Error Free Performance
You know do what you do so well that they will want to see it again and bring their friends become advocates ken blanchard causes raving fan and ps the net promoter scoring system calls them promoters these air folks out there just so wowed by whatever it is you're doing there talking about your services in your your your value proposition at cocktail parties that lunch is on airplanes and hotels and I know exactly who used to do business with you get and they're actually handed out your car your website or your numbers for you and your phones ringing and you're like wow uh thank you so that coupled with a lot of the online and social media type opportunities we have today is a great way to use creativity before capital all right I mean I used to have printed brochures they weren't free you know and then you had to mail him out that wasn't free I would be out I would be mailed fifty thousand brochures and one mailing that's like dropping brochures out of an airplane it seems like peopl...
e are gonna read these things and now they're going to sign up for the course you know um these days a lot of that stuff paper listen a lot cheaper so it was a take a big gulp to do that kind of an investment to get people to come to seven years fifty thousand grocers at once is not cheap now print it, tio bulk mail it okay and then tow until then to deliver on it. All right, so the purpose of this session oopsie is to go back to domestic and focus again on improved so we're going to do another point, guys, in this case we're going to take that wobbly catapult we're going to figure out a way to stabilize it, try about the variation, improve the customer experience because right now the customer's not getting up a precise, timely, uh, accurate delivery think of the catapult to rep represent perhaps the diamond department around one or two where we was we were getting variation in our and our diamonds all right, same same idea prove our accuracy, precision and our performance in the results and again this in this case use the status post toe bring the lessons to life so that's where we're going, we're going to shift back to the point kaizen from the flow kaizen not great after this particular point in the process and, uh and nail it all right, so that's, where we're going to go with this? So we pull out our domestic and we looked, we now we go to the analysis phase and we say, well, let's start with the top beauties, those top undesirable effects, what is it we're tryingto to eliminate or reduce we could pull out our aipo diagram to learn that transfer function the input process output diagram what's what's going on with that, we could pull out a two like our fish bone charter ishikawa diagram and say all right, uh let's put it the head of the fish, the the variation in our distance and then let's go back and explore the different causes through the different fish bones. All right, let's, do it let's do a thorough process analysis to understand her voice of process and what's going on there. Let's, take that and then find the leverage points let's, let's find those, uh, those major leverage points stabilize those and see if we can't measurably improve our performance. All right, so there's, our wise a function of x transfer function when we say, you know, um take this catapult and, uh all right, we've got shooters because everybody shooting the ball the same way, you know, when we couple it with methods and procedures, we all following a very standardized procedure and answers no, we people did shooting at different ways, loading the ball differently, you know, pulling it back differently. So we had variation in our way had variation and our people because not one person was shooting at the whole time we had six different people shooting it they said well, we could reduce that variation john by just getting rid of the other five and having kurt you with all the shots or kate shoot all the shots that's not gonna work business today infact I fly I fly every week pretty much so it's interesting because every time I get on a plane I see a new pilot that's like I'm not the guy that flew the gloomy last week the week before that so I'm seeing different pilots all the time that might scare some people doesn't bother me because I know they have very rigorous standard operating procedures and methods to take the variation out all right and operated above seven sigma in the airline industry on safety not on luggage but safety all right? They operated a very high signal level because it's tze life all right, so we've got to make our process robust to the fact that we got different people shooting it material what do we do about that? We're getting material issues here the rubber band uh the ball okay, the equipment what about the catapult itself? All right, any variation going on with that? Any variation and policy and procedure what's this, you know removed the rubber man and we hook it every time thing all about we really need that interrogate that you know our pullback method or load the ball method are released method our measurement method ok uh the environment itself does that have anything to do with it? The carpet the wood, the floor, the table, the temperature, the humidity we interrogate our process to the point where we're finding those critical ex variables that are propagating to the y variation. So like we talked in the last segment, identify the variation in your business wherever it is and it's it's everywhere to a large extent but sort of short sort through it and figure out where do we want to start? Think about it from the customer's perspective first so value in the eyes of the customer we've talked about that repeatedly and value is ahh it's a moving target but what they value on time performance is there variation in the time of delivery? Right? First time was the variation in the quality okay, did did they get what they expect? So there's variation and we want to make sure we pay attention to that starting with the customer, I always start with a customer and work back. All right. By the way, what kind of variation is the customer experiencing it's? Not just about variation from us handed off to them. What kind of variation are they experiencing and are we contributing to that? Are we actually making their lives more variable and putting them at risk? That's a great question think about your customers variation and sell for that all right that's again a that's a real partnership kind of question how do we find a win win so we mapped the current state and we did that earlier and we said wow as we look at that map um you know we got variation all over the place get variation in the way we're set up on the ball pulling it back we get variation in our release we get variation in our measurement system but could even propagate to our documentation our interpretation of that um our whole systems loaded with variation all right, so let's map that out let's call it out to be brutally honest with it. Okay, well, if we have ah statistical process control and play let's use that as a tool to evaluate where we're experiencing the variation by the way this is you know, if you're an automotive um if you were not a motor in the nineteen eighties you had to have this teo even bid on work. This is nothing new you want to do business in the automotive industry. You have to have everyone on your company trained in spc a minimum of eight hours you have to have documentation and proof that you're using it and your facilities or you can't even bid you can't even get business new business if you haven't got this play that's how fundamental it isthe so if the big best in class cos they're using this stuff maybe there's a reason why maybe we should pay attention to that and even if we're running a little water to person show okay, are we measuring our pro performance on a regular basis and learning from that to get better and better and better at whatever we do? It's it's critical all right, those owns all right? So we've got our mean our average we've got, you know zone c zone b zone a those air each typically those each represent one standard deviation so you've got six standard deviations okay across your three on each side of the main basically we know that ninety nine plus percent of our data is going to fall within those control limits inside those owns but we can now start to look at, uh, symptoms if you will that our system is not in control so we could certainly we could do this we could populate it ah control chart with our data from around one of the of the catapult and we'd start looking at things obviously if one or more points is outside the control in it we're out of control sometimes something funky is going on but there's other tests we can run here and now that clever software have these built in so you can actually run that to run the test for you and then throw a flag up when there's a when there is something that requires your attention all right but if they're second seven consecutive points on the same side of the center line we talked about that last segment um that's that's the likelihood of seven heads and a tail flipping a coin that's not that's not normal so check it out all right seven consecutive intervals on either side increasing are entirely decreasing the probability of that's less than one percent check it out so this thes tests are basically saying to us check it out there's something funky going on in your in your system you might want to check it out and know about it okay, so this is again this is where this is world class performance where we really get to know our system so well that we can predict with a high degree of confidence and accuracy what's gonna happen before it does and our customers will trust us when we were with that smart about whatever it is we do with that reliable okay and then you know when we have out of control and play some of the most common reasons for that are listed here we have inadequate s o p standard operating procedures that we don't have him at all or we haven't but nobody's following them we have noise variables which we talk just a little bit about early we have variables that are beyond our control that we haven't learned howto deal with or protect ourselves from that cannot that could certainly hurt like I said earlier we might have s o p is that we don't follow all right or we could actually have made true process improvement I think about that you mean we made things better and now we're statistically out of control and the answer's yes what that saying is that we need to now re calculate our control limits with the new data to read to now create new control limits because the new process improvement in relation to the old control and that says we're way off we're we're way out of control so I mean that could look something like this you know where we reduced our cycle time from this average to something less we reduced r variability from something that looked up like that to that that that could say what we're no longer in control using the same control limits but we're better than we were before so it's statistically in control or out of control doesn't necessarily mean good or bad it really means stable are not stable and we just want to know that so we want to know okay um this is again this is like uh checking out your uh your heart rate this is like checking out your your body in some way it's an indication of are we are we healthy in good shape? We're every at risk so it's again running running your life with good data personally and professionally right? So these are just some symptoms and summit visuals if you will of out of control uh performance so here's an example where it says I don't take action now because something weird is going on we've got you know we've got two out of three right here in zone air be so something weird's going on that's not supposed to happen the probability that's very low and check it out all right? Simple tests the questions for out of control these would be very important as you is you think about your business in your process and your customers and everything um and you've identified variation that's important sir donnelly think about variation from the customer's perspective again, what were they experiencing it and then start to ask yourself if you know their differences in our measurement tools or measurement systems accuracy something statistically we call messa measurement systems accuracy we're going on there if we go back to our catapult example um we had a real problem with measurement system we really didn't know exactly where the ball was landing so some of the variability in our process was was coming from our measurement system now what if we're driving down the road and a police officer you know, pulled you over and says, you know you were speeding I got one my speedometer said I wasn't well, I got you on radar and you were but my measurement system says I wasn't far measurement systems are inaccurate, all right? We could be operating with lousy data and, uh, putting ourselves at risk about methods are there any differences in the methods people using the pullback to release is the process affected by the environment in any kind of way humidity, temperature things like that other differences in the machinery, the tools, equipment being used? I think about this this should start to look like a pattern we've seen before, and that would be that fish bone. Remember the six m's so measurement method mother nature uh, machinery uh, man, man, power, things like that. So, um, this would be a great way to start to say, okay, well, let's, let's pull out a tool to go after variation reduction and gain knowledge, and this is where we could pull out the tool we talked about earlier that the issue cow in the fish bones, this is a great little six sigma type of cause and effect we've got one ut not ten like we had earlier that we're really going to go after here that one happens to be this variable distance where the ball's going, we're out of spec customers sending it back, telling us it's no good so we have a high pasta poor quality it causes all kinds of other things so has it got anything to do with material we're using the rubber band the ball anything to do with the catapult itself anything to do with the different methods we use the people shooting training things like that what about our measurement system? We just talked about that a little bit and in miscellaneous or mother nature we get into you know whatever else way think so where you gonna discipline method now to drill into what you know what's causing this variation and shifting that again from the I think I know so back teo improve now we've been through this once before this is just another little cycle to use it in a different kind of way we got the problem statements we've got the ideas and the solutions along with solution specs things like that the risk analysis and I'll introduce you to a new tool in this section I've talked about it of reference that thie f mea the failure mode in effect analysis is great little too especially if your business start up for an entrepreneur solo and entrepreneur with an idea and you want to be a six run a successful business you get you have to know the f mea that's absolutely key tool I'd call it a top ten if not a top five tool in the whole toolbox it sze that important and then kaizen will come back to hk eyes in which we've learned about and well actually won a little simulated kaizen here with our with our catapult so uh an example problem statement well, it could be something like this. The problem is that we have excessive variation our process this causes miss deliveries lost revenue excess inventory piling up high cost, two poor quality high carrying cost and negative customer feedback. The cost of this problem is x and this illustration a million and a half dollars annually. So this variable process that we have, whether it's the catapult, it's what we've experienced earlier with the lean sigma game is it it's a trigger a whole lot of beauties beauties in this case would be the miss deliveries the mist hits, which we saw yesterday as well, if we miss the delivery we missed the target, we don't get revenue for it fact we cost us more money now to try to fix it or scrap it or whatever we do with it. Meanwhile, we got inventory piling up so it's the same kind of problem statement we had before and we'd say to the guys in team I was pretty clear on what our mission is here the problem is what we're trying to do yeah okay, we're good so we go to solution specs alright, what kind of solution specs do we have? Well the customer could give us some customer could say you've got it, you've got to get your range down to twelve inches at the must that's a must have and you're twenty nine now so you've got a lot of work to do you basically you need toe you cut that range by more than half incidentally you have to do it um by the end of the week I'm going to another vendor the time could be a factor we gotta have this done fast we've got to really blitz it oh, and by the way, you have to do it without any money. So creativity before capital we're not gonna just throw a lot of money at this so tell you what we will do weave we found a few things that we might be that might be helpful with we found some duct tape you can't travel without duct tape you so you gotta have some duct tape on your handy you know, we found a sandbag that could be useful, we found things like aluminum foil, okay, so we didn't go out and spend a whole lot of money but way have a few basic things around here that we could get creative with and, uh but you can't go out by new catapults and you can't drill it to the floor and things like that all right, so uh we've got the customer specs we've got the at the time, the budget things like that uh and we have a way even have a bottle cap. So what do you know? Well, you know, we're going to do with that thing, but we'll figure something out. Okay, so, uh we got some, uh we got some resource is handy this's creativity before capital if you ever saw the movie, apollo thirteen there's a wonderful illustration of this where it's like we dump a bunch of parts out on a table and say, you got this much time, uh, to figure out how to solve this problem or people are going to die if you've ever seen that. If you haven't seen that movie, you've seen that scene it's it's amazing! Because that's kaisa and that's amazing it's amazing what creativity khun khun do sometimes. Uh, so we brainstorm alternatives. All right, let's, get our post its out. Let's, start right now on what? You know, what could we do? How could we stabilize that catapult? You know what we could do with that sandbag? What about the duct tape? What are we gonna do with that? All right, so let's start toe let's start to brainstorm alternatives and then a finito eyes those, if you will, so we start to brainstorm and uh we get these ideas you've seen this before so it's a great way in five ten minutes to get a whole lot ideas from the team to say let's just open brainstorm let's not try to get to over analytical here let's not judge write down with whatever comes to mind let's capture it and then let's affinity eyes it and put it in grouping so now we say well you know what one thing we could do is is partition and control those x variables let's sort through him and let's put him in tow uh categories it's partition um so some are going to be noise variables and some are going to be constant or controllable variables and some might even be what we call experimental variables I'll come back and cover that so we could we could start by doing that we could figure out a way to mistake proof the pull back well what do you mean by that? Well what if we could figure out a way to pull it back to the same one seventy seven every time even with our eyes closed what if we poke the yokes to pull back to the point where we didn't have to look we designed it into our process well that's pretty clever we could do something like that what about improving our measurement system? I mean we really had a lot of very was probably had at least a six inch very very variation right there so that's not going to help what could we do to improve the the measurement system and maybe one of these ideas is you know what if you're what if we came up with a way to put something on the floor so when the ball landed it pinned it it made a dent uh you know it told us that's exactly where it landed we don't have to guess it there's a big dent that wasn't dead in the carpet and it wasn't done on the uh the wood but I'm sure there's something around here that we might be be able to use that would show up like a ping in it and I'm saying, well, we'll have to look around and see if we can find something oh, there is we could probably stabilize that catapult you know what? We know how we're going to do that but one of those post its might say duct tape on it you know or sandbag or something like that and standardized that work we might wantto have ah, standard process that everybody follows to make sure that we're all using the same pull back same release uh technique, that kind of thing, same load the ball technique matter of fact we might even look at our ah process map here and we might discover that uh we don't even have things in the right order necessarily unnecessarily okay because I'll tell you what if we uh we load the ball and then we pull back what's a potential failure mode there we dropped the ball blown up sir so you know what if we wait we pulled back first alright pokey poke this whole thing and then we loaded the ball that might outsmart the system so to speak so we start looking at our process and say we might not even have a curtain I talked about this at lunch time in a a major project he was involved with some years ago where the way we sequence the work we might have think about a sequence it might be things we can dio in parallel or in a different sequence that would make the process a lot smarter okay, so those are those are some thoughts that they could show up in our brain storming all right, so uh we've got all these ideas this is the fun part of kai's in because I think that people sometimes people say well I'm not creative I'm not a creative type what was the one of the takeaways this you know earlier was what we're whole brain we're not one or the other so so yeah you're created he just might not realise it or nurture it or tap it yeah kurt uh you know, one thing you talking john boy this morning was to really think outside the box when you taught me about diamonds actually being circles, I'm thinking that way now and I'm looking at the variables that we have here and I'm wondering if getting rid of the most number of variables would actually be moving away from a catapult entirely having a person takes the ball and place it at that point they're still using the same inputs on you're getting extremely controlled output that's what I was thinking subtraction, why don't we just get that whole catapult right out of the way? You know what? We've had some wonderfully clever when we do this in a workshop that's um you know where we have teams and we like to compete set up little competitions, we'll have the teams go out and I'll actually brainstorm all kinds of stuff we have to set certain rules to keep the thegame honest, so to speak because we've got people trying to shoot the ball up against the wall and drop it into a funnel which goes into the cup it's like watching miniature golf or something it's like, well clever creative it's beautifully it's brilliant but no that's not allowed in this long lives not so much of a question is more but thank you for getting me to think really different oh good yeah good that's the idea of this whole thing this whole course thank you appreciate that kurt could good feedback so here's mory partitioned the variables this is where we're staying all right? You know we got that why output but we have now taken the ex inputs and we've partitioned them, so we've got variables that we can control and then we've got variables that we cannot control all right, the noise variables and then we could also start to identify some very variables that we could actually then use for experimentation. So this is where we run designed experiments so uh, the key here is if we're going to run a designed experiment, which is what what we do when we intentionally mess with certain variables to see how it impacts the output. So this would be like if I'm if I'm bacon pumpkin pies and I'm trying to reduce my cycle time to get more pies out, what might I be able to do to reduce my cycle time but still maintain good quality? And we might start saying, well, what if we turn the oven to temperature up five degrees and reduce the amount of time in the oven by three minutes with that correlation of variables, get the pie out sooner and still still with the same integrity and efficacy and taste everything that we want look so we could start to manipulate variables to see what happens all right, we could start to manipulate pen heights and pull back toa tio to predict with great accuracy where the ball's going to land because in this in this round of the catapult we have much more advanced grounds and this round all we're trying to do is stabilize the process and get the range down to six inches from twenty nine but in ah uh in future events I might say well, I haven't told you where I want the ball the land what if I told you I want you to hit one thirty five with your first three shots then I want you to hit one twenty two with your next three shots then I want you to hit ninety five and one twenty five oh, man, how are we gonna hit those numbers? Well, you're gonna have to run a little regression modeling or design of experiment toe to figure out what to pull it back to it. What pin height t hit that target and then so you're you're purposefully changing your ex variables to hit your white targets that's that's really intelligent business management and systems and, you know and that's what the that's what we're seeing in the military and that's what we're seeing in a lot of world class companies is they know if I tweak this and I do this it's going to do this they know that's the knowledge transfer transfer function they know they're not guessing they know so um at what speed think about this the next time you're on a on a flight the pilot comes on and says all right, we're taking off now um our flight time we should be arriving at two thirty eight this afternoon that's based on the headwinds that's based on the course that we're gonna have to take those there's a storm up there and we have to go around it um and when you land it's two thirty eight they know that they weren't winging it now they might win it at the counter when you walk up there and there's no plane and they say you're leaving in ten minutes but I'm talking about once you're on that plane and your taking off that that's tightly controlled all right? So this is what we're doing with partitioning the variables and this is what we'll need to do of course with our their exercise incidentally, we don't mess with these until we have a stable process until we have a stable pumpkin pie coming out of the oven every time. If we start messing around with up in temperature and things like that, we don't know we're confounding things we don't know if it's the oven temperature we're messing with or some other variable that way we don't have control over that's causing the change in the output so step one is always stabilized the process so that it's predictable it's repeatable and then you can start to interrogate it a little bit, tweak it to figure out a way to reduce your lead time, reduce your, uh, loss rate, improve your wind raid if you're selling something solid, you know, selling more that that it's running your system, your process, your business with intelligence and knowledge, not harder work and more effort. So we find our best option. Hey, maybe it's a combination, maybe we should stabilize the catapult and pokey yoke the pull back, not one or the other, and we should probably do a little measurement systems analysis, improvement exercises. Well, so this isn't just a single point. It might be a point the process, but there's several caissons we want to make we wanted make good change to our measurement system, make good change to our procedure and our method, and at the end of the day hit the target, shoot the ball in a cup, so to speak. So what are the forces for all of these things? Where the forces against it, you know, this the forces against could be, well, budget, we don't have any money, so we're going to have to come up with some something else, okay, we can't drill the catapult to the floor there's policy, all right? The create apply folks don't want us drilling the catapult to the floor and screwing it down that could certainly help but not the stage so what you know what kind of forces against against this we have and this is a key takeaway I urged people to because we always want to know what are we up against to be successful if I want to know if I want to write a book but I don't know how to write a book that ignorance that lack of knowledge is a force against me I got to figure out a way oh I do that get back to our top girl there j k I bought a book uh on how to write a book I read a book I read a book on how to write a book great book by the way two but it walked me right through everything I needed to know and then I continue to make a lot of mistakes but uh I went to a conference actually in the orlando, florida this was after my second book was published and this was a conference on book publishing. I wanted to learn more about book publishing and copyrighting and legal and distribution and all of this all the the book industry didn't know anything about it so I signed up for a conference it was down at disney was very cool I went down there and I brought brought down by the two books that I had published and I learned in that conference that I did just about everything wrong except I sold out all my books s o the experts were telling me about everything I did wrong, the only thing they couldn't explain is how did you sell out all your books? Okay? Because they were second printing already at things like I didn't have bar codes on my books on the covers, but you can't sell a book without a barcode on it, so they had the label them all and things like that and, uh, you know, I couldn't distribute through certain channels without barcodes hoops don't miss that one and, uh so it's, a lot of mistakes, but you learn from those mistakes, and then the reality is that the book's all sold out and the first book pulling together has been reprinted now by four different publishers and a couple different languages and sold hundreds and hundreds of thousands of copies, so something worked, you know, but, uh, ignorance was a big factor for me, and I had to figure out ways to overcome that this where our countermeasures come in so you know what? The forces against us back to that tool let's, come up with the best solutions these could be separate projects are all one project, so the project would be we need to stabilize that catapult doing measurement systems analysis tighten up the measurement system implement all these changes to a kai's an event to prove that out on demonstrate that it's better so that's actually what we're going to dio this afternoon so here's the f me and I said this was a top ten tool in my in my mind so in the f mea we list the component or the step uh where there could be an error so this could be uh you know a jet engine what's the failure mode what could go wrong? This is where we at what could go wrong? It stops working you know catches on fire something like that what's the effect of that well okay um that's not pleasant and it could be very serious so this is severity to pickle fm a use a scale of one to ten or one two five aye like one to five and so we rank this a five would be someone's going to die or somebody's going to prison it's very, very, very serious ah one would be it's it's noticeable it's it's aggravating but it's it's it's not really that serious what's likely to cause that can we didn't dig into the cause of it always oh currents so how likely is it toe happen? Is this a five it's it happens frequently it happens all the time is it a one? It hardly ever happens what's the occurrence detection is how likely are we to know when it happens like right away so in this case way reversed the numbers a little bit because the higher numbers are high risk so if it's going to escape us, maybe it's going to get into the market end up in a recall and that's going to cost us millions of dollars or lawsuits are all kinds of very pain? Okay? It got past us, we didn't detect it that's a high risk that's a five if we detected instantly, we've got a visual control system that kicks it out to something like that we're going to know instantly that it's it's not right that's a lower risk because now we can least kick it out and deal with it. So this is a risk analysis in mitigation tools, so r's risk sometimes referred to his r p and risk priority number and I risk priority number is r s times r o times are d it's a number and it just helps us prioritize all the different things that could go wrong with whatever it is we're doing so drop the ball blown up, sir that's serious that's gonna that's gonna crush us in terms of our process or performance with their customers that's a serious thing likelihood we didn't see it happen in round one but let me tell you something happens a lot happens in almost every workshop I do when we have multiple teams out doing this it's almost always some team that drops the ball once or twice it does happen a lot I didn't do it on purpose it just happens all right detection well if we dropped the ball going right away so that's a low lower number but the key then is to come up with a foolproof plan the fool proof pants says well how do we make sure that doesn't happen can we design in or polka yoka a uh a solution and this might be where we say you know what? Why don't we do this? Why don't we have the operator pull it back and get it to the pokey poke the one seventy seven and have someone else actually load the little dismissal or load the ball so to speak that takes a lawyer a ah huge potential for error could we do that? Sure fact you'll notice a lot of times in the military uses that strategy where it's a team approach it's not one person loading their own ammo a lot of times things like that all right so here's the scale of this is in your materials where you can see that one to five scale and run fm is it's nothing nothing teo too tough fact we use sometimes a very simple version of me which is just three questions what could go wrong so I urge anybody at home to just released run a simple f emma and your business. What could go wrong list there, all the different potential failure months? What could go wrong? Um, I don't I don't. I don't get sales, okay, I get sales, but I can't deliver. I could be making promises I can't keep because my process is messed up. So what? What could go wrong in your business? There's a lot of kinds of things, okay, and what's the risk. If it does go wrong, how serious is it? And then the most important question really is, um, what are you going to do to make sure it doesn't what kind of ah, what kind of a solution are you going to come up with? It helps mitigate or really reduce that risk. You can't make your business is completely risk free, and every entrepreneur knows that. But you could certainly take out a lot of the risk, and wise business leaders do that and central preneurs.