Streamline pipeline scada alarm management for corporations

How to create outlook signature

okay good morning everyone my name is Joe Shera I'm the VP of marketing here at novatech I want to extend a warm welcome to everyone who has joined us today for our fourth and a six-part webinar series entitled alarm management Essentials presented by novatech and our strategic partner Pas we'll save all the questions for the end of the webinar if you have additional questions after the webinar Kevin Johnson's contact information is listed at the end of the presentation Kevin is our VP of development here at novatech and is responsible for the partnership between us and PAs you may also contact your current novatech account manager who will be happy to assist you in arranging a demonstration of the novatech Pas alarm management solution if you were not able to attend any of the previous webinars you can view them in entirety on our website at .n novatech web.com to access the recorded webinars click on the alarm management menu item under the process tab as a reminder the next webinar is scheduled for April 11th at 10:30 a.m. eastern standard time and will cover control Loop performance monitoring in June of 2009 Isa released the 18.2 standard entitled management of alarm systems for the process industry this Landmark document is the first true standard on how companies are expect to design Implement and maintain alarms generated by process control systems Pas and novatech have formed a strategic partnership to bring our customers state-of-the-art alarm management software and Analysis tools these tools are designed to enhance overall plan efficiency and operator Effectiveness Pas is a recognized leader in the field of alarm management and novatech is known for its commitment to delivering worldclass and value added Solutions the Pas plant State Suite allows novatech to provide a featur set of alarm management and Analysis tools when coupled with our novatech V3 DCS furthermore and just as important it allows novatech to support customers with a wide variety of control system platforms as well as provide a single point of contact for project planning installation training and followon support today webinar entitled real-time alarm management system control and maintenance will focus on steps six and seven which provide for a dynamic alarm management during the various stages of plant operation such as startup shutdown running and equipment maintenance techniques included in this section are state-based alarm management alarm flood suppression and alarm shelving we are joined today by Mark carrian Mark is the vice president of business lines at Pas he has over 20 years of experience in the process control industry including engineering and Commercial positions he joined Pas in the year 2000 and his roles have included alarm Management Consultant managing director for Middle East operations and manager of our of the Pas Consulting operation I will now turn the webinar over to Mark who will guide us through steps six and seven of effective alarm management very good thank you very much Joe for that introduction um I I thank everyone for attending and taking some time uh for this webinar as Jill mentioned today we're going to focus on steps six and seven uh we went through the other steps in the previous webinars uh but just for those who have not attended this I will briefly describe the first few steps and then we'll get into more detail on six and seven as it relates to six and seven I'll spend a little bit of time on the reasons why these steps are important and the benefits you can expect to get from this and then we'll delve a little bit into the actual technology that we offer to help you accomplish this so basically uh step one the alarm philosophy is a critical step uh for basically setting a standard for how you're going to design Implement and maintain the alarm philosophy so that everyone in the organization's on the same page in regards to how we Define an alarm how we set them and how we maintain the system steps two and three uh collecting data uh to venture the system and establishing kpis for performance uh and step number three bad actor alarm resolution are quick ways to understand the current performance of your alarm system as well as with relatively little effort uh get immediate Improvement to the alarm system um for resolving the bad acting alarms we typically find for most Control Systems a relatively few number of alarms say 10 to 20 alarms cause anywhere from 30 to 60% of alarms generated by the control system and by identifying those and addressing those you can quickly uh improve the performance of your system step four alarm documentation and rationalization is essentially a re-engineering of your alarm system to match the standard you set in your alarm philosophy fully documenting when you set alarms what the operator response should be and being consistent in regards to how we set alarms what should be a high priority versus a low priority and taking the guesswork out of it if you will relative to we set of learns and when we don't step five uh audit and enforcement technology is a way to maintain the system that is to have an improved alarm database uh and ensure that any changes that may be taking place were properly documented and agreed to and with that we'll get into more details on step six and step seven so Dynamic alarm management one thing that you will find uh in a typical processing facility is that the validity of alarms of an individual alarm is very dependent upon the state of the process as a very simple example if I have a compressor and that compressor is running I may have 150 alarms associated with that compressor that are meaningful while that compressor is running I can have you know flows and Lube oil uh pressures and vibrations Etc all of which mean something when the compressor is running but when an upset takes place and the compressor trips then I may get a whole bunch of alarms that are now meaningless so the compressor is tripped and I start getting low flow alarms low Lube oil pressures all kinds of alarms that at that point in time really are meaningless because the compressure is down the problem you can run into is that flood of alarms that is coming in during that condition may be masking an important alarm that the operator should be responding to and he may not see other alarms that are coming in that are actually meaningful from some other part of the process and uh there have been numerous um industrial incidents where an operator has quote missed an alarm that has led to an unfortunate uh you know whether it be a plant shutdown or environmental incident Etc H but I would submit that that is not an operator error if you flood him with a whole bunch of alarms you know several hundred alarms in a few minutes and he misses one you set that person up to fail you cannot expect them to Wade through that much data that quickly and find the one important alarm out of many many that are not important Dynamic alarm management or real-time alarm management uh is is the two names for basically the same thing which is designing alarm strategies to match various plant operating conditions this might include plant startup plant shutdown various upset conditions and having different alarm strategies for different operating conditions from P's perspective we have our plant State Suite software which uh sits on a server on the process control Network and interacts with the control system whether it be the novatech D3 system or others and provides the strategies to read and write from the DCS to change alarms to match current plant operating conditions and make the information available to the operator through their DCS graphical HMI so the main uh uh applications there in including the alarm audit and enforcement alarm shelving and dynamic alarm management so just briefly what those applications do uh alarm shelving is typically used by an operator uh when there is an instrument malfunction or alarm is is is frequently enunciating but really isn't providing any meaningful information the operator should have a way to temporarily suppress malfunctioning alarms but it should be done in a controlled and a safe way such that when he suppresses that alarm we won't forget about it later again we've seen incidents where an operator has temporarily sheld an alarm he makes a note in a log book and 3 months later no one knew to go back and reenable that alarm once the instrument or the piece of equipment has been fixed so shelving is a way for the operator to select specific alarm or group alarms that will temporarily be sheld then the operator typically has to put in a reason why that um alarm will be sheld and then the system can set a maximum time when that alarm can remain on the shelf and then typically speaking it shift changer at the right time the operator is going to receive a message from the system that says here's all the alarms that are on the shelf and here's why they are on the Shelf but if the operator does not uh respond and reshel them they'll automatically turn back on again so the operator gets what he needs uh is a way to temporarily turn off alarms but in a controlled way so that we don't forget about them and and forget to reenable it Dynamic alarm management are basically there's two two modes of it State based alarming different alarm st strategies for start out shutdown Etc and flood suppression to deal with upset conditions for suppressing alarms as we'll talk about in the next slide Dynamic alarm management isn't necessarily about turning off alarms it's about the right alarm at the right time and here's an example of what I'm talking about uh in this particular case uh you can see this tank here and when the we are operational with this tank I may have a high pressure alarm at 90% And or high level alarm at 90% a low level at 5% and a high pressure alarm at 250 PSIG but if I've taken this tank out of service I may not necessarily want to just get rid of those alarms but change them to match the condition so in this case I have tried to isolate tank 405 and may be doing maintenance on it I want to know if I have lost isolation because that could be a hazardous condition if people are planning to work on it or renter the tank I want to know wait a minute here I've lost containment and now the the vessel is starting to energize again so in this case I may set a high pressure alarm at 5 PSIG and a level at 2% uh and get rid of the lowlevel alarm so again I have a alarm strategy set for this specific process condition again the shelving we talked about this already uh it's the way for the operator to um quickly um suppress meaningless alarms but the safeguards are then in place so that we do not forget about those and they can be reao they can be automatically re-enabled if people forget to go back and change it now here is one of the key differentiators that we offer as a solution and a real value ad we bring to the DCS system uh sometimes we have seen in other control environments where a number of independent applications whether it be custom built application inside the DCS or a pat nwork of different applications serve as a real-time alarm management solution but there are some issues you can get into when trying to have these things separate from each other so for instance here if I have one application doing an audit and in force and a separate application doing Dynamic alarming if my auditor goes out and does a check on the system say 8 o'clock in the morning and it sees a bunch of alarms that are currently in a different state well the dynamic application may have changed the state if the auditor does not know about that then you can return a bunch of false positives it's the same thing with the dynamic application I may change a state and enable a bunch of alarms but the operator may have shelv the alarm through a shelving application and they come back on again when they should not be back on again because they were properly shelv using the controlled mechanism so what we offer with our PSS Master alarm database is a Consolidated application for audit and enforce shelving and dynamic alarm management and basically if you think about it if I'm going to have more than one application that could write out to the same tag. parameter or the same alarm parameter I had better be sure that those two applications are aware of what the other one is doing so they don't fight over and make inadvertent changes to that same tag. parameter so our audit and enforce application basically knows um that uh if I go out and do an audit it's going to check with the shelving application and it will check with the real-time application and when it does its audit at say 8 o'clock in the morning it will audit for the alarms based upon that state the shelving application that we have if it's say noon and it's time to reenable a bunch of alarms that were on the Shelf the shelving application will check with the dynamic alarm management application and restore the alarm to the condition it's supposed to be in based upon the current state similarly the dynamic application say there's been a change of state and a compressor has been restarted or maybe we're back up at full rates and we have different alarm settings or I'm in a different product grade and different alarm settings for different product grades when it changes the state the dynamic application will check with shelving and if there's certain alarms that are supposed to be off due to maintenance purposes it will not reenable those alarms because they have been properly sh so the software platform as I mentioned earlier is our plan state Suite software it is an integrated platform with all the various applications inclusive of alarm and event analysis the documentation and rationalization and then the applications we've been talking about here this morning alarm and and and enforce Dynamic alarming which includes shelving State suppression and and flood suppression Etc as Joe mentioned earlier we'll be talking about some of these other modules in other webinars I did want to speak to another application that the alarm audit and force is used for which is event triggered audit and let me describe what that is oftentimes when a shutdown takes place we expect a shutdown system to take a number of actions close this valve open this valve move this controller whatever the case may be a shutdown system takes a number of actions to put the plant control system into a safe location there could be 50 or 100 actions the shutdown system takes to take the plant to a safe shutdown condition oftentimes the current procedure at a plant is that an operator reviews a list of actions by the shutdown system to make sure everything that was supposed to take place did so the operator has to weed through and go through a whole bunch of different logs to find out okay this happened good this happened good this happened good but what the operator really needs to know is did anything that was supposed to take place not happen the operator needs an exception report not a list of all the things that were supposed to happen that did so the event triggered audit you can set up an audit to go out and look at every tag. parameter following a trip and poll is it in its proper location is it in its proper setting is it in its proper State and then provide an exception report to the operator that would say hey this shutdown just took place and here are two things were supposed to happen that did not please go out and take an action on this it's just another application of the same technology the alarm audit and enforce to help the operator uh times when they are very busy during upset conditions to give them the right information moving on to so that's the uh the discussion around step six which is uh Dynamic alarm management moving on to step seven control and maintain like any alarm excuse me any system at your facility you must maintain you must have a maintenance plan in order to maintain any gains you may get by embarking upon an alarm management project and and uh so it's critical that you put work processes in place to uh maintain the performance of the alarm system but one thing you don't want to have is some kind of work process or not have the technology where it takes a lot of time to maintain the system so there's several things that are very important to put in place both from a work process standpoint and Technology standpoint to keep the alarm system working well first and foremost was the management of change policy and process by which uh we decide what constitutes the management of change who has to approve it and how we implement the changes permanent changes to the alarm system that may fall outside of just a temporary shelving from the shelving application the next very important thing to do is to have metrics that is what constitutes good performance of the alarm system what metrics do I want to monitor Isa 18.2 spells out very good metrics that you should be using to monitor the performance of the alarm system some of those will include number of alarms per day per operator position what percent of time your alarm system is in flood how many chattering alarms do you have per day how many um uh standing alarms do you have uh these things can be monitored and tracked and as the performance starts to go in a bad Direction you can solve those problems while they are small rather than waiting until you have severe alarm system problems again the other thing that we recommend implementing at most sites is part of say a weekly operations meeting uh that can be appended to include a short discussion maybe 15 or 20 minutes about any alarm system issues and typically speaking um the software will produce a report of say the worst acting alarms a top 10 report of the worst acting alarms that are causing some problems and then the team can discuss those and send someone off to go fix those specific problems so again uh there are some relatively easy things to do work processes as well as technology to put in place to monitor kpis and keep the performance of the alarm system where you want it to be here is just an example a generic example and in a future webinar it will be much more specific to the novatech D3 system of how an operator uh may be receiving information uh from a uh system like plant State Suite so this is just an example of a of a of a of a graphic we've designed for a customer a level what we call level one overview graphic uh that that provides monitoring information for an operator uh but if an operator say gets an alarm so I can see that I currently have a like a level two alarm that's in that yellow triangle uh the operator can click on that alarm and then we can configure a right click menu in the DCS Graphics system that can pull relevant information from our plant State Suite software so as an example here there's one of the items is alarm details so from the operator perspective I may want to know what am I supposed to do about this alarm I want to have the alarm response menu if you will Dynamic and in the graphical system and so by clicking on alarm details uh he can pull up from the plant State Suite software all of the documentation that was done during step four the documentation and rationalization what are the causes of this alarm what are the consequences of not responding and what actions am I supposed to take when this alarm comes in Additionally you can pull up you know standard operating procedures documents anything else you may want to have linked to a specific piece of equipment or to a specific um um uh you know tag or whatever you can link specific documents they can call up on demand if required um and additionally you could shell the alarm from that graphic and pull up other detailed information as as be required in regards to any next steps that you may have if you are interested in learning more or if you're interested in finding out am I a good candidate for alarm management man agement is it something that I should be worried about we currently do uh offer a number of different things first of all you can contact novatech for additional information uh we're happy to arrange for on-site demonstrations of this solution also a thing that we are offering is a complimentary analysis of your alarm data to provide some insights relative to how is the performance of your system and what are some steps that you might want to take as an example we'll produce a report on the number of alarms per day some of the worst acting alarms and what alarm flooding problems you may have and action you may want to take to solve those so again if you have any questions please contact your local novatech account manager or please contact Kevin Johnson directly you can see his email address and his cell phone number here and he'll be happy to give you more detail and information about the solution and how we'll be able to help you so thank you all very much for your time uh we appreciate you spending it to learn more about this and I I think we're going to open this up now for anyone who may have any questions okay Mark we had a couple of questions come in during your talk the first one is can the Pas system connect to multiple Control Systems simultaneously the answer is yes we have many installations where I have a single server that is pulling data from multiple Control Systems um and you know I I could just off the top of my head I can think of sites where I have a single system pulling data from a Honeywell and Inventus and Emerson and uh a skada system all come into the single server for a common reporting environment and everything else okay and the second half of that question was do you have a list of all those control systems that you can connect to the answer is yes um we have a list of what I'll call Standard interfaces but we're building new ones all the time uh what I would suggest is the person who asked that question um i' I'd be happy to provide them with a specific list or if even they want to say the system they're they're interested in we be happy to answer that one specifically I'd like to make a you know we'd be happy to send out that information to the individual that asked that i' request that that person um you know contact Kevin directly and we'll be happy to provide that list okay next question was with respect to the management of change how does the system administrator get notified when an authorized an unauthorized change is made very good question the audit and enforce module what we do is we go out so the audit enforce module has what we call a master alarm database the approved values that everyone has signed off on and said this is the right value then on a scheduled basis the alarm the auditor will go out and pull the DCS system to find all the current alarm settings and compare them to the master alarm database if it finds a mismatch you have three options on what you may want to do with that mismatch the first option is you could just send a report to a specific individual that says here are mismatches or unauthorized changes currently existing in the DCS system take an action and do something about it option number two is excuse me you could send a message to a specific person that says here are the unauthorized changes and if you give me permission I the audit en Force application will automat atically change the control system back to match what is currently uh in the master alarm database the third option you have is that with no human Intervention when the audit and enforce module finds a mismatch it will automatically change the DCS back to the approved value what you do with individual mismatches is configurable down to the tagged do parameter so you can decide that certain alarms are automatically changed back so maybe all of your high prior priority emergency level alarms if there's a mismatch between the DCS and the master they're changed back with no human intervention and maybe the rest of the alarms just fall under report so you can decide for yourself how you want to deal with mismatches whether it just be an email or change of control system back okay I've got another question uh two another two-parter uh it says on slide 15 you displayed alarm count can I do that within total vision um and that's probably something that I may need to help on that answer I'm not sure if I can actually access that information I don't know if you have that answer Mark um on that specific type of information let me let me read the second part of the question um which I do have an answer for and maybe that will help with the first part it it goes on to say on slide 15 you right Mouse clicked and had a list of options is that something total vision can be configured to do and the answer to that is yes we've um in in our uh 14.1 uh software we've actually um implemented a right Mouse click functionality that will let uh a user or a designer to set up a menu like Mark displayed there and um we have direct linkage to the um documentation and rationalization information inside of Pas and we have uh demos of that um that we can show you uh in terms of the actual getting the alarm count I'm not sure we can drill down and get that I don't know if you have that answer Mark sure the answer is yes if you wanted to we could create a link in here where it would run analysis from our software and provide you with a historical you know how many times this alarm has gone off of the last day week month or year most customers don't do that in this graphical UI because it's not necessarily something the operator needs dynamically that is when he's making decisions about the process and what do I do about this alarm he doesn't necessarily need to know that hey how many times alarm gone off in the last week month or year that's not something they typically need they typically might just go to the normal plant State site interface to get that information but if someone felt that that was important and I wanted to provide that to the operator we could configure that as part of the rightclick menu it's just another option we would do to a different call to our database to bring different information back okay uh let's see another question here uh would the audit and force module be kept in sync with any Dynamic alarm strategies that might be in place at the time of the audit absolutely that is one of the key I'm gonna I'm G to scroll back up here to the slide that talks about this it's one of the key things that we think offers differentiation it's exactly what the auditor is going to do it's 8 o'clock in the morning it's time for an audit it's going to check with the dynamic application and the shelving application what are the alarms supposed to be in right now this compressor is off these various alarms are supposed to be off ordered against that these 10 alarms have been sheld by the operator they're allowed to be off ordered against that this particular tank is offline so the alarms are different there's a different alarm priority a different alarm set point it will audit against that and so you can be very confident you're not going to b a bunch of false positives from the auditor because it knows exactly what the dynamic application is doing and the shelving application is doing because this is all a singular software application within plant State Suite okay I have uh no further questions at this time um I'll give people another few seconds to to add in any more if they like if not I would like to exess my thanks and um happy that everybody attended the webinar I wanted to remind you that the next uh in the series webinar number five will be held April 11th at 10:30. Eastern Standard Time and that will that will cover uh control Loop performance monitoring the video uh will be available and posted on our website for those of you that may want to review it or share it with one of your colleagues and again uh thanks from novatech and PAs and uh have a good day thank you all