How to Calculate OEE and why you need to

all right the new studio believe it or not this is our third try at this first video because believe it or not drawing in this space is hard and distracting and Zacks on the other side of a piece of glass but to get started I'm going to do it just a warm-up video about OEE which is a response to a conversation we had earlier this week so this video is why you absolutely have to be calculating OEE it's not optional take zero so to start we'll start with our automation stack because I think it looks beautiful and the automation sack applies to the holy grail which we're going to talk about this week so this is what every business is striving to have mes in the middle supervisor control and data acquisition and PLC HMI ok anyone who watches our videos should be familiar with the automation stack we use a five layer automation stack the traditional stack is a six layer where PLC and HMI are broken out they've actually merged into one another so whether the state state of your business is wholly industry 3.0 or your state of the business is wholly industry 4.0 the structure of your business is this stack there's data at every layer and none of it is integrated or unified okay the traditional environment I have my PLC's out on the edge my PLC's are doing all of my automation to keep my employees safe and to keep my equipment running uniformly and repeatedly ok the HMI gives us the ability to control the automated environment the supervisory control and data acquisition layer gives me alarming and real-time monitoring of my entire operation at the plant level in it from a control room manufacturing execution allows me to Tran to convert a sales order for a bunch of widgets that's in the ERP into manufacturing on the plant floor that's what this layer is and this is where oh I lives ok and cloud is where we dump all of our data and our information to retrieve for reporting and to do what's known as post processing data analytics that is what can we learn from the way that our operation our business operates here's what you almost certainly have you almost certainly have this in actually you don't almost certainly you absolutely have PLC HMI and you absolutely have a RP you may or may not have SCADA in some way shape or form if you have alarms that you can see centrally not just on the machine then you have skated in some form and very few people have mes a software layer mes and very few people are using cloud-based analytics although everyone wants to be using it everyone wants the holy grail which we'll be talking about in a minute but this video is about whether or not you have an mes system a software based mes system you absolutely have to be calculating OEE and it's not optional okay Oh II II the OEE calculation overall equipment effectiveness is a percentage between zero and a hundred percent there is really two metrics there's teep okay t p-- is based on a 24-hour 7-day schedule that is if i was able to run 24 hours a day seven days a week if my machine produces one part per minute so I can produce one part per minute that means that I could theoretically produce 1440 parts in a day the calculation the teep calculation without taking into account quality or availability the teep calculation says if I produce 90% if my teper is 90% then it means I produced 1296 if I produce 1296 in a day then it means my T per one ninety percent I assure you no one is operating at 90% teep o ee is overall equipment effectiveness and it takes into account your actual schedule so you don't operate 24 hours a day in seven days a week you have safety meetings you have shift change you have change overs from one product to another so you can't when you're doing a changeover you can't be producing one part per minute when you're in a safety meeting you can't be producing one part per minute what oae does is it takes into account your actual schedule it also takes into account changing rates your teep number is always based on the theoretical rate of a machine ok and and I'm going to get into why this stuff actually matters here in a second for the closed loop holy grail that is I want to use machine learning to make optimal decisions for my business and tell me what to execute on the plant floor what order to run next whether or not the stop disorder when to schedule my my shutdown which is ultimately what your goal is you want machine learning and artificial intelligence to make those decisions for you that's what you want one of the things that the machine learning and AI algorithms have to have is OEE they have to know how efficient are you operating ok they also want to know this teep number that's an easy one to calculate it's basically actual versus theoretical in a 24-hour seven-day-a-week schedule so OEE is takes into account your actual schedule and it also takes into account your changing rates it doesn't use theoretical rate I can create a standard rate for how fast I'm supposed to produce a certain product on a line and that rate is a function of lots of variables we know that the tank heater doesn't operate that well at this temperature and therefore I need to run the line slower so the heater can keep up those types of things Oh II takes into account this so what is OE OE e is a QP availability quality and performance okay these numbers are all on a scale of 0 to 100 percent very simple availability is what percentage of the time that the machine was scheduled was it available to run so this takes into account downtime when they're seen broke when the machine was down this number will tell us what percentage of the time when I was supposed to be running was I down because of some maintenance issue ok quality is very simple it's the number of good parts divided by the number of total parts how much waste did I produce ok and performance is what did I produce relative to what I was supposed to produce performance only takes into account production when the machine was available so the first thing that we have to do with OE e is we have to know how much did the machine go down well here's the thing it's not as simple as it looks availability has two points the first thing you do is you track all down time number one the next thing you do is you say was it planned or unplanned ok if it was planned downtime if it's planned downtime you subtract that number let's say 10 minutes you subtract from the scheduled time how much I was supposed to be running so if I was originally scheduled to run eight hours and a shift and I have a 15 minute planned safety meeting that's still a planned downtime event I need to subtract those 15 minutes from the 8 hour shift to get 7 hours and 45 minutes of available time to run so I can use that baseline to do all of my other calculations ok if it's an unplanned event it gets stored as an unplanned event and it's used to calculate availability ok so if it's an unplanned event we have to store it so that we can use it to calculate availability now getting that number is not always easy knowing whether it's planned or unplanned is not always easy so that's the availability part quality is always the easiest one to calculate what percentage of the parts that I produced were good parts performance is another really hard one so performance takes into account a scheduled time so how much time I was scheduled for that shift but remember that number is always moving based on planned downtime events so as I accrued downtime planned downtime events during the shift it gets subtracted from the scheduled time but its scheduled time times standard rate how many am I supposed to produce equals our target and 2nded to calculate performance we divide actual that is what did we actually produce divided by what our target was actual includes good and bad parts bad parts are only ever applied to the quality number good and bad parts go to the performance number they both get added in together because it's a total amount you produced at the end of the day we end up with three numbers we end up a percentage of our availability was I was available a hundred percent of the time I had no downtime never happens ok this is always the highest number we'll say I had 100% quality generally we see quality numbers 99 whatever and then our performance number we'll say was 50% so that is I produced 50% of the parts that I was supposed to produce which means I was really running the Machine too slow or my standard rate is too high when I multiply all those numbers together I get an OEE number of 50% here's why you need to be calculating oae here's the reason you have to be calculating ohyeah and why it's not an optional number okay it has to do with the fact that you want machine learning an AI to make decisions for you and send those decisions back to you and the reason you want to use them is because they can process a lot more variables than a human can write one of the key numbers that machine learning and AI needs is OEE the machine learning and AI algorithm needs to know is how efficient am i operating for two reasons number one it can make a recommendation to tell you whether or not you should change that by standard rate your standard rate is not real it's not an optimal standard rate it's unrealistic and there's a lot of people watching this video right now that'll say yeah we have standard rates that are unrealistic we say we should produce X number of widgets per minute and we know that we can for a lot of reasons the raw materials are defective whatever the deal is so machine learning at AI can tell you whether or not your standard rate is realistic it can tell you whether or not you need to change your maintenance plan you need to do shutdowns more frequently you need to do preventative maintenance more frequently it'll tell you whether or not you need to change your q IP your quality inspection plan okay you are not doing enough inspections in line on your line to lower the amount of waste that you're producing and it can also tell you which of your operators are the highest performers you have opinions about whether or not they're the highest performance or not but they tell you which operators the highest performers but more importantly which shifts are your highest performers which raw material vendors which machines are your highest performers so when you know these numbers this efficiency number you can use this overall efficiency number to make determinations about whether or not I should invest an additional ten or fifteen thousand dollars in my next machine build because that one time that we spent an additional ten or fifteen thousand dollars building the machine it yielded eleven or twelve percent additional OEE overall equipment effectiveness which translated into plus X gain and overall profitability for the company these are all numbers that you think you're calculating these are all things you think you have a good grasp on that you don't and the reason why is because you are not grabbing that information from the edge and seamlessly integrating it up through the stack to be consumed and produced at all layers and then sending that information back down to all layers of the business for decision making okay the reason you have to calculate OEE is because a you don't know how well you're running until you're calculating OEE real OEE this real calculation number one and number two you can't achieve the Holy Grail without that number you have to have this number in order to be able to achieve this Holy Grail this closed-loop integration of your entire business okay and we hear it all the time and in fact we had a meeting this week with a potential client we were talking about the value of capturing OEE and a short time to value as part of a complete digital transformation project for their entire business okay and many times during the conversation the potential client said that they didn't care about this oae number that this isn't what we cared about we want this and so this video actually is sort of a response to that meeting what's crazy is the potential client they were all really really smart guys guys and gals they were very very smart they got it they understand what it is we've been preaching for the last I mean what I've been preaching for the last six or seven years of my career well we've been putting in our videos for the last year they get it they're not one of the clients who thinks they know everything and that they don't need to optimize their business long term they know they need to do this and they know they need to have a vision and they're taking all the right steps to get there but they did say they didn't care about this number and it got me thinking we have to let the audience know this is not optional this number is not optional you may not want this in a pretty dashboard for your whole plant to see but calculating this in real-time calculating it accurately and getting it into the hands of the nodes and people who need to consume that number is not an option and that is the reason that you have to calculate Oh yi all right thanks all right so this month we're gonna be shooting videos on what is IOT 20/20 we're going to be talking about the manufacturing Holy Grail which will be a reference to this structure here I'm actually sketch the whole thing out for you we're gonna shoot why you need OE which we just shot and we're going to do an example of industry 4.0 versus industry 3.0 so an IOT implementation versus an industry 3.0 implementation it's two of our clients two projects that we've done in the last couple of years and we're going to go ahead and just oppose the reality of one versus the benefits of the other in the last video that we shoot this month if you enjoy the kind all the normal stuff like subscribe share but what I would like for you guys to do since this is the very first video we've ever shot in the new studio using the electronic whiteboard and the whole deal the neon markers and me facing the camera the whole time and which by the way it takes a lot to get used to this was definitely weird for me to do it this way I would like to know what you guys think of the new studio and this new setup do you prefer the low-tech whiteboard environment or do you like this environment better maybe you need to see a few more videos to make the final decision but I would like to see comments there a couple other things I'm going to touch on and just like an ad-hoc video this month we're gonna talk about some of the comments that we saw this month based on the content last month I'm going to answer a couple more questions about is OPC UA the future of IOT etc etc and then one final video is going to be sort of my commentary on the way that we do business the way that we go to market versus the way other companies go to market and just some of my observations about the current state of industrial automation and the industrial Internet of Things which should help to shed light on some of the challenges that everyone is facing trying to decide what product to use what architecture to use and what partners to choose so anyway that's my spiel I'm sticking to it and that was one hell of a warm-up boom [Music]

Channel: 4.0 Solutions

Views: 11,863

Keywords: intellic integration, walker reynolds, intellic, iiot, industry 4.0, digital factory, oee, overall equipment effectiveness, how to calculate OEE, MES, manufacturing execution systems, plant performance, continuous improvement, oee training, what is oee, oee calculation case study, oee calculation example, overall equipment effectiveness calculation

Id: cFCtT71be40

Length: 14min 45sec (885 seconds)

Published: Mon Mar 02 2020