Design of Experiments (DOE) for Injection Molding

my personal goal is to provide easy-to-use tools for molders that's one big thing that I want to do right there's this fantastic research being done there isn't acne fantastic technologies out there lot of the stuff but I want to somehow translate that to the production that's what I've been trying to do and of course to have more does make data-driven decisions right so with the software which I'll show you which is the view software right when I when I first said I want to make my own Neri software because I was using Minitab which is a fantastic software right for those of y'all who use me that you know yes know the capabilities of that it's a fantastic software but it's not easy for a molder to use okay and so I said how would I come up with something simple that anybody without any math background can use it and that's how the software is today so when I started looking for somebody I found Robbie right which I'll introduce here right so Robbie is my partner and the software is mechanical engineer by trade and he has specialized in data and analytics and quality systems the last 30 years he's actually online and so when I met Robbie he even being a business man he said show us why do you want to develop a software he did not say okay let's do it together says why do you want their episod but I said I wanted to he says because there's so many of them out there I said I want to develop my own software only because I want to make it easy for me right and that's how we became partners in this stuff and is in Caesar then now I have known Robbie the software is whether were 20 years old and she moved on that and I've known driving for the last 12 years and we have had a fantastic relationship on this right so what are the typical so let's get into presentation right there just introduction there so what are the typical problems of a molding company and I'm sure whatever relate to this definitions to constantly adjust the molding process parts have to get sorted out at the Machine the parts they need to get sorted out by cavity don't tell me you guys don't know this right sometimes cavity for is the one that's bad so it was sorted and float away measurement frequency is very high possible customer returns right that affects a reputation you will rerun the parts all that stuff right to legend you're frustrated because it had steel to the two there's only some of the problems by the way right this they had the tool done correctly but then you said no no no this part is now too big so they go you know she is steel but now you to go back and weld it it's some typical things right parts were good CPUs are acceptable acceptable during qualification so you qualified the more fantastic 1.67 CPK now you go in production my god now I see because they're pointing 97 what happened here right so all these things are typical problems in the production floor simple question if you want to drive your car in cruise control mode at 80 miles an hour right with no stops for four hours let me just want to keep driving Yeah right which of these roads I would you prefer the answer is very simple right you want to basically drive on the edge of a cliff or do you want a nice wide freeway like this right so of course the answer is obvious you need to have a nice wide freeway so you can have careful carefree driving is your left limit right right limit you get a phone call they won't drink water or something you know a bottle or whatever it is and if waiver a little bit this is going to be okay right so the choice is always having a nice wide freeway for cruise control and I say there for molding processes should also run on the cruise control mode and also should have a nice wide freeway right we do not want the process that is at the edge of a cliff right and therefore molding processes they must be robust it must be repeatable and they must be reproducible all the three hours are extremely important okay now having said that if the processes on the cliff then all those problems that we talked about you're going to see all those problems all right so robustness is one thing I've talked about and I said the process should be robust so then you say that I mean that process should produce all the parts under normal variations in the production environment okay so if you see what is that if you look at the if you look at what does it take to mold and cart right this little pad to that right so each of these things is gonna have variations coming from sorry the process is going to have variations coming from different locations machine the personnel the process and all these things and at the same time each of them are going to have their own fishbone diagrams right there's going to be a should process this is going to be a shin the measurement system so that's what I do gauge R&R it's going to be the variation the viscosity of the material right so all these things are going to happen so they all going to have their own fishbone diagrams now apart from this variation is also remember that to come up with a good molding process there are five elements that are important part design the material the mold design and build the molding machine and then last is the molding process right so all these elements also this is pre-production or pre molding okay all these things are or also these five elements are also important it's like a relay race right all the runners in the team have to run fast enough to win the race so this is very important you know so if you look at the aim of the molding operation what do you want to do you want about parts consistently you want to make the mold all the parts that meet all the quality requirements and you want to run an efficient process consistency three types of consistencies the first one is a cavity cavity consistency second is a short the short consistency that means every shot has the same dimension and then you have got a run tool and consistencies consistency that means you more 100,000 parts you change the mold about something else would you know change the mode back to the first mode if this quality of the parts and this quality part is the same that means I have mentioned the variation that you had in the first run is the same as the average and the variation that um the second run then we say of also the achieved at the rental inconsistency this is the most difficult consistency to achieve for Lord lot of molders okay bye the main reason is the word I use something back robustness the process was not robust to begin with it is not repeatable is not reproducible okay of course you want to meet all the quality requirements you want to have a good efficient operation and you don't want to process the funds like that's right so this is where you see a lot of loss and efficiencies you see lot of returns and so on and so forth so something like this is not is not acceptable all right so going back to this the source of variation can come for a molded process part it can come from various locations however we are going to focus on the process right so this is the first one that you show you the process right so now if you look at the molding process what are we where where is the variation coming from number of things but we are now going to be talking about the process conditions for the molding process parameters right how many are there I have identified them as eleven plus two molding parameters in the motion machine okay and eight outputs so these interval plus two I don't say thirty because if you don't split the back and hold in to do and lot of companies don't split back and hold into two then you have eleven and those if you do spill them to back and roll then you have thirteen parameters oh they're right and decompression is there it's a question mark again I cover this in my four-day class I teach a whole board a class about it so let's not look at this but I say that is 11 plus two morning parameter so variation and this is going to have is going to get keyword variations in there the quality of the molded product right so here as I judge the complete molding cycle therefore in order to have a good robust process and to have a good cruise control process then you need to optimize a molding process right and so what are what do people do it look at the process development procedure right now you have got the injection back and whole cooling during its cool recovery takes place this is really a complete plastic conditions that you only you need to optimize what the molders do the first thing they do is the mental patient studies I don't know a lot of people do this since now but again there's something that I've been teaching for last now five six years is that as a man preparation study so you do med preparation screw rotation speed study a back pressure study for coming up with this part of the cycle and then you get into the scientific molding side or which I call is a six step study the only reason I call it six step study because so we know how many steps there are it's nothing yet it's not that I want you to come up with something else the term scientific molding was coined by mr. John Roselli and the folks at rjg right so and these are the six studies for that right so people will do these studies and they come up with the the process so here are some graphs from that well preparation study the six step studies viscosity cavity balance pressure drop process window gate scene studying and cooling study so they do these studies and then they come up with the molding process most companies right so here is the molding process that I've come up with after I have done the Center in Buddhist studies okay good so I have a process sabi done the answer is no right why is that because what about long-term variation how the dimension that we measured there are they are they on the cliff right lot of people will do a CPK study right there the dimensions get see but it is flawed because if you take thirty parts back to back and do a statistical analysis chances are they'll be extremely consistent but that is not the representation of your complete production run in the future or the next one that you're going to get right so other dimensions you measured are there on a cliff are you going to be fighting dimensions and cosmetics at the same time yeah I'm going to give an example of this right are you setting up realistic alarm limits right alarm limits very important you say that along should the cushion values plus 0.05 inches then you get at point zero five from the pressure values should be plus minus 100 psi where do you get knotted psi from right so there has to be some data-driven decisions you need to make data-driven numbers you need to come up with if there's a dimensional issue what should I change out of the 11 plus 2 parameters right as an example look at this right the dimensions below the lower spec limit so if you want to make a part bigger increase the dimension of the part then if I say go increase the dimension part what will you do possible solutions are you ready is a mole temperature you reduce the melt temperature you increase the pack pressure you increase the whole time you increase the cooling time all these things will in theory increase the dimension of the part and therefore we need to understand the effect of molding parameters on the part quality which is the one that is going to give you the biggest bang per buck when you go change is it melt temperature is it more temperature and there's no definite answer because it is far dependent it is dimension dependent and it is material dependent okay all these answers are correct but they're all dependent upon the particular things I just told you about okay so we need to understand the effect of molding parameters on the pot covering and that is what is design of experiments we play with the process what do we do let me go change the coding time and see what happens let me go change packing pressure what happens right so we played the process playing with the process and systematically finding the effects of what the what the effects are is basically what is called as design of experiments right so if you look at definition it is a systematic set of experiment that involves changes to factors of the process to study and recording the corresponding effect to the responses on the responses right so that is what a URI is right now do is is not nothing to do with injection molding only by the way right it has been around for a very long time all right so I got this from online from psu.edu where you can see there how the the history of you is came about right from our efficient in 1980 to 1994 TZ use it in agriculture so studied you 'yes you probably already know this right then afterwards as the first part and they use it in when the war broke out they started using it in the world increases and all that stuff and then he party to mister Demings came about and he started implementing the good to the quality systems and all that stuff and now because we are using it more and more and one of the reasons one of the reasons by the way the biggest one of reasons why the covert vaccine is getting delayed is not because they're not able to make it but the testing is going to take a long time and how do they test it they go through a lot of neo ease they have to go through a lot of factors thought of variation you know underlying conditions and all that stuff right so do is have been used in several industry yeah my favorite wine is Yami Pinot Noir okay so if anyone of all want to send me a gift that's what you need to send me so if you look at this right if you look at the this this wine in in 2014 and 2018 I got this from the websites you can see that they're all blended differently from different grades right while you do that in order to maintain the consistency of the taste okay and once you have that consistency when people start liking what it has right is that taste tell you to keep on maintaining it right and that's what gives them success right in fact I when I was doing this research I found that this company offered just one brand of wine right just this one was sold for three hundred fifteen million dollars so there's a lot of money but they do they use this in the in the wine industry food industry everywhere yeah so why don't would so if it's so useful why don't more does use its often because of the myths misconceptions right number one they say it's too complicated one needs to understand math and statistics right then they say it takes a lot of time to perform these studies there's a lot of waste of material resources it takes a lot of time to measure all the cavities all the dimensions are all the runs then I need to learn how to use a software right and interpret the results yeah I mean Minitab is a fantastic software right I mean if you guys know how to use it more power to you guys right but always say for what we want to do you don't it's like taking an airplane to the grocery store right you want to go to the grocery store you don't need to go buy an airplane right it's just like that right so use something simple right so I then other thing they'll say you know what I don't need to do all this new ease because I can come up with a same process by doing without a new e ok yeah but this you can't come same process but there's a lot of other things that that the benefits that you get which I'll talk about here they're useful right so for all these reasons people don't do new ease right so let's go to molding numb right so effect of package if I want to study the effect of packing pressure then if you see here if I increase the packing pressure from 4,000 plastic psi to 8,000 plastic psi then the dimension goes from one point seven two six to one point seven thirty-five what does this mean an increase or decrease in the packing pressure oops not 4400 it should be 4,000 and increase or decrease in the packing pressure by 4,000 psi increases or decreases length by nine thousands of an inch right so what did we do here they systematically went and performed two experiments one at low pressure and one at high pressure right and looked at the dimensions and part of the results right this is a dotted line that you see here because this relationship can be anyhow okay there's always a curvature there if I have the time you can talk about it especially in molding there to be very careful because all there's a physical limitation to the tool you cannot keep on packing in and dimensions keep on you cannot go on increasing right because of that thing right so this this relationship can be anything in there and so what have we done we have done two experiments one low and one high but now I say that so this is the base most basic nearly right because your sister Mary done these experiments then I say that I want to also now look at the effect of men temperature on this right I did packing pressure first so did melt temperature second so what do I do I repeat the same high and low experiment with the high and low the melt temperature so I end up with four experiments then I say I want to add more temperatures and now I add another for experiments because I'm going to repeat the same for at the hind or the other one so I end up with eight experiments right bad one more to it it becomes 16 experiments right so as I keep on increasing the number of factors or processing parameters the number of experiments keeps on becoming double becomes studying low and high they're called as levels at two levels it becomes they keep on becoming double right in fact by the way we have all those of you all those do scientific modeling studies all if you have already done a de right look at your gate scene study part wait was this time what have you changed your system and he changed at times and ended up with and collected the part weights that's an AoE okay so all these things are do is also right so it's changing xi plus two factors and coming over the dimensions and finding the dimensions is basically what is called as a designed experiment right but here is a problem is that I've got one factor and two levels end up with how many two experiments right if I go to two factors systems for then that becomes eight right that becomes 16 if I now decide I want to do three though medium and high and get three experiments and then if I decide to factors becomes nine experiments right so as you start increasing this you find out the number of experiments that you need to run is it low number of levels raised to a number of factors that's a formula so now if you do that calculation you will find that at 13 factors at two levels you end up with 8192 experiments and at three levels you end up with almost one and a 1.6 million experiments which is impossible for you guys to do so what do you do in this case that is one reason I think a lot of people have not one more reason I should say that you have not done they don't do these industries as - to me factors and injection molding first thing that only 11 plus 2 by the way okay I've heard numbers like when I start my class people say how many practices that we practiced what if I doesn't know its just 11 cuz two factors okay so if you say this to me factors right on the stuff it has the first reason people don't do the you is one of the reasons that s that they don't do it right but that's becomes an impossible job so what do you do in that case I'm not going to get into any of the details but this is where where our friends the statisticians they come into a picture and people like Robbie who do data analytics they come in the picture and say okay you know what it's not that there's something called as orthogonal arrays and something called x' interactions what we do is that we take all these things and confounding their you know and then basically now instead of running all those experiments 8,000 experiments you can run few of those experiments and those few are now going to predict what the quality of a part is right and so that is that is what statisticians do like I said I'm not going to go into details of this before the time limitation here I do it teach some of it in my class if you want something really detailed of course we can come and teach that we will for you guys also right so they do these things called as fractional factorial experiments and there's several models out there for you is by the special methods that which you method so on and so forth right and you can do all this analysis and then come up with the what is the factor that most are most important for you okay like I said Minitab is one software that you can use Nautilus software which I'm gonna show you here which is which is which I'm going to show you guys right but then how do you write how do you select the factors and I'm sure you guys are interested in this right how do you select the factors right so I can't run these man experiments so first thing is that you understand you're the molding setup everything comes from the MEP diagram you've got a pressure volume and temperature graph okay that's what you do right in the injection molding what are you doing you're taking plastic and melting it so giving it temperature you melt it and then you pressurized it and inject into the mold and as it is being injected inside the mold inside the mold its part is cooling down so the change in the volume which is shrinkage right so anything that is affecting the PBT is what is extremely important in the theories okay now now let's so so if you look at it right mainly it is the pressure volume temperature so pressure and volume volume is related to the the pressure all sizing right all these things of course related to these things right so pressure volume and temperature are important and therefore basically if you look at this we do our scientific modeling studies and we start looking at eleven plus two factors in this case I have made only eleven factors keep it simple and I put shorts and transfer together so you see them as ten factors here so if you look at this right how do you select a shorts are in transfer position it's fixed at 95 to 98 percent flow so do you use it for do is no you don't give it for your ease because you want to keep that percentage for the same please find out injection speed into this caustic curve and you select an injection speech so do you want to go slower no you want to go faster no right you're set your injections to viscosity curves the injection speedster injection pressure on closed loop machines is fixed machine takes what it wants right from there okay it takes what it wants from from what's available to it so if you change that up and down it's not gonna matter I've called as the holding pressure as this is a she pack and hold pressure yes uses for new is very important for tea time is fixed for cold run assistance for hot on assistance you may have to use it holding times use it for you is yes because it's a cooling rate when temperatures are yes one or more first materials more temperatures I guess for crystalline materials okay and again why is that covered in the class we have set ask rotation speed based on the cement melt temperature study the pain the homogeneity and we also set our back pressure is based on scientific back pressure study right so this is a date based on data we have set those numbers there and therefore if you look at it you can basically factors that you want to study for use for you ease it's basically always your back and hold pressures all these at time always you melt all more temperature effects and organs you may have to this word both make the word important is me you may have used both of them and may be a holding time if it's a photo assistant and therefore in most cases if your record on a mold right you can 3 factors is more than is more than sufficient right and therefore like I said you need don't have to do all the 1.6 million or 8,000 experiments you can get away with man-made experiments and that's gonna give you all the low-hanging fruit that you need right is ingestion speed important in some cases yes right it will have to go on case to pace case basis but in most cases no ok it's back partially important for doing do is absolutely not okay why is that because the effect of back pressure let's say on scale of 1 to 10 is 2 whereas packing pressure is 8 okay so what do you look go change that means Russia in that case right so having said that I'm going to quickly go to my my software it's gonna be very quick there's a nice 23 minute video on my website that's a short version that's how powerful my software is but I'm going to go to the software here and show this to you guys and this is my software there's a home screen there's a database here all the databases on the project you can see all these icons out here meditation studies six-step study this anything bowling studies plastic process sheet and so on and so forth of which I'm going to go into the AoE and will show you how easy the do a portion have designed this air is you can see you in the do e the first window is with a drop-down list you select a factors cooling time Mel temperature holding pressure of course in type l1 in here but notice this I only had five factors you cannot add more than this is it easy for for us to put say all the eleven factors here yeah but we don't need it I want to keep it simple for you guys okay this is step number one step number two put in your highs and lows numbers what do you want to study i want to study a holding pressure from 800 psi 2 and PSI because when I did my process window below 800 I got short charts hundred I got basically flash so these are my this physical limits in here right I'll tell you what this position number here is soon so this is two number three but in all the dimensions that you want to check in nominal values and tolerances sample sizes this is not statistical analysis very important very important you just need to check three to five parts that's it you don't need to check 30 parts okay if it's a family mode then select to which part it this this dimension belongs to and you simply click on worksheet that is it that's how simple the software is guys okay it spits out these eight eight experiments here you can see there's eight runs we do do full factorial experiments three practiced two levels so that becomes eight experiments you don't want confirmation run and these are the basically the eight explains that you run here plus one nine or the molding machine run your parts collect all the data put all the data here in here you can also collect on Excel Sheehan and dump it in here if you want actually preferable and then so experiment one through eight seven sample1 sample2 sample three four all the dimensional you see here and then you simply come in here and click on analysis that's it right like I said I've achieved my goal of making life simple for for the molder okay so what does it give you I will tell you the most important several tabs here on the top we can see will tell you the most important tabs donator chart what does it tell you the most important factor for that dimension that that cavity seats as you see for interest as the most importantly okay so now if i tell you go change the dimension of the part what are you going to change even go change holding pressure are you going to touch me cooling time no because kody that makes no difference on that part so don't you touch the cooling time so now see are helping you're making data-driven decisions on this based on this okay then you come to the process windows ok because we know melt temperature and holding the ones that are important right this is my cosmetic process window that means inside this I can make good parts below 800 I got short shorts but about 100 I got bad parts right so now if I want to it nominal on this dimension this solid green line that you see if I click on that anywhere I click on that it tells me at five thirty nine sixty two and eleven point five seconds I will hit nominal on that dimension this is your data this is your machine this is your insistence the softest when you're analyzing it okay how accurate is this will show that you very soon here right so now that means what I can come in and draw this window inside this which is the dimensional process window inside which you know all these new old parts that are to spec so if I maintain my temperatures between 486 and 544 and 834 and 1060 I'll always make parts that are going to be mystic and that becomes my dimension process to you okay so you can do this you can also simulate the process so this is cavity for this is caddy five this is a length of the bar is not lower spec limit normal and upper spec limit at a setting of five ten eleven point five thousand psi the predicted dimension of cavity for is fifteen point zero five eight nine for cavity five is fifteen point oh seven whole line I can now come and change the process right here right if I change the process what happens you can see the X is going up this will become red out of spec I can decrease my cooling time I increase the melt temperature right I can do a complete simulation of the whole process right here right this is the length of the part a diameter the part is are the diameter the part that I'm either to and you can see here very clearly that this blue column that represents the change you can make in the molding process the dimensions basically right that is completely outside my spec limits what does this tell you that's it's impossible for you to move these parts to spec because you're going to get flashes short charts you can go down so you have to have a scene change in this case right or look at this dimension no matter what you do it will never go to spec so you can actually eliminate instruction from this one okay so the software you can come in and say okay tell me the most the best process we can also optimize based on the dimensions and the cavity that we select it tells me my most often process if I can come to nominal here right it tells me basically that my most often process is five fifteen fifteen and ten thousand ten twenty five years yeah okay so we also give the most optimum process but okay that's often process what about these parts here okay five fifteen fifteen and ten twenty five we have something called a suggested dimension changes come in here five fifty right ten seconds sorry ten seconds and ten twenty five see now it does me diameter to is fifteen six and fifteen age about sixteen thousands are they from normal if I go change the steel on that cavity at by sixteen thousands I will again fit normal on that dimension I know now the software also tell you how much steel you changing so the several things that you can do the software basically now it tells you what is your example come so what is it it tells your process Windows it can tell you we can find out the most optimum process it can simulate the process and all the stuff soft identity yeah and then that's how you come up with a preferred process right and then you come up with the processor that you really need to run and give dimensions and see what the windows are stuff right once you have the molding process you can record the molding process in the plastic process sheet and under compatibility the software will also now tell you right selecting that thing here I'm going to do this big compatibility the software will also tell you which machine other than the machine that you ran it in which machine will it fit in as far as percentage short size goes and residence time goes and all the stuff but that is one tab the second tab is process machine compatibility so machine five is where I came up with a good robust process based on scientific holding and new ease and now I can now predict give you this the process for every other machine in the in the database what is read is not acceptable what scream is acceptable so now if I decide that I want to write in this is H machine here that coming here select H that click on go the software will now come up we take with the process sheet for that machine where you know all these are calculations were anyway in this case right so this is a process sheet how did he come up with this you know that this process is robust because you have done the new Union this right most importantly okay so again that is a it takes me almost one and a bus to show you the software there's a very quick 23 minute video please look at it on the software side of it but I want to get back to my my my presentation here because you know take some questions also so a good example is this does a company there is all these making bad parts and they always had issues customer returns right so what did I do I went so this is a thought that had sink the length was important so we did this first thing I did is a process window sink and flash and decreasing the pressure would bring the parts in spec but then the chances increases chances of sink I'm sure you have seen this right so this is how the process in the worst so how did I how did I know only because around the do in this case right and so what are you doing this case you simply come in here center the process right in this you know center the steel down here based on the div data again and now first thing you choose a robust process with a nice wide window on both sides as just a steel and now you can have a good robust process in fact in this case there are three dimensions that the customer that will be stocked measuring up to this and before that they would always get me checks from the customers because either sink or dimensional respect right how good is my software here is from my seminar one of my seminars I get this every single time but this the one that I actually made up the PowerPoint for you can see here that the actual under predicted so we'd run this this is run number nine which is the the confirmation run look at the actual and the predicted that is almost a thousand actually less than thousands as rounded up there right that's how good we have our chakras it predicts the dimensions out there oh you're right then the process transfer how does how does how good is that right so it'll Academy a custom of mine from fifty to hundred and machine using that printout that I showed you they were able to get the dimensions to tens of thousands and almost a thousand nine tenths of a thousand step right so again excellent prediction is what we get into the software okay again everything is data guys right there's no going and let me change this let me change that that's the biggest thing that is very important right another customer they say four hundred thousand dollars is right here irrigation company they save a lot of money on this by doing again is all being based on trees they found out that one of the parts much better have reduce inspection all that stuff again without doing a do you never know how fast you can run a mode you never know where the dimensions are and how soon they finish this in water spec right the cost of software by the way is equal to the cost of one more trial on the send about a seventy five hundred it's two thousand dollars a year we constantly add stuff to it and that is a subscription-based model just in case you guys the one thing why there's lots of stuff that Robbie and myself are working on very cool stuff that should be coming up here very soon also yeah company in the UK they ended up saving just inspection just an inspection right they save almost thirty thousand pounds you can see this number in yeah okay so I always do a two stage approach scientific morning is one one part of it and then basically I've brought the second part which is the that is an experiment so I highly suggest that you guys do these things again these flowcharts are available at no cost and the phrase of send me an email any more than happy to send you this information absolutely right again one more thing I'll tell you that's fantastic technologies out there example there's pressure sensing technology but again if you don't have a good robust process those technologies will not be useful for you guys so I think this is this is great great stuff to maintain and to monitor your processes and look at the parts and and you know so I think if you're using this I think you should be using the doing Yogi's also right so it's very important to set alarm limits for you set along limits right is all should be based on again one day they're right again you may have the world's best machine out there okay you might have very good machines now I know I've heard very good things things about sonic machines have a customer down in Texas that uses I think they're for Smurfs you know a lot of machines down there they are mainly the world for small parts but I know that sort of makes machines that are you enlarge it on machines and I'm sure the technology transfers in the same technology so definitely know like I said you may have the really good machines you know but again you need to also have good processes so this is also very important right so why don't more does do nearly very often less let's look at this right it's very complicated is it it's not right it's very simple I showed you thanks for click sorry ever analysis we need to understand that you don't need to understand that let's take a lot of time to perform not really 9 runs or 5 shots usually in two as it should be finishing the runs you do the calculations guys okay in within within two hours it should be running all the do any parts ok waste of material resources yeah it is that I'm not you use if you don't do you is now you will be wasting all that material in production because you're constantly running took the presses and to and having bad parts that right it takes a lot of time to measure all the cavities that is absolutely true but there's a way that I can tell you how to reduce measurement you can take it let's talk about that offline but there are ways yes measurement is the one that takes the most time but again the small dimensions and all the stuff are got a way that you can get around and I'll talk about that right I need to learn how do you software is that true no just look at my software is like four flicks yeah I can come up with the same process without going to the do II yes but how robust is that process right what do you know how do you know what you need to change in there right is it process efficient all the stuff will not find out until you do URIs right one main important thing that I have seen very important is that you have to have management support your management has to understand that yo is available to you if the managed management does not agree with that then it's going to be a challenge for you to that in your company okay so that's it I think I made time yeah so that is the thing so now having said all this stuff right I just hope that you guys agree that URIs are not complicated

Channel: FIMMTECH

Views: 6,247

Keywords: Design of Experiments, DOE, Injection Molding

Id: NwoNIs6rLms

Length: 41min 17sec (2477 seconds)

Published: Wed Jul 22 2020