach english but you never know what fate has in store for you and right now i joined exeter and i'm also a certified safety professional and i also take part in a lot of the risk analysis like seal verification workshops um verification the calculations some of the a lot of the different alarm management projects and also a cyber security side of it as well although i'm more involved in terms of the cyber security audits for the time being instead of a lot of the risk and assessment in the context of cyber security which i do hope that one day i'll be getting more into so that's about me um an exeter we were actually founded in um just before 2000 um and you as you can see from this map this world map we have a lot of different offices scattered around um in the world but we are big in terms of we have a lot of experts around the world and we have this focus on functional safety and our management management and also cyber security and we're also proud to say that we are very customer focused uh company why would we say that because our work are evolved around customers we have our own enterprise tools pixelentia and also other modules that goes into its lecture my own software we also have a certification and assessment services to help you to have compliance with isa iec standards and also we do consulting services like we help you or facilitation for alarm management workshops alarm rationization workshops or help you to build your alarm philosophy or help you for your hazard and seal selection need loper and we also do cyber security or chairs or those kind of risk assessment so all those consulting services we also catering towards the customer and other than that um you also publish a lot of books and database for example the scrh is basically a database for all the instrumentations about their failure rate and how their different failure modes can go in it is listed within those ser h books we also have books on cyber security how exactly you can approach cyber security based on the iec 62443 standard which is a standard cyber security standard just for the ot space the operation technology space and we also have books teaching you how to do risk assessment and seal selection from hazop and also lopa we also have sim verification telling you how to exactly do those pfd average calculations or pfh calculations based on your demand mode and this is the book for that and you can understand why we are very proud of the math that we do for zero replication because we are probably the only one company has the tool excellencia to actually do uh serial verification calculation using markov movement a little bit more about our certifications so we have different kind of certifications we have certifications for different instrumentations for the oems so that would group into the iec 61508 certifications we also have iec 61511 certifications which is a little bit more towards the end users about how exactly they how they perform or performing some of the sat mats at the site as some kind of a functional city assessment we also have automotive focused certifications ing to the iso 26262 um and also sometimes for automations that's also related to 662 as well we also have cyber security certifications we have personnel um certifications for as a kind of uh for your resume that you can show that you have this particular competency to work in the functional safety field and we are actually the global leader in terms of a lot of the different specifications as well for example we have the larger market share in terms of the safety devices certifications and our main competitor the troops which is based in of course everyone should know them based in germany and we also have in terms of the specifically of our logic solvers um we have uh overcome a 20 years head start to actually become the leader overtaking truth and other competitors in terms of that and we also one of the leading leader in terms of the cyber security certifications we actually pioneered that with the issa to form an alliance to help other people and also different organizations to understand how exactly they can do certifications in that sense in the field or in the context of cyber security and we are really really proud of that in fact we are getting a very big customer base in saudi um to work on a very big cyber security certification project for the whole site um and as i've mentioned we also have our own software which is excellentia and the most i would say the most famous part mojo of excellent should be the silver silver part of it because silver is probably one of the only software out there actually uses uh markup modeling to actually calculate the piv average or pfh and also help you to do a lot of the same failure rates as well but other than silver we also have other modules and we're very proud of the fact that because everything is built into one particular software and with different modules and as you might be familiar with the iec section 511 standard we have a particular life cycle and to bridge different life part of the life cycle we need information flows we have different data flowing between different parts of it like for example from lopa we go into the srs so um what kind of information do we need to actually go into srs then so our tool is actually also built in the view of that that information from phaas for example what you did in the whole hazard can easily flow into the lower eggs when you try to do the lopa and finally when you try to select your selection you can basically do all basically import all the maths and data from your local eggs and then everything can be automatically populated into the srs tool where you can actually print out the document of the safety requirement specification and based on that you can have all the input information or actually in a way that what kind of test what kind of level that your set needs to pass and then use that as the basis for your superapplication in that sense and also we are very happy to say that i'm very proud as well to say that we have good integrations with other tools as well for example ourselves module our cell alarm software has good integration with different control systems like plant paax control system or the amazon's delta v system as well and we can import alarm data seamlessly from these control systems whenever you control whenever you create any kind of function blocks with any alarms associated with them and they can be easily and seamlessly be imported into the single alarm tool where you can basically manage your alarm information other than that we also have oem targeted uh tools probably uh because this particular webinar is more focused on the end users so um it might not be of much use to you guys but um it is very useful especially for the familiar eggs for the oems where basically we are adopting a method which called fema from the us military back which was infected with vector invented back in the 60s and 50s i believe that they tried to investigate failure modes so we use that to look at how different components within one instrument can actually fail what kind of vitamins they have and then try to see how their failure rates would become afterwards that is the tool for that for medicaids so that is a brief introduction about exeter now let's go into what we the main dish of today's topic then so we want to look at alarms so why do we actually need the alarm management in the first place that was because we see a lot of different alarm problems for example so this is the current situation in the industry right now we have um different uh different types of plants processing plants out there oil and gas petrochemical powers and although some maybe some other processing plants and what they're facing right now is that they actually have about at least about a thousand alarms per day and on an average speaking uh per 10 minutes interval they have more than five alarms annunciated to the operator and these kind of situation is very very dangerous when you have these kind of alarm overload situation as the term that we calling for that you basically don't know what these alarms are doing when they come coming at you at that large quantity and that is creating a lot of the problems in the operation because of that um and these kind of datas are not really just some isolated cases in fact every single plant that i went to to do alarm rationization if they had not performed alarm rationization or law management they do not have that particular scheme of management in place they have at least a thousand alarms per day coming at them and after the alarm rationalization the number of alarm comings and would be dramatically reduced afterwards and this is one of the problem and one of the most dangerous situation is that at one certain moment within 10 minutes you might have a situation where you have hundreds of alarm coming in and that is the situation whenever you have some kind of upset that would lead to very serious safety consequences like what happened in deepwater horizon where they were facing more than a hundred alarm coming in during the incident and that causes the delay for them to actually hit the emergency evacuation alarm if they hit it 10 minutes earlier than they actually did they could have saved lives there wouldn't be people dying in that particular incident during that situation so this is a situation that we really want to prevent that is really closely related to safety and other than alarm floods we're also also very often uh facing a situation where you have a lot of alarm just standing in the background not really knowing what they're trying to do um and then they're just catching your attentions and then distracting you and the operators doesn't even know how exactly what exactly the action is in order to clear them it's like the cat sitting on your keyboard trying to distract you sometimes so these kind of alarms are called nuisance alarms up one of the news in salaam the news alarm is actually a very big group of problematic alarms and standing alarms is one of them um and these kind of alarms are usually the ones that without operator actions or with our operator actions documented so therefore operators action by documenting them or by knowing them we can actually eliminate these kind of alarms or maybe in the first place if there's no action for them maybe we don't even need them and there's another big problem which might not sound very serious but it can be in during dangerous situations and also it shows a very poor management or poor organization of your alarm system which is the incorrect priority so incredible correct priority is referring to that you have a lot of the alarms portion of alarms actually at the higher priority than the low and mid priority the ideal priority ing to the standards the european standard or the isa standard is that it should always form like a shape of a pyramid that with the lowest priority taking up the top of the pyramid taking up the least portion of all across all your alarms and um sorry the the highest priority taking the the least portion of all your alarms taking up the top of the equipment and then your lowest priority taking up the most of your alarm so that in this situation let's think um when you have two alarms coming in there's a lower chance that you'll have two high alarm coming in at the same time because priority as it indicates means that when two coming when two different priorities coming in one lower priority one high priority coming in at the same time you should actually prioritize the high alarm rate so in that case that if you have all the alarms as high priority then that is no point of actually having the priority as an indication so in that case we actually need to actually use and make good use of this particular tool called priority when in our alarm system then what it means by whenever we have a low priority coming in at high priority coming we should always try to handle the high priority alarms in the first place so and and there's another big problem which is the duplicated alarms these kind of alarms um definitely they would clock up your your screen like what that is doing right now that you have high alarms with a series of different um reactors or even maybe some of the same d reactors but they have different measurements so these kind of duplicated alarms are usually um they usually comes from i would say the wrong situation or wrong presumptions about how alarms should be there was a philosophy i would say maybe borrowing from the common cause factor in the when we think about electronics or instruments from the nazis this government that they should also implement some kind of redundancy for alarms as well but human operates differently than the then electronics we can only process information more or less one at a time the idea of multitasking is basically how the cpu does it where you shuffle one task at the side and then you do one task and then you start one task and then do another task at the same at the other time so it's like an alternating shuffling time slot that when you whenever you try to do your multitasking and that kind of idea comes in whenever you see a redundant alarm that redundance alarms like that so the human brain can't really process that really well because of that and it also create another situation because it's right now clogging up the screen that if you have redundancy alarms that might comes in as a cascade from the top upstream of the process up to the downstream of the process then in that case that you basically lost you're basically lost in terms of what exactly is the root cause i would say that what exactly is the originating alarm that actually started everything so and that is also going to create another situation which i mentioned is the alarm flood situation where you have a very large quantity of lungs coming in at the same situation at the same time and then usually that you have a very very complex problem and the operator usually doesn't know how exactly they can actually handle that and they might easily just miss one particular very very very important alarm and that's probably lost in the within this particular hmi screen and then maybe it's pushed down um to the to the bottom of the screen and then they ignore it and then what happens next is a very serious consequence because of that um and this is also because usually these comes in in a very short period of time and coming in a very large quantity and these also are the most critical time time critical situation for the operator to actually do anything so by eliminating these kind of redundant or duplicated alarms we can actually help with the situation and one of the key thing to actually read to remove these is actually to understand the root cause of these kind of alarms and there is another types of alarms that are really really problematic for the operators as well which is the which would be the alarms that are without a response so basically the operator they don't know how exactly to respond to these alarms there's no action documented and uh maybe it just identified one routine event instead of an abnormal situation and sometimes it might even be some kind of diagnostic alarms that's not meant for the operator instead it's meant for the maintenance team or main perhaps either even meant for the system administrators not really for the operators not really for the operations who needs to operate the plant and alarms itself as you will see later when we review the definition in the isa standard alarm should always be catering to the operators this is not a tool um meant for most other people it's the main target of alarms are always the operation so we have to make sure that alarms should always have have an action from the operation to perform else they should not even be an alarm and that is the situation with stale alarm or standing arms a lot of the times so these kind of alarms usually they remain there for more than one hours more than days and i've seen myself an alarm standing there for more than a year where there's no authorization and whenever i ask the operator they just say that okay it's just there it's just hanging in the background we don't need to do anything i don't care about that it's just something over there and they don't even understand what exactly it means a lot of times so these kind of alarms would usually um [Music] just take another place on a place on the alarm state plate and in a way that it would also uh try to draw the operator's tension taking up some space it might also push at alarms to the bottom of the screen as well and that creates a lot of problems as you might think and how exactly we can actually try to eliminate these factors one of the key thing is actually to document the operator section so by documenting the operator's action and understanding whether there is actually an action in the first place for the operators we can easily eliminate these kind of stereotypes so there are far more problems of alarms so what i am doing here is just to mainly focus on alarms that have problems that's related with the problems of root cause and also going on with alarms without a response without business actions um and because um the time that we have today is really limited so i just focus on these two things because i'm trying to focus on the main two topics today how exactly we can apply the philosophy from the data science to actually tackle these two types of elements so after talking through all these problems of alarms so what exactly is our management how exactly we can perform this alarm management to help ourselves to help us to address these kind of problems so there are few purpose of alarm management and very the first thing that will pop in your head i would say is probably to ensure safe operations but i would say in my experience for sabot while it is very important aspect of alarm management but safe operations is actually usually handled by the safety system and that is if you have a safety system i've been to so many plants that they don't have as you can see where they only rely on alarm system to actually perform all the safety operations and in that case you pretty much stuck with the alarm system i know but if you have a safety system that means that your safety system would kick in and make something drastic or something disastrous would come in so in that case what we try to do is actually try to prevent the activation of the safety system because whenever you try to activate the safety system you will end up with unplanned shutdowns and you might have damage to equipment and also you might have some kind of processing incidents as well like you have some kind of emission um that might breach your environmental permits and then you end up with a phone so that is usually the case why um we need the safe why we need an alarm to prevent the activation of a sis or maybe some other kind of trip or maybe it's like burner management system to prevent these situations and also the alarm system itself is should be actually a tool to help the operator to perform some kind of roles this is especially true when we have a batch plant where sometimes they use the alarm to actually indicate when the or when they can actually go into the next phase of the reaction like if they're cooking maybe brewing them bring some kind of um like a vehicle where they brew alcohol so they need to keep the temperature maybe at i would say 60 70 degrees c celsius for a period of time and then they need to drop it down uh cool it down to maybe 30 degrees c for a period of time and using these kind of alarms temperature alarms it would help them to understand when they actually go into space and also the alarm management itself is being outlined by standard and also these kind of standards will help you to basically fulfill or comply with some of the regulations based on your geography then if you're in the u.s it would be the osha dsm that you try to fill and try not to basically about them and also it would help you with the bullet producer quality products um because you might degrade if you have a wrong temperature set when you try to basically do or one so this is the full um i would say poor life cycle in the isa standard for alarm management and today we'll mostly be focusing on the rationalization because we're trying to look at the root cause and also trying to talk about operators action and those are the stuff that are being discussed in the rationalization where you basically try to document all the aspect of an alarm and if the alarm doesn't really fulfill some certain requirement then they will basically be degraded that's not and the result of that is rationalization is to produce something we call the master rounding base so ing to the standard an alarm should always follow this three folds of definitions the first one is that it should be something that's targeting the operative and also catching their attention it should be an audible or visual me and then we should be indicating an abnormal situation something that is going wrong and finally you should need a tiny response if you it needs an action and the action should be within a period of time um for example if you have an action like writing a particular report are maybe within a month then i would really not say that it's a proper of the proper operator's action because that's not the type of response a timing response usually is within 30 minutes or within them now that's the type of response and therefore if we realize that the alarm doesn't fulfill any of the definition right here we can actually put it as alert prompt or message based on how exactly uh or what kind of condition they are or what kind of actions that they need and as you can see that um these kind of alarms they usually should indicate a male function aviation they also need a timely operators actions and in order actually to help the operator to understand what this alarm started from where it is originated from we also need to improve and also try to investigate what exactly is the good cause for these arms and this these kind of ideas are actually being mandated by the isa standard um especially when it comes to the consequence of inaction and also the operator section they are required information by the isa that is should be within the master alarm database and this master alarm database should be well kept well catered and well maintained and it is also recommended that you put some pauses in order to help the operators operate in that sense so we talked about alarm management how exactly we can do it let's see what kind of steps or what kind of information or philosophy fundamentals that we can borrow from data science in order to help us to flush out these information so we talked about root cause and the root cause is also very important in the in the data science discipline i would say um as i have explained before ing to the definition it is basically outline the reason why one particular problem occurs and within the field of data science or data analytics they actually use a process that's called five why's to identify that particular impulse so how exactly do they do that on the paper is simply as just asking why five times it is just to try to actually find out what links up to is the fundamental problem of fundamental reason so you what we want to do is try to ask why repeatedly to unmove the call of the problems for example why are you late for work the first one why are you late at work because i missed the bus and why did you miss the bus because i left the house late why did you leave the house late because i got up late and why did you get up late because i was up one night and why were you up all night that was because my children my child was ill and that is the cause of your problem let's see how exactly we can apply that in a process since in in a process let's consider that there's a compression session drum here and there's a high level alarm right here at this lc101 and there is also a compressor trip at this lsh203 um and once this switch is activated and on high level then it would basically trick that compressor it also has a a drain box here right here that's controlled by the c101 so now considering that is a higher level market let's ask first why do we have a high level alarm and actually believe alone whenever i ask that question to the operators this is the answer they get because there's high level in the compressor compression suction drop and they just stop there they don't they don't think deeper into the answer so i had to a lot of time ask them a second why so why why is there a high level there that's because the intake is actually higher than the outflow from within this structure um and why is there an imbalance of intake and outflow that is one of the reasons could be because you can't drain the vessel so when i ask this third why it can actually branch out into different other causes as well so one of the branch would be because you can't drain the vessel the other branch could be some kind of offset upstream that would lead to more flow from the offspring into this electron but let's stick to this calm during the vessel branch and last unless asked the fourth one why why can't you drink vessel something wrong with this valve so the valve is stuck closed so that you can't drink faster and why you have this problem with the bar it can either be the mechanical problem with the bowel itself the bowel fails itself or it might be the failure of this particular particular controller that actually is telling the valve to close so there are two different root cause for this particular branch where you can't treat the vessel and that is how exactly you can arrive at that kind of good courses now let's take one more step we have the root cause now so as you can remember might remember i talked about that there were alarms that were up without responses without operating sections and we that's and that's why we end up with a lot of the standing problems so within the data science field that is actually the time that when they use the get analysis so in the data science what they try to do is that they have a course for one particular business plan for example they can't sell um a blueberry this year and then they try to breach that particular gap to actually address that problem that they can perhaps sell more of that particular product so to in order to breach that particular gap they need to derive different action plan and these kind of action plan exactly is if we put that in the context of allow management is what the operator's action needs to do in order to correct the root cause of the alarm to put that alarm back to the normal situation to basically clear them now let's see how exactly we can do it on that example that we have discussed so we know that there were two reports the first one is uh the valve thing so what could be the action of that so because right now if the vowel is spelled then in that case we have to do something about so we can either manually try to operate the valve in the field in that sense because it's right now stuff so we might need to train uh crank it open um and we need to someone in the field to actually perform that um how about the other results when when you have a failure of the controller to actually help with that situation you can actually have two different actions depending on how this particular controller goes for example if the controller fails in a way that it's just filling its automatic function um in terms of how the set point is basically uh set at then you can actually put that controller in manual and then basically telling the controller that okay right now we have a high level situation we need no survive the other thing what you can do is based on the design of your hmi screen maybe there is a button for this particular file on the hmi screen on your control system that you can click that can send a signal to the vowel that would actuate it and then close about from the processor so that is another action another way and if you don't do this action by addressing these problems then the result would be the compressor trick and that is exactly the consequence of inaction that the master number faces so as you can see by understanding the root cause we can understand how where this alarm originates from we can start to divide the action path and then in a way that we can document the operators action to tell them to let them know how exactly they can troubleshoot these kind of situations and as we have mentioned there were situations where you have a lot of duplicated alarms um by understanding root cause we can actually eliminate another already especially when we have alarms that's within one particular reactor but you have different measurements like here with a measurement three measurement four because they have the same root cause then why don't we just take one of the lava and then silence basically take out the other arts because that is basically different a lot or in some cases we can actually also use something we call the advanced alarming technique so that would be something like uh doing some kind of programming in order to help the situation um i'm not going to go through in details about advanced alarming techniques because it's a it's basically a number of slides for a whole afternoon when it comes to other onslaughting techniques because we are talking about how exactly we can do different programming and to silence some some kind of alarm based on their state or the state of the operations or maybe we can do some kind of first out suppression for example within a compressor trip then you usually would end up with a lot of alarms on the discharge and also maybe the info and even some kind of vibrations or electrical circuit search alarms as well so you try to move whenever the compressor's trip that comes in first and try to basically silence the rest of the alarms so those are a few of the examples for advanced alarming if you guys are very interested in details how exactly we can do these um different methods in terms of how we do the advanced alarming techniques of what exactly suppression means now you're welcome to join our courses online these are the online self-paced training these basically online courses where you don't really need to follow a schedule you don't need to do it within a week whereas there are also courses which allows you to do that and this is these are the upcoming courses we have um in the upcoming months within this year within 2022 that we have pauses for uh like for the ic6 1911 standards and also we have some kind of life cycle trainings uh or maybe machine safety training and usually they are online and we also do have like courses that is in person um like we have some forces in december that's in our headquarters in salisville pennsylvania that we have this episode 100 bonus um so if you guys are interested um welcome to uh send us a question i'm also welcome to follow us up on twitter on facebook and we also have a lot of information like the white papers or also reference materials on our website so though white papers are really good materials and resources for you guys to actually look out to to look up specific topics because we try to make them like academic papers to address some of the questions of some of the problems that we see during operations and um and some of them can actually uh maybe address some of your problems that you see during your operations so um we do have a bit of time so we'd like to see if you guys have any questions um regarding um the information that i talked about today like how exactly we uh how exactly these kind of uh life cycles are being mandated in terms of management of change or monitoring assessment how exactly they were done within the isa stem right i have i think i have a question about our tool interesting so yes yes so silence itself um suicide right now itself is uh still um a standalone tool but we have plans to actually build them within the facility suite um but solar itself the main purpose of it is actually to help you with the um i'll say the rationalization process in a way that we can actually because you can outline the flow of how exactly we want to do the alarm measurization and the flow and the steps that are listed within the sew alarm and also rigidly are being implemented throughout the interface of the silver is actually ing built with the i would say with a mindset that was built um that's streaming back to the i'd say standard about what kind of information are needed what kind of workflow would be best or recommended during the rationalization process um and then we also have different fields that you can fill it in that is be basically outlined by the standards like what is the priorities what are the calls and consequences active actions but those are very intuitive ways for you to conduct rationalization but most importantly sue alarm itself is also a database because the deliverable for the alarm rationization should always be a master bomb database and for the master alarm database we need to make sure that any changes because it's the all the alarm data within the master lab database are actually they are actually being confirmed and they are also being processed by everyone in the team in the um alarm right transition where you have operations where you have process engineers instrumental instrumentation engineers you have control engineers you have other kind of personnel so everyone agrees on so in that case um any kind of changes in this you actually need to go through change and the zero alarm software itself is actually trying to help you to make that particular mastermind database to stay static i suppose and that is what i see the main strength of the serial knob you can have any kind of database even like a sequel to do llamas transaction but you can do very easy changes and they are not really compliant with the iec or isa standards so that kind of a locked in the information or not institution can actually help for the compliance with the isa site um so that i hope that answer your question i do have another another question on um on some of the data that we show some of the target within the is different isa standards so yes um so these kind of information these kind of recommended practice or the numbers um every time when i finish an alarm rationization process i every plan that i went to after they finish everything they are easily achieved these kind of information that they have average alarms about one to two alarms for maybe 10 minutes in the ball and a lot of the times i actually would be able to see the number of alarms within the day less than 100 a lot of the times and the distribution itself is very easily achievable once you start documenting the upgrade the operator's action whether these are right for alarms or not so we can actually eliminate a lot of the high priorities of the phrasing there so i would also uh maybe start another webinar to talk about priority specifically and these would be um after you understand how priorities is being selected you understand how easily it is to actually achieve these kind of numbers for expression of priority distributions and i hope that answers your questions um and i have another questions about the rationalization objectives across that process so um as i mentioned um as i've explained through why uh how the zero alarm could help there's basically the rationalization on the purpose ing to the standards is basically trying to document and all as much as information for the law the first thing we try to do in during the rationalization is to see whether this alarm is better if the alarm is not valid we just take it out and then if it's better we start to drop all these informations we have the calls consequences actions and also the time to respond and then we try to select the priorities and then also to see whether it's um what kind of classification it has and then see whether there's some kind of a maintained schedule specified for these kind of test classifications as well and then all these would form gradually into a big advanced alarm database where you try to keep them as static as possible and why exactly why we need to keep them anesthetic one of the reason as i mentioned because these are authorized alarm attributes that you have the other thing is during monitoring and assessment process is that you need to periodically compare the alarm setting within your control system compare them with the master alarm database to see whether there are any unauthorized changes within your control system so if there are any important authorized changes then you actually need to go through the management of change to address whether there are extra risk associated and that process is part of the ultra dsm and that's why it's very important if you don't do that you're breaking the regulations by option and i hope that would answer your question um i think i don't have enough time we've been going through about more than an hour so if you guys have more questions please feel free to send me an email um my email is basically as soon as 4.10 at exeter.com and uh i'm happy very happy to answer any extra questions that you have and you can also go to our website check out our white paper white papers and books and see those extra guidance just shoot us an email we're happy to help and i'm really really grateful that you guys come to this webinar and i hope that you guys have uh enjoyed the rest of your day and have a good it's nighttime for you guys please enjoy your evening and have a great night thank you very much i'll see you next time