 Hello everyone and welcome to the third sessions of this quality control and improvement using Minitab. Last session what we have done is that we have discussed about voice of the customer and also how voice of the customer is related to technical specifications of CTQ, how we can convert the voice into technical specifications of CTQ. And we have also discussed about how the voice of the customer can be stratified using a cano model and stratification and prioritization of voice of the customer that we have discussed like that. So, today what we will do is that we will try to discuss about quality function deployment or which is known as House of the Quality that relates between voice of the customer and CTQ that building between voice of the customer and critical to quality characteristics like that. So, how it is done we will try to give an example out of that. So, let us start our discussion from there. And so, this is session 3, this is session 3 what we will do. So, this is session 3, so voice of the customer over here. So, quality function deployment basically which is known as House of the Quality what does it consist of basically. So, you will find that on one of the box over here you will find that the voice of the customer. So, there will be voice of the customer one like this there can be n number of voices over here. And the CTQ will be written that CTQ dimensions will be written over here one two. So, this will be done by the engineers and they understand that in these voices has to be translated in the products in that case what is the characteristics that has to come out of the products basically that will basically link the link to the voice of the customer basically. So, voice of the customer n number of voices will be addressed by m number of CTQs let us let us try to think about that scenario ok. So, this two box will show you the CTQ and this box will show you the VOC. And the correlation between VOC and CTQs will be shown in this box what you see over here. So, they can have strong relationship one voice can be addressed by CTQ, but relationship can be strong or weak like that or medium relationship can exist over here. And this will be shown in this matrix over here. And relationship between the CTQs that how CTQ 1 and 2 are related like that. This relationship will be shown on the top of the top of this house of quality and this is known as roof of quality we can think of. So, what we will see is that there will be a roof of the matrix which will show the correlation between the CTQs also ok. So, how CTQ 1 is related with 2 like that ok. And then you will find that each voice of the customer will be evaluated by the customers and the average rating will be shown over here. So, if our company is X and other companies A or B like that. So, where we stand in a scale of 1 to 5 like that. So, average rating may be for a particular specific voice over here can be we are rated as one as the lowest rating over here and they are given a high rating over here. So, then this is a concern for us. So, the way we are addressing the voice may not be sufficient and we need to improve our ratings and in that case we have to follow what A and B is doing basically ok. So, we have to improve that one in whatever way that is that is possible. So, so similarly, so this will be done by the competitor each voice will be survey will be done and each of the voice where we stand as compared to a competitor that will be evaluated in case it is a existing product that we are trying to launch over here and we are trying to improve on that. So, quality function deployment can be done at design stage also can be done at various stages like that. So, even after implementation Q every can be done because every time CTQs are keeping on changing because the voice of the customers are. So, this is a dynamic one. So, voice changes, CTQ changes and we may have to replace some CTQs because voice has changed. So, in that case we need to continuously improve on that and then competitive evaluation also changes with time. So, in this case also and the CTQs also will be evaluated based on the engineers will give you inputs over here. So, in that case if we have defined a values over here for any of the dimensional CTQs. So, which will have specifications like that and in that case engineer will say that our specification how it is performing as compared to the other competitor specifications like that. So, this is the technical input that we will get from the engineers or process experts or anybody who is product experts like that. So, in that case expert opinion will be taken over here. So, in this case what we will what we will what we can see in this in this box is that technical evaluation on every CTQs. So, all all the CTQs and what is the evaluation in a scale of 1 to 5 again. So, where we stand as compared to the CTQ that that is provided by the competitor. So, this information will be used maybe in reverse engineering when we are doing reverse engineering this information will be useful over here. So, voice of the customer will be translated into CTQs CTQ dimensions over here. So, now let us take an example to understand this one what what is happening over here. So, in this case what what we have taken over here is the card do design example over here. So, one of the examples from book we have taken like that. So, what what we are trying to what we are trying to show over here is that this is the roof of the matrix what you see roof of the house of the quality over here. So, this is the box where voice of the customers are noted down over here and this is the CTQ box that you are seeing over here and this is the competitor evaluation box that you are seeing over here and this is the correlation box that you see over here evaluation of VOC and CTQs like that and here we will have targets for each of the CTQs like that target values we will target values will be set over here will be shown over here and the technical evaluation will be in this box. So, where we stand as compared to as in the CTQs basically. So, this matrix will say in terms of voice of the customer where we stand this this below box will say that CTQ stands with respect to competitor where we stand basically and this correlation matrix what we will show is that relationship between any voice voice of the customer one let us say and CTQ one let us say over here. So, here one of the voice that is shown over here is easy to close. So, card do design. So, customer wants that it should be very easy to close like that. So, then technical engineers like that. So, then they have defined that easy to close is highly related with energy needed to close the door basically and also door seal resistance is another parameter which also influences or that is another CTQ which influences how much easy it is to close the door like that. So, these are the CTQs CTQ one and CTQ two we can think of. So, these are the two CTQs what we can think of over here energy needed to close the door and door seal resistance basically. So, energy needed easy to close is highly related with energy needed to close the door. So, that is why this symbol is used delta and this delta is equals to strong relationship is rated as 9 over here. So, we are giving some numbers over here this is arbitrary over here. So, any company organization to organization this can differ. So, what we are trying to say is that what is the interrelationship or correlation between voice and CTQs like that. So, this is shown over here as they have a strong relationship. So, energy needed is highly related with easy to close like that. So, similarly easy to close is highly is mediumly related over here this is based on previous evidence or technical literature or something like that they knows that door seal resistance is high and in that case what will happen it is very easy to close like that ok. So, they have a but they are saying that this is not very strong relationship but energy needed is highly related with voice of the customer 1. So, voice of the customer that is first voice of the customer we can think of ok. So, this is based on experience based on feedbacks based on expert opinion based on literature like that. So, we will be able to know what is what is a strong relationship which is having strong relationship which is having a weak relationship over here ok. And then they can be given a scale like 9, 3 medium relationship is given a scale of 3 and this is 1 equals to number is given as 1 like that if the relationship is not so strong like that ok. So, it can be positive or negative any direction of relationship we can think of ok. So, then energy needed to close the door and door seal resistance you see on the top also there is a matrix which shows the relationship between these two. So, CTQ 1 how CTQ 1 is related with how it is related to CTQ 2 over here. So, how they are interrelated basically that will be shown over here let us say the symbol that they are using over here is cross that means they have a negative relationship. Negative relationship means one if you try to improve one the other one will deteriorate like that. So, in that case they have a correlation within the CTQs like that. So, energy needed to close the door and door seal resistance. So, you see if door seal resistance is high and then in that case energy needed to close the door will be will be small ok. So, there is a negative relationship that means one in if one increases other will decrease basically ok. So, that is CTQ relationship between them that will be shown in the roof of the this matrix what you are seeing the or roof of the house you can think of. So, all this will be defined by the quality personnel or production personnel or the experts opinions that will say based on evidence what we have literature or other references we can we can also relate CTQ how they are interrelated like that. So, we can just have a picture of the I mean idea about that one ok. So, voice of the customer any voice of the customer will be related with any of the CTQs over here ok. If the voice of the customer is not addressed by any of the CTQ that means we are missing something in the design basically ok. And or maybe this voice of the customer is not relevant so much relevant because we are not interested in this voice of the customer while we are designing this one. So, every CTQ is important and it is addressing a voice if CTQ fails then customers will be dissatisfied because it will impact their voice of the customer and it will impact the customer's opinion or feedback about the quality of the product basically ok. So, this voice of the customer will be related like that way and this matrix what you see technical evaluation over here where we stand where B and A stands as compared to our product which is X maybe. So, we are X company similarly competitors evaluation which will be done by customers. So, in this case what will happen is that customer will evaluate this one and maybe we are standing. So, this will be in a rating of 1 to 5 that you are seeing over here and maybe we are standing at 1 near to 1 and A and B is standing near to 4 and 5 like that. So, they are getting highest rating as compared to our products in this case for specific voice number 1 over here. So, where we stand is also important as compared to our competitor and that will basically dictate that what changes are required. So, if I have to address this one then I have to go back which is the CTQ is addressing this voice and then compare the CTQs with the competitor CTQs like that and what is their target value which is known as reverse engineering. So, that we can also do something on this so that our technical feedback is also saying that we are lying at the lower level over here. So, we need to improve to A and B. So, maybe our specification should be closely following the specification what is given by A and B like that and let us try out that one and experiments and find out what is going wrong basically. Any company does this reverse engineering you go to any companies they will do reverse engineering and they have ideas of the competitor products also. So, that is also reflected over here so in this matrix over here. So, this matrix how it helps basically so I want to voice is prioritized over here. So, then you will see an importance matrix importance to customer over here. So, there will be some rating that is given over here 7, 5, 3, 3, 2 like this this is a rating that is given over here. So, this importance rating may be based on customer feedback they will they will get some customer feedback over here and average customer feedback will be noted down over here it may not be integer over here it can be any numbers like that. So, in a scale of let us say 1, 2, 10 which is most. So, some numbers can be given over here importance number for every VOC is like that. So, this input will be taken from the from the customer and also we have a Kano model which will also which also says which voice calling involved what category maybe this is attractive features or this must be category. So, we can also differentiate one dimensional category like that. So, this can be hypothetically this is a hypothetical one I am just arbitrarily mentioning this one, but you can think about this and which will fall in what categories like that and it will be shown by Kano model when they are doing surveys they will figure out which is following a which which strata that means which category basically. So, in that case that is also possible. So, Kano model will also say so if it is attractive features maybe we can give a also a score to this and we can also give a score to this and we can we can give a score to this and must be components can be given a score of 1. So, 4, 2, 1 scale we can use for this attractive must be an this is also one weightage we can think of over here and also we can ask the customer to give the rating like that. So, this is given by the customer rating and Kano stratification also says which is important which is less important like that ok. So, this numbers will be used to so I want to prioritize because voice is already prioritized by Kano model I want to prioritize the CTQs also CTQ 1, CTQ 2 up to M CTQs which is important which is not important. So, when you see many of the one voice is addressed by many CTQs or maybe one voice is addressed by only one CTQs like that that can happen. So, that is that is also possible one voice is addressed by many CTQs or many voices are addressed by one one single CTQs like that that is also possible reverse way, but the voice that is noted over here basically what is important to the customer and CTQ is those CTQs which are important and addressing the voice of the customer ok. So, nothing is unimportant over here that is there is no redundancy over here ok. So, not on the CTQ not on the VOC. So, every voice should have a CTQ like that. So, that is what is emphasized over here and then then this what you see is that let us say the CTQ number 1 let us say consider the CTQ number 1 and it is having a strong relationship with the first voice of the customer. So, voice of the customer 1 is having a CTQ with 1 is having a strong relationship that is symbolically shown as delta over here ok which is given a rating of 9 let us say which is given a score of 9. If it is strong we will say this is about 9 we can think of a number which is arbitrary we can which is defined by the organization like that ok. And the rating that is given over here let us say customer average rating is 7 or approximately and this is a attractive feature and we we are giving a rating cano rating of 4 over here. So, let us try to see how this calculation happens over here to calculate the CTQs over here. Also the engineers will rate the CTQ in a scale like that. So, engineer has rated 10 he has given a score of like this score is given in a scale of 1 to 10 like that. So, it is given a score of 10 by the engineers for this CTQ over here. So, then what happens is that for CTQ 1 I have to calculate the total score of this. How do I calculate like that? For voice number 1 the weightage is approximately 7. So, 7 multiplied by and this is a strong relationship that we are seeing 7 multiplied by 9 is the rating over here. Then then for the CTQ I go column wise over here there is no other association between this CTQ 1 with any other voices over here. So, all other will be equals to. So, all other will be equals to 0 over here. So, this will be continuing up to this n number of voices that you see over here. So, this will be continued and nth observation will also be 0 over here. So, in this case and this will be so summation of all of this will be 7 63 over here and the importance rating is approximately equals to 10. So, this will be score will be around 630 over here for this CTQ the overall score will come out to be 630 over here. Similarly, for the second CTQ for the second CTQ CTQ number 2 what you can see is that 7 will be multiplied. So, this is 7 rating for this voice of the customer and medium relationship exists between this 7 multiplied by 6 and then what we can do is that this is easy to open is also related to CTQ CTQ 2. So, in this case this will be 3 multiplied by and this relationship medium is shown over here which is given a rating of 3. So, 3 multiplied by 3 like this and similarly third one what you see is that does not leak in rain over here. So, this is equals to what you are seeing is that 3 rating is given over here. So, this is 3 multiplied by 3 multiplied by. So, this will be so we will go back to this. So, this will be equals to 3 multiplied by strong relationship like this. So, this will continue like that and total summation will be multiplied by importance of weightage that is 6. So, this is this will be equals to 6 multiplied by this. So, this overall rating can come out let us say this is some number over here which is a number some number is coming out. So, we can calculate this one and we can say this is n 2 like this. So, this is n 1 over here. So, we will get for every CTQ a number over here n 1 n 2 like this n m. So, this is the total rating we can think of total total rating number what we are getting out of this for CTQ 1 and CTQ 2. Then what we can do is that from highest to lowest we can arrange the CTQs and we will get which CTQ is getting the maximum maximum ratings like that and we can prioritize because if the CTQ is more important that means more rating it is getting and so we need to concentrate on those CTQs and those CTQs should not fail because if that CTQ fails in that case voice of the customer will be will also fail. So, in that case that we have to keep in mind. So, this is the overall house of quality or quality function deployment it is done at the design stage it is done at production stage for any improvements we always do this QFD. So, QFD can be in shop floor also say every time what is the CTQ and what is the voices and conversion of this. So, existing products and also for new products like that we can we need to do this QFD analysis. So, that we understand where we stand as compared to our competitors like that. So, there is no voice that will be noted over here that does not relate to any CTQs like that. So, there will be no CTQs which does not address any other voice. So, redundancy will be not be there. Now, these ratings that we are getting over here. So, this is 7 rating over here. So, again this can be multiplied with 4. So, overall importance of matrix may be 7 multiplied by 4 which is coming from the Kano model like that. So, that importance it can also be multiplied over here and then summation of this will be multiplied with 7 7 into 4. So, that score can be taken into consideration. So, many of the research articles also consider that Kano rating along with customer rating can be multiplied together and that number will also can be noted down and that overall rating can be given based on the inclusion of this Kano rating also. So, that is also possible and but all these things are somewhat some numbers are arbitrary because 9 3 I can change this number over here. So, I can make it 6 3 1 like that. So, it depends on company what marks they are giving like that and that can influence that can that can totally dictate the scores like that. But more or less the analysis will be it will it will reflect the same thing because if there is strong relationship it only the numbers will somewhat change and it will change for every every locations like that. So, strong medium small we can change the numbers but overall which CTQ is important which is not important will be always reflected because importance is given by the customer and Kano importance we know. So, that will be multiplied with the correlation matrix that we are getting over here and based on that overall score will come. So, which CTQ is important that will not change basically top priority CTQs will not change over here even if some number changes you are here and there you are making. So, that will not impact the overall importance of CTQs where every CTQ stands. So, that will not change much. So, this is what I was mentioning over here. So, Kano ratings can be multiplied with the independent survey ratings like that and Kano can be taken as 4 some of the authors have mentioned like that. So, if it is a basic requirements it can be given a rating of 1 because we are interested in attractive one dimension and basic. So, here we should not fail anywhere we should not fail. So, reverse we are not interested and we are indifferent also we are not interested because in the QFD we only put attractive one dimensional and basic because that is the core. So, that is the core and that will define the CTQs basically. So, if it is indifferent that means that dimension is not so important we can also drop that from the design. So, that was we should not address we are not concerned about addressing that one and reversal anyway they will not include in the design. So, most of the time. So, in that case we are only interested in attractive one dimensional and basic. So, ratings can be 4 to 1 and that can be multiplied with the survey ratings in a scale of 1 to 10 like that. So, that I have shown earlier in the QFD matrix what we have drawn over here. So, this number 7 which is given by the customer is can be multiplied with 4 over here and then it will be multiplied with or overall I can multiply increase the number over here. So, this will be 7 multiplied by 963 and this will be multiplied by 10 630 630 can be multiplied with 4 because this is attractive VOC. So, we can multiply again with 4. So, numbers will increase. So, changing numbers will happen. So, I am giving some weightage to the Kano analysis also over here. So, that is considered over here and that is what we can think of as Kano model that we can think of as Kano model in brief idea that we are placing over here ok. So, and more can be studied in books you can see books and there can be many other views like this ok, but overall idea is like that I want to relate the voice of the customer with the CTQs like that ok. So, then CTQ has to be translated into products like that. So, how this is done in manufacturing like that. So, we will have this is the diagram what we can use as a basis like that. So, there will be some inputs over here there will be some inputs that will come in the process like that it can be subassemblies it can be raw materials like that it can be any other inputs like that. So, then what will happen this is the process when it goes inside the process like that you can think of machines like that. So, whenever the raw material is entering into the process or maybe a continuous manufacturing a steel industry there will be inputs over here. So, in that case raw materials will be provided and the process will run or valuation will happen and then we will have products as an output and products will have some characteristics maybe strength or something other which is a CTQ basically. So, CTQ can be strength over here strength ok. So, CTQ can be any other dimensions over here strength elongation or any other parameters which is important to the customer basically. So, these are the CTQs which is coming out of the process. So, product will come out then we will measure in that product these dimensions over here is the strength ok as per specification is it produced as per specification is the elongation ok as per dimensions like that what is given by the customer. So, then we will do measurement and evaluation over here and accordingly what is coming out of the process the CTQ dimensions what we are getting what we can do is that we can change the process settings like that. So, this is a controllable inputs conditions or inputs or variables we can think of and these variables can be changed like knobs over here. So, like in machine speed feed depth of cut this can be changed like that ok. So, similarly there will be inputs in a process there will be output will be basically quality characteristic what we are measuring what would you are interested in that is the CTQ what we will measure we will not measure everything like that. You go to inspection and you will find that they are not measuring everything they are measuring only CTQs like that which is important to the customer. If deviation from there happens then it will affect the customer it will not affect much if some radius is gone wrong or something like that in a product dimensions like that ok. But if the diameter has gone wrong it will it will create problems if surface finish has gone wrong it will create problems like that ok because it will impact the customer basically dissatisfaction will improve if if it fails in the dimensions what is provided by the customer ok. So, customer wants this this dimension which is coming from the voice of the customer. So, those are the CTQs which we are trying to emphasize over here and those are the CTQs we will try to improve like that ok. So, there will be some variables which are known as controllable variables over here and this can be changed this is in our hand basically this is in the hand of the operator who is operating the machines like that. But there can be also uncontrollable variables over here uncontrollable variables what we are seeing over here or uneconomical to control or maybe unknown variables which which we are not able to locate like that which also influences raw material we can understand maybe supply to supply a variation also we can understand like that, but there may be hundreds of variables which which we have no ideas about in the process like that. So, that may be very minor effect of that maybe from previous process or any other any other thing we are not able to locate that one. So, those things which has little impact or if it impacts in that case we should be able to locate that one, but in case it is we cannot do anything about that those are known as uncontrollable variables or it can also be somebody can say noise variables like that. So, their impact is very less like that. So, in that case we are not so much concerned, but later on we will see that this also needs to be considered in in making a product robust like that. So, in that case that that is also considered like. So, these variables are uneconomical to control and maybe unknown no information is there about these variables and even if information is there we cannot do anything like that. So, these variables we are not, but if if we can control these things at least our output will be within the specification or or whatever is required like that. So, in presence of these variables which cannot be controlled which is known as uncontrollable variables, I need to set this machine over here with these parameters so that I get the Y which is required which is close to the target over here which is close to the target that that is our overall objective. So, CTQ idea of CTQ in a process we can think of like this there will be inputs, there will be output a product is coming out, but we do not measure everything out of the product we only measure the CTQs which is important because CTQ will address the voice and then how to improve the CTQs in presence of some variable which we have no we have no ideas there can be sample to sample variation over here. So, everything is there and in that case we we have some control on some variables which is known as controllable variables and those things we will change in case there is deviation from the target values what is defined by the customer like that we will measure and we will change that. So, at the start of the shift we will find that this operators are changing the knobs like that if you go to the process at the start they will see one product what is the outcome and then they will see they will make some changes and then they will do some twist and turn the knobs like that or change the settings like that. So, this x1 to xp what you are seeing is the controllable variables these are the critical x which we are which we are concerned about and this is the critical y which we are concerned about. So, basically if you know the functional relationship between x and y then there is a favorable situation. So, I can I know the relationship. So, in that case where to change the x so that I get the best y. So, but mentally sometimes shop floor operators you will find that they are they are doing it mentally basically ok they have something in mind. So, in that case how this parameter influences how this x1 variable influences the process like that and how it influences y they have some ideas. So, which one to change which were not to change. So, they have a strong idea about that. So, some does not have qualifications to understand optimization mathematical part of that, but mentally they are doing calculation and with experience they have some idea about if I change this variable what will happen to the y like that. So, that is a inbuilt knowledge you can think of which they have, but overall this is the process scenario what you see. So, I want I want the CTQs to be to the target and for that what should be the setting that that is our overall goal in a in a in a process and for that input conditions will change input will have variation and there will be some variables which we have no ideas like that, but which have little impact like that ok, but that can be exploited also to understand the behavior of y like that. So, that we will try to figure out in subsequent subsequent lectures like that ok. So, this is all about this is the CTQ diagrams we can think of process diagram. You can think of this is one process this manufacturing process you can think of service process also. There will be some control variables, there will be some uncontrollable or noise variables like that, there will be certain inputs and there will be certain CTQs which is required by the customer like that. It can be 8 dimensions in quality it can be 5 dimensions of quality service quality like that. So, you can think of, but we need to measure this one CTQ whenever I have a CTQ I have a dimensions to measure I have a specification spec to that is given for this and then I measure this one and then what I can do is that I can check whether this within specification or whatever is the requirement of the customer and accordingly I will change the settings of these variables and so that I get the best output like that ok. So, that is the idea what we are trying to emphasize over here now CTQ can be strength like that what what I have mentioned earlier ok and you may have done some basic course on statistics. So, the process outputs that is coming out the process can be let us say 100 dimensions I have 100 100 products I have measured and I have measured the strength like that then I can plot that into diagrammatically I can represent that one. So, that is generally done. So, CTQ is one of the this is a continuous variable. So, this is a continuous variable. So, and this variable can be plotted in a chart which is known as histogram and in basic statistics you will you have already done that one and plotting of that is frequency is plotted as compared to the dimensions like that. So, this will be the CTQs so which is the Y measurement that we are getting over here and this is the number of times it is coming that observation. So, 70 to 90 if you can see the observations is around 3 observations are within this zone like that. Similarly, 90 to 110 may be this is 4 observations are within this like this this is the highest what we are seeing. So, maybe 22 observations are over here which lies in the range of 150 to 170. So, this is the maximum frequency what you see. So, most of the observations are around 150 to 170. So, this plotting of the data that is coming out of the process or CTQs can be done using histogram if it is a continuous data type. So, this is the this is one of the thing that we can we can plot this one. So, now we are just trying to see understand and visualize the things like that what is coming out of the process can we measure that one and if we can measure that one can we see that one or visualize that one that is also important because it is 100 times better one picture is 100 times better other than only seeing the data information because if I give you 1000 data points and you cannot visualize that one it is difficult to visualize in excel sheet if I give 100 data points like that. So, I need some visualizations tool. So, we are entering into that phase where we will use this visualization tool. So, in our session 4 what we will do is that we will start with visualization. So, we visualization of the data basically. So, quality talks about visualization of the data. So, more data you will have. So, data and interpretation it is not all about engineering it is now about how do we interpret the data how we based on the data what action we take. So, that is important in quality nowadays. So, data will give you provide you information for decision making basically. So, from engineering standpoint yes is required because you need to understand what specification is important what is the CTQ which CTQ is important that addresses the voice like that that conversion is happening at the design phase if the design phase goes wrong everything goes wrong like that, but whenever I have a strong design I know the CTQs and then I have to improve that one. So, in that case what is required to be done that we have to do in the process like that and then what we what we see that quality of conformance that is the second phase of our discussion. So, quality of design and quality of conformance are we adhering to the specification that is given by the designer for a specific CTQ. So, that will be our agenda for our next session session 4. So, we will continue from here. So, in session 4 we will start with visualization of the data. Thank you for listening, meet you in the session 4.