**SESSION 04 — Transcript & Summary | SozAI** Source: https://sozai.app/transcript/i-can-t-create-a-meaningful-slug-from-session-04-alone-since-the-title-doesn-t/ Prof. Marcelo Machado Fernandes presents Session 4 of a Design for Six Sigma (DFSS) course: design phase, concept selection with Pugh matrix, hybrid concepts, certification exam logistics, and DFSS vs ## Key Takeaways - DFSS is a design-oriented methodology best suited to creating new processes, products, or services, while Lean Six Sigma targets improvement of existing ones. - A Pugh matrix with a sound baseline and clear criteria drives objective concept selection, and hybrid concepts can further strengthen the winner. - Rigorous VOC-to-CTQ translation, including prioritization, is the foundation for reliable design criteria. - Modern tools (simulation, AI, free apps) make redesigning from scratch faster and cheaper, supporting the instructor's view that DFSS is increasingly relevant. ## What the video covers - Session 4 opens with audience engagement checks and a recap of the DFSS journey: Define, Measure, Analyze completed; Design and Verify/Validate remain. - Certification test details: 60 multiple-choice questions, 2 hours, passing score 60% (36 correct), and an e-certificate (PDF) upon approval. - Concept selection uses a Pugh matrix: any concept can serve as the baseline, and if all other concepts score negatively, the baseline is the best option. - After selecting the best concept, the instructor recommends building a hybrid 'super concept' by challenging zeros and minus-ones to strengthen the winning concept. - The VOC-to-CTQ process is outlined: define the niche, collect and classify the voice of the customer, prioritize with pairwise comparison, translate into CTQs, and build a criteria list. - DFSS is contrasted with Lean Six Sigma: Lean Six Sigma improves existing processes, while DFSS creates or redesigns processes/products from scratch. - The instructor argues that in 2026, mastering methods for creating things may be more relevant than improving things, due to cheaper technology, simulation tools, and AI. - An ER waiting-time example illustrates the two approaches: mapping and improving the current process versus completely redesigning patient reception. - The session includes a pizza-based experiment: identifying critical factors, modeling the process, using main-effects regression, and optimizing for CTQ targets. - Sigma-level calculation for two CTQs is introduced, with the note that 3 sigma is often acceptable in practice. ## Chapters 1. 00:00 Welcome and Audience Engagement 2. 05:00 Certification Test Overview and Logistics 3. 10:00 Where We Are in the DFSS Journey 4. 15:00 Pugh Matrix and Baseline Concept Selection 5. 25:00 DFSS vs Lean Six Sigma: The ER Example 6. 40:00 VOC to CTQ Translation and Prioritization 7. 50:00 Modeling, Critical Factors and Pizza Experiment 8. 90:00 Process Optimization and Sigma Level Calculation Answers ## Questions about this video How is the certification test structured? The exam has 60 multiple-choice questions and lasts 2 hours. A score of 60% or more (36 correct answers) is required. Approved candidates receive an e-certificate in PDF form. What is the difference between DFSS and Lean Six Sigma according to the instructor? Lean Six Sigma aims to improve existing processes, whereas DFSS is used to design or redesign a process, product, or service from scratch. The instructor argues DFSS is increasingly more relevant today due to technology and simulation tools. How is the best concept chosen and refined? Concepts are compared in a Pugh matrix against a baseline and a set of criteria. If the baseline outscores the others, it is the best option; the winning concept is then strengthened into a hybrid 'super concept' by addressing its zero and minus-one scores. ## Full Transcript — Download SRT & Markdown 00:04 Speaker A Am I talking to the best of the best of the best of the best? Yes or not? Please type BBB. 00:20 Speaker A [laughter] If you are happy to start session four, session four. Yes. Amazing. Four Bs. Four Bs [laughter] from 0 to 10, please. How is my audio [snorts] from 0 to 10, please? How is my audio? 00:54 Speaker A [laughter] If you are happy to start session four, session four. Yes. Amazing. Four Bs. Four Bs. 01:30 Speaker A design and also verify or validate if you prefer. How exciting is that? And today I will be giving all the details about the certification test. Okay. So it's a very very very important session. 01:47 Speaker A [laughter] From 0 to 10, please. How is my audio? [snorts] From 0 to 10, please. How is my audio? 02:09 Speaker A All questions are good questions. I'm [snorts] glad to know that tresaras. Yes. Multiple choice. Yeah. You will have 60 questions. 60. Okay. And you need to score 60% or more. Yeah. So, you need to score 36 questions or more. 02:38 Speaker A Amazing. Amazing. Where are we in the DFSS journey? Where are we in the DFSS journey? 02:47 Speaker A Okay. The second D stands for design. For design. Exactly. And I don't know if you guys saw, but I put our template. 03:03 Speaker A [snorts] We have covered define. We have covered measure. We have covered analyze. And today I will be finishing up training content. So today I'll be talking about design and also verify or validate if you prefer. How exciting is that? And today I will be giving all the details about the certification test. 03:20 Speaker A Yeah, I'll be talking about the logistics of the exam. Okay. How do we identify which is the baseline concept? Great question. In fact from pure matrix you can take any concept from the pum matrix any concept and then you put everything 03:44 Speaker A Okay. So it's a very, very, very important session. Very important session. 04:01 Speaker A So it doesn't matter. Sometimes people take let's suppose you are redesigning a process or a product then sometimes people take the the current product the current process. That's an alternative too. But to be honest it doesn't matter. 04:20 Speaker A Um, let's just run a quick content review. Let me do it differently. Please type here your questions. All questions are good questions. Type here your questions. All questions are good questions. Type here for me. All questions are good questions. 04:31 Speaker A Yeah, a proper a reliable list of CTQs. It's crazy, right? We define the niche. 04:37 Speaker A I'm [snorts] glad to know that Tresaras. Yes. Multiple choice. Yeah. You will have 60 questions. 60. Okay. And you need to score 60% or more. Yeah. So you need to score 36 questions or more. 04:49 Speaker A have our list of criteria [snorts] to help us selecting the the best concept. Okay. 05:03 Speaker A Um, being approved, you will be receiving a PDF. Okay. So it's an e-certificate. 05:21 Speaker A that? Yeah. Uh another important thing is I saw a question about lin six sigma and design for six sigma. Link six sigma is to improve existing processes. 05:37 Speaker A Okay. The second D stands for design. 05:47 Speaker A and or redesign. And because we are in 2026 in October almost November 2026, in my humble humble point of view, humble point of view, mastering a method to create things may be more relevant than mastering a method to improve things. Why? 06:21 Speaker A For design. Exactly. And I don't know if you guys saw, but I put our template. 06:44 Speaker A Yeah, I love both. You guys know that I love both. I love both. But to be honest, if you ask me, you know, today, Marcelo, what's your opinion? Your point of view. It's just my point of view. You 06:59 Speaker A Yes. Our template is in the description of the video. Okay. [snorts] Amazing. Let me see. 07:19 Speaker A nowadays and based on my experience in the field in the field I I do I I go to gamba I go to gamba I have an impression that today the FSS is more relevant than Linix Sigma. That's just 07:43 Speaker A Yeah, I'll be talking about the logistics of the exam. Okay. 08:05 Speaker A limit in the exam? Yes. So you have uh 2 hours for the test. Okay. But if you are taking notes, if you are taking the sessions, uh if you are taking this journey seriously, you will take the testing honestly 08:22 Speaker A How do we identify which is the baseline concept? Great question. In fact, from the PUGH matrix you can take any concept from the PUGH matrix, any concept, and then you put everything in zero. Okay. You can pick anyone, and the thing is, if all the other concepts get a final negative score, it means that your baseline is the best. Okay. 08:37 Speaker A Yeah. It's not the FSS is better than Lin Sigma. It's more relevant. What do I relevance changes more relevant? Um because see at the end of the day we want to impact processes. 08:52 Speaker A So it doesn't matter. Sometimes people take, let's suppose you are redesigning a process or a product, then sometimes people take the current product, the current process. That's an alternative too. But to be honest, it doesn't matter, because what you need to have is a reference and a good set of criteria. That's the thing. 09:13 Speaker A In which sector are you? Banking, healthcare logistics manufacturing IT construction. Yeah. Yeah. So I I believe I can use Okay. Healthcare I can use um process. 09:29 Speaker A Yeah, a proper, a reliable list of CTQs. It's crazy, right? We define the niche. We collect the voice of the customer. We classify the voice of the customer. We prioritize the voice of the customer. We duplicate. We shrink down the list. We translate the voice of the customer into CTQs. We prioritize the CTQs just to have our list of criteria [snorts] to help us select the best concept. Okay. 09:47 Speaker A map out the process measure everything in the current process in terms of CTQ. Then you find the root causes. This is beautiful. This is valuable. This is amazing. or or you can imagine that you don't have an ER you 10:06 Speaker A Yeah. And now that we have the best concept, we do what we call super concept or hybrid concept. That is to challenge all the zeros and minus ones in order to make the winning concept even better. Did you guys understand that? 10:23 Speaker A Because nowadays we have crazy, you know simulation resources. We have apps. We have so many apps for free. So many apps for free that we can simulate. We have AI, you know. So um if you absorb if you absorb 10:45 Speaker A Yeah. Another important thing is, I saw a question about Lean Six Sigma and Design for Six Sigma. Lean Six Sigma is to improve existing processes. Design for Six Sigma is to create a process, a product, a service from scratch. Okay, or redesign. 11:16 Speaker A tell me honestly, okay, did you have at least one aha moment, at least one eyeopening conversation that Marcelo that you were able to directly or indirectly apply in your work? 11:34 Speaker A And because we are in 2026, October, almost November 2026, in my humble point of view, mastering a method to create things may be more relevant than mastering a method to improve things. Why? [snorts] Because redesigning and creating things from scratch became cheaper and easier because of technology, because of the resources, the amazing resources that we have nowadays. Do you guys understand that? From 0 to 10, how much do you understand that? 11:51 Speaker A the value is not only in the projects you know I want you to run projects but um you can use the tools in independently the concepts independently. 12:08 Speaker A Yeah, I love both. You guys know that I love both. I love both. But to be honest, if you ask me, you know, today, Marcelo, what's your opinion? Your point of view. It's just my point of view. You may disagree. You may disagree. In October 2026, with all the amazing things that we have available for us in terms of technology, everything, sensors, components, in general electronic components, AI, you know, based on all the things that we have nowadays, and based on my experience in the field, in the field, I go to gemba. I go to gemba. I have an impression that today DFSS is more relevant than Lean Six Sigma. That's just my humble point of view. Okay. 12:29 Speaker A That's the thing and that's why I love to simplify to use simple words [clears throat] you know to shrink down [clears throat] the complexity you know at the same time I keep the relevance do you guys understand that 12:54 Speaker A I have a little bit of experience in both, a little bit. [clears throat] So my opinion may have a certain level of assertiveness. Maybe I am wrong. Yeah. 13:14 Speaker A mind. I know I know about the voices of your mind. Yeah, you need to silent the voice of your mind when we are talking about designing something. [clears throat] Yeah. 13:32 Speaker A Is there a time limit in the exam? Yes. So you have two hours for the test. Okay. But if you are taking notes, if you are taking the sessions, if you are taking this journey seriously, you will take the test, honestly, in 30 minutes, maybe 30 minutes. Yeah. But I give you two hours. Two minutes per question. Okay. Two minutes per question. 13:54 Speaker A procedure. Yes. And for governance and compliance, this is key key. And so how can you make how can you write down, you know, a governance and a an SOP or a series of SOPs truly and fully connected to your 14:12 Speaker A Yeah. It's not that DFSS is better than Lean Six Sigma. It's more relevant. What do I mean by relevance? Changes, more relevant. Um, because, see, at the end of the day we want to impact processes. At the end of the day our mission is, in your company, if you are in the automotive sector, in which sector are you? Type here for me. In which sector are you? 14:29 Speaker A Seamok is uh CM OC. It's the the largest it's a Chinese neobium lining. The largest neobium operation in in Brazil. I don't know if [snorts] you guys can see uh this is in Gois Catalon Gois. 14:54 Speaker A In which sector are you? Banking, healthcare, logistics, manufacturing, IT, construction. Yeah. Yeah. So I believe I can use... Okay. Healthcare, I can use... um, process. 15:12 Speaker A And um just to give you an example uh [clears throat] maintenance lead time or delivery time simple like that simple like that not preventive maintenance okay corrective maintenance. 15:30 Speaker A So imagine that you are improving a process. Okay. I don't know. Let's suppose in a hospital, you want to reduce the waiting time in the ER, emergency room. You have two ways to do that. You can apply Lean Six Sigma and map out the process, measure everything in the current process in terms of CTQ. Then you find the root causes. This is beautiful. This is valuable. This is amazing. Or you can imagine that you don't have an ER, you don't have anything, and let's forget what we have today, and let's completely redesign the process of receiving people in the emergency room. Do you guys understand that? 15:52 Speaker A but we had to redesign the process [clears throat] we redesigned and it's funny because sometimes when you are on a root cause analysis, another another amazing project from Fernanda. Um, we optimized the consumption of dies in the trucks, you know, the 16:15 Speaker A Because nowadays we have crazy simulation resources. We have apps. We have so many apps for free, so many apps for free that we can simulate. We have AI, you know. So if you absorb the concepts, the principles, the tools, and techniques correctly, you have a powerful machine of value in your hands from the content that we saw so far. Okay. From what you've learned so far, tell me honestly, did you have at least one aha moment, at least one eye-opening conversation, that you were able to directly or indirectly apply in your work? 16:27 Speaker A And many times when you reach u uh analyzing lin sigma the root cause analysis depending on the root cause that you find it's better to start from scratch to redesign it's like me playing soccer you know I [snorts] don't know I don't know if I am 16:49 Speaker A Did you have anything that you've learned in these three days that you could apply? Marcelo, I had a chance to apply this in a meeting, in a project, you know, because that's the thing. [clears throat] When you learn these things, the value is not only in the projects. You know, I want you to run projects, but you can use the tools independently, the concepts independently. 17:11 Speaker A Marceline. No, Reborn is too dramatic, you know. But maybe it's better for me to forget anything that I know about soccer and start from scratch learning, you know, a new method. For example, yeah. 17:32 Speaker A Uh, I hope I am able to change your mind. I hope I am able to contribute for you to see things differently, to see things differently. 17:37 Speaker A Soccer. Yes. Beautiful. So that's our template. That's our template. As you guys know, we had a chance to we have this important road map. Yeah, this important road map and I think just to guarantee Yeah, just to guarantee that I truly want you to 18:02 Speaker A That's the thing, and that's why I love to simplify, to use simple words [clears throat], you know, to shrink down [clears throat] the complexity, and at the same time I keep the relevance. Do you guys understand that? That's my learning strategy, because if you understand the value of first listening to the VOB and then the VOC, if you understand the value of listening to the VOC and not the voice of your boss, or not your voice, the voices of your mind. I know, I know about the voices of your mind. Yeah, you need to silence the voice of your mind when we are talking about designing something. [clears throat] Yeah. 18:39 Speaker A Yeah. So um yeah so we have defined uh customer segment customer segment here we have collected the voice of the customer. 18:59 Speaker A [snorts] Does it make sense? As head of the governance and compliance department, what I learned from you, Professor, applies to the streamlining and creation of organizational SOPs. SOP is beautiful, beautiful, beautiful, beautiful. I'm glad to know that SOP is Standard Operating Procedure. Yes. And for governance and compliance, this is key, key. And s 19:28 Speaker A that things are not random. Once again, nothing here is random. There is a method in place. Okay? Because if you master the method, there is a very very very high chance that you will get better results in comparison to trial 19:51 Speaker A [clears throat] and error. Stop trial and error. Type here for me. No more. No more trial and error. Type here for me. 20:02 Speaker A No more trial and error. Type here for me. No more trial and error. Try and error is a thing of the past. I am doing this so you fully understand. 20:20 Speaker A Uh even when I use different words. Yeah, you fully understand that where we are. Okay, it's important to know where we are. It's very much important to know where we are. 20:38 Speaker A Okay, so once again we have defined the customer segment. We have collected the voice of the customer. Yeah. The final voice of the customer. Then we have classified used using kenno model. Yeah. 20:54 Speaker A Then we have prioritized the voice of the customer using pair-wise comparison. So at the end of the day we got a list of prioritized voices. Yeah. 21:09 Speaker A And then we translated VC into CTQs using uh QFD quality function deployment. So this is the list of uh the the matrix you know uh of um VC's and CTQS and then at the end of the day we got a list of prioritized 21:31 Speaker A CTQs. Okay, this is very much important. And then yesterday during um yesterday during analyze phase we are we were able to identify potential concepts. And what is a concept? A concept is just an general idea, an initial idea. initial idea of the design 22:03 Speaker A that we want to create. Okay? An initial idea of the design that we want to create. Is that clear? From 0 to 10. How clear is that? And you should be proud of your professor because men cannot multitask. I am conducting the review at 22:22 Speaker A the same time I improve the material for you guys. [snorts] You should be proud. Only women can do that, you know. So you guys should be proud of me. [laughter] We cannot multitask. 22:40 Speaker A We don't have this ability. This is a superpower. This is a superpower totally concentrated. 22:49 Speaker A Yeah. [snorts] in the amazing women in this planet. And by the way, by the way, what would be you know what would be what would happen to us if we without an amazing [snorts] wife or an amazing mother or an amazing grandmother. Come 23:19 Speaker A and make some noise in this chat for all the amazing wives, all the amazing mothers, all the amazing grandmothers. 23:29 Speaker A [applause] Make some noise in this chat. [applause] Guys have a special mission in this planet. Okay, you know that guys. So, how excited are you for us to start on design phase? How excited are you to start on design phase? 24:16 Speaker A Amazing. So remember, now it is time for us to identify critical factors. Oh my god, what is this? 24:44 Speaker A Identify critical factors. How comfortable are you identify critical factors? How comfortable are you with the concept of y as a function of x? Okay. 25:11 Speaker A Y as a function of X from 0 to 10. How much do you understand that Y is also called the dependent variable and X is the independent variable. So Y is the dependent variable and it's our CTQ. 25:39 Speaker A Yeah. And X is nothing but the process or product characteristic. Okay. Okay. I need to know for example and by the way [clears throat] we should take all CTQ's all CTQs one by one and do what I'll be doing here 26:29 Speaker A okay because of time because this is a Green Bay education program. I'm going to pick two. I'm going to pick two. We are going to design the final pizza in details based on two CTQ's. 26:50 Speaker A Are you guys following me? Are you guys following me? and so remember that CTQ number two is that CTQ related to quality. 27:32 Speaker A Yeah. Do you guys remember this CTQ here? CTQ number two is pizza delicious. But this is like that reaction. Wow. You know, imagine the student went there and took a bite. 27:55 Speaker A Yeah. And then the reaction was, "Wow, this split is amazing. It's yes or not. 28:09 Speaker A Binary. If you like stats a little bit, it's binomial probability distribution. And so how can we identify the critical factors, the characteristics in the pizza? Which pizza? Which pizza? 28:40 Speaker A The c the campus harvest. Remember that this is our winning super concept. You guys remember that, right? 28:54 Speaker A How can I identify the critical factors? Critical factors that will drive critical factors characteristics that will guarantee the highest percentage possible of Whoa, the pizza is amazing. from 0 to 10. How much did you understand that? 29:23 Speaker A Why as a function of x? It's like my friends calling me to play soccer. 29:28 Speaker A Marcelo, let's play soccer. My answer is I don't know because if I will play soccer or not is a function of my wife. 29:40 Speaker A [laughter] Let let me check with my wife if I want to play soccer. Okay. Is that clear? 30:17 Speaker A Is that clear? [clears throat] Okay. [snorts] And then I'm going to pick we need to select. Let me write down this. Let me write down this. We we should perform this analysis for all CTQs. 31:14 Speaker A Um because of time constraints I will be selecting two CTQs. Okay. [clears throat] I will take CTQ number two that is is pizza delicious. 31:36 Speaker A Is pizza delicious? Uh [snorts] uh CTQ and I'm going to call this as CTQ1. Okay. is it's the percentage of pizzas that got a feedback like uh a wall feedback. Do you guys understand that? 32:20 Speaker A And [clears throat] this is connected this is connected to the second CTQ in our list. 32:31 Speaker A Okay, just to make it super clear. And I'm going to take another CTQ [cough and clears throat] CTQ2 that is pizza campus campus harvest pizza cost. 32:57 Speaker A Okay. Uh in US dollars. Okay. Um and this is connected this is connected to the to the price CTQ that is one two three four five six six six connect to the CTQ in our list that is um 33:39 Speaker A just to guarantee it's super clear for you for everybody. So is pizza delicious? Is pizza delicious? 33:51 Speaker A Is pizza delicious? Uh yes or not? Yes or not? And um the other one is pizza price. Pizza price. 34:18 Speaker A Pizza price. Okay. From 0 to 10. How clear is that? [clears throat] From 0 to 10. How clear is that? 35:06 Speaker A These are the two CTQs that I'll be working here to model. Let's model the process and the slash product based um on these two CTQs. 35:20 Speaker A Okay. So now I am not talking about customers anymore. Come here with me. Now I am not talking about customers anymore. 35:34 Speaker A Now it's an internal conversation. Internal conversation inhouse. Type inhouse. Is it possible to consider? No, you must consider all CTQs here. Just because of time I'm going to pick two. But you should do in your real project ideally 35:55 Speaker A ideally ideally um this very same exercise that I'll be doing here for all the CTQs. Okay. 36:06 Speaker A And I selected one that is discrete data and another one that is continuous data. 36:17 Speaker A Okay. And so here [clears throat] we need to select we need to select [clears throat] we need to select uh the factors the factors that are critical per CTQ. Do you understand that? 36:46 Speaker A So it's a simple question to the pizzolo to the chef to the assistant. It's like tell me three things three things. Yes. So the question is tell me three things three factors that impact you know factors that impact CTQ1. 37:15 Speaker A Okay. Okay. So you guys tell me what what do you think would impact you know would impact the percentage of pizzas that got a wall feedback. 37:31 Speaker A What? Type here for me. Type here for me. condition of the ingredients. Fresh, not fresh. What else? 38:10 Speaker A Uh, pizza temperature in uh Celsius. Size of the pizza. Size of the pizza in inches. 38:29 Speaker A Uh, let me see what else. Appearance. Appear. appearance. Uh, exciting. Colorful. Colorful. Not colorful. 39:00 Speaker A What else? One more. Just one more. Yeah, I will try to separate uh waiting time to price to quality. Okay. So, let's focus only on quality. Condition of the ingredients. Pizza temperature, size of the pizza, appearance, like colorful, not colorful. 39:23 Speaker A Let's see one more. or greasy. Yeah. Or smell. Smell. Aroma. Aroma. Yeah. Yeah. Aroma. Beautiful. 39:42 Speaker A Beautiful. Um. Aroma. Pizza. Aroma. Roma. Um. [snorts] Good. Not good. Yeah. Because sometimes the aroma is bad and the and the flavor is good. Right. Beautiful. And again guys for this exercise you can place I don't know 100 factors 40:15 Speaker A and nowadays because we have so many alternatives to analyze for free or at very low cost with AI you know [clears throat] this is this is amazing and then I will do the same I'll do the same for the other city queue. Are you 40:34 Speaker A guys following me? Can you feel the flow tonight? Can you feel the flow tonight? 40:44 Speaker A Yeah, I hope you can feel the flow tonight. [snorts] I'll do the same for um I'll do the same for factors that impact CTQ. 41:15 Speaker A Okay. [clears throat] [snorts] And to be honest, sometimes a factor that impacts CTQ1 can be the same of CTQ2. 41:27 Speaker A Okay. Now, I need you as pizza experts. Imagine that you work in the operations of Alys's Pizzeria and you tell me what would impact, you know, the the cost of the pizza. 41:44 Speaker A What would impact the cost of the pizza? What would impact the cost of the pizza? 41:50 Speaker A Type here for me. Yeah, probably ingredients is a very important one. What else? Let me know if you guys agree. Let me know [clears throat] if you guys agree. 42:38 Speaker A Baking temperature. H baking time. Baking time in uh seconds. Do you guys agree or in minutes? Yeah. 43:02 Speaker A Yeah. To be honest, appearance colorful. Not colorful, I believe. It may impact you know. 43:20 Speaker A Yeah. Let me see what else. Wow. Oven. Oven. Oven. Because we have let's suppose we have a new oven and I have uh an old oven and uh people are saying that we should change all the old ovens 43:40 Speaker A to the new ovens. Guys here do you agree that we are building hypothesis? Do you agree that we are building hypothesis? 43:53 Speaker A Do you agree that we are building hypothesis and now I need to model to model my process? 44:07 Speaker A Model my process. Okay. Model my process using um these factors. Do you guys understand that? 44:21 Speaker A Do you guys understand that? Let's model mathematically model. Okay, you don't need to guys you don't need to know the formulas and everything. I will be talking a little bit about about um math here but in a very simple 44:43 Speaker A way very simple way. Okay. And there are many different ways to do that you can do with softes. I will prefer for green belts to use AI to [clears throat] use AI. Okay. To help us. So take a look 45:01 Speaker A here with me. [clears throat] Yeah. We need to build to build an experiment. We'll be modeling designing an experiment. 45:16 Speaker A Okay. Very simple. very simple. It's almost ridiculously simple. Okay? So, come here with me and type simplicity is power. 45:33 Speaker A Simplicity is powerful. Simple is powerful. Okay. [clears throat] Very simple. Very simple. [snorts] Very simple. Come here with me. 46:00 Speaker A I'm going to take Okay. And and by the way, now I am already modeling my process. Okay. I am modeling. I have already identified the critical factors and I want to model my process. So I am in this step model process 46:24 Speaker A model process [snorts] okay model process and then there are many different ways to do that but the one that I'll be recommending is you just place all the factors all the factors Let's do let's do first of all 46:47 Speaker A for CTQ1. So condition of the ingredients. Beautiful condition of the ingredients. And I'm going to call this X1. This is X1. 47:03 Speaker A Okay. Simple. [clears throat] And it's fresh or not fresh. Amazing. Amazing. Simple. Beautiful. Then I have another one that is pizza temperature. 47:18 Speaker A That is X2 pizza temperature. Beautiful. Excellent. Amazing. Uh size of the pizza. It's X3. 47:35 Speaker A X3. Size of the pizza. Amazing appearance. colorful. Not colorful. Amazing pizza aroma. Pizza aroma. Good or not good. And then I will be creating different conditions. 48:06 Speaker A Interesting. Yes. Different experiments. For example, for example, um just for me to save some space here. 48:25 Speaker A Condition of the ingredients. Pizza temperature. Yeah. [clears throat] Maybe like this. So take a look condition of the ingredients. I can have a fresh ingredient. 48:48 Speaker A Uh pizza temperature. Yeah. So pizza temperature. I'm talking about the temperature when I get there. Okay. So let's define two different levels. Two different levels like a super hot and hot maybe. Um, best temperature to eat a pizza. 49:15 Speaker A [cough and clears throat] Okay. In Celsius. In Celsius. Uh, 55 to 60. 55 to 60. And um, and 40. 40 may be strange. Okay. Interesting. 49:34 Speaker A Interesting. So, let's test. Let's test 60 and 40. So, let me run a test with 60. Size of the pizza. [clears throat] Amazing size of the pizza. 49:52 Speaker A [clears throat] [snorts] Yeah. So family is 18 in large is 14. Okay. So let's test with 18 and let's test with 14. Okay. So 18 appearance colorful and pizza aroma. 50:16 Speaker A Good. And then I'm going to measure. This is uh x4. This is x4. [cough and clears throat] x4. And this is x5. 50:31 Speaker A And this is X5. Okay. And then here I have my CTQ that is CTQ1. 50:42 Speaker A CTQ1 [snorts] that is [clears throat] take a look here the percentage of pizzas that got a wall feedback because ideally we should design an experiment. 51:01 Speaker A We call this experimental condition and send [clears throat] 100 pizzas. I know it's expensive, but send 100 pizzas to my customers with fresh ingredients at 60° uh the size of the pizza 18 in colorful [clears throat] and good and then 51:23 Speaker A measure the percentage. Do you guys understand that? I [clears throat] know it's time consuming. It's it may be very expensive, but that's the right way to do percentage of pizzas. 51:39 Speaker A that got a wall feedback. Because if you run a good experiment in the right way, you you may gain so much knowledge, so much knowledge that you have a kind of formula formula, you know, a formula for the rest of your 52:04 Speaker A life. And obviously nowadays you can simulate that guys. How many factors do I have here? 52:12 Speaker A How many factors do I have here? How many factors? [clears throat] How many factors? 52:37 Speaker A And how many levels, how many levels, how many options I have in terms of condition of the ingredients. 52:56 Speaker A I have two. So there is a very simple formula in order for you to to design this experiment. And guys [clears throat] in our black belt course, I teach design of experiments from the very beginning. 53:15 Speaker A Okay? So here for green belts I'm trying to simplify as much as possible. There is two power five levels number of levels power the number of factors. So you guys tell me two power five what's the what's the [clears throat] 53:33 Speaker A result? Two power five correct two power five it's 32 32 so uh it's very simple I need to have I need to have 16 freshes. So, 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16. Okay. 54:19 Speaker A And and 16 not fresh. Okay. And by the way, we can have this [clears throat] just to show all the 32 all the 32 experiments. Okay. 54:49 Speaker A [snorts] Yeah. Some people call this a matrix, a contrast matrix. Pizza temperature 60. What's the other one? 60. What's the other one? The other temperature 40 60 and 40. 55:26 Speaker A So I got eight and eight. Yeah. Size of the pizza 18. I have four with 18 and four with 14. Remember that. 55:45 Speaker A Yoooo. And four with 14. Okay. [clears throat] Colorful. Um, two. Not colorful. Not colorful. 56:20 Speaker A So see that I did 16 by 16 then 8 by 8 then 4x4 2x two and now good not good and now one by one. Yeah. 56:34 Speaker A Uh guys we have softares that do this for us so easily. I am doing here by hand so you understand the mechanics of the experiment. So type mechanics if you understood that we have factors here. 56:55 Speaker A We have factors and I want to know how does my how does my the percentage of pizzas that got a wall feedback you know behave. I want to know how does CTQ1 behave as a function of X. 57:21 Speaker A Simple like that. [snorts] And then again, in order to collect the percentage of pizzas that got a wall feedback, we would need to run 32 experiments in this case. 57:36 Speaker A Yeah. And then let me place some some values here. Um yeah. Um, I'm going to show you guys in one second. 58:41 Speaker A One second. [snorts] Yeah. So let's suppose we have collected these percentages. Do you guys understand? Like we ran this experiment here with fresh vegetables at 40 Celsius degrees. The size was 14. Colorful, not good uh in terms of aroma. And we got 59:40 Speaker A 90% of the students saying, "Wow, this is amazing." Yeah. And uh let's suppose we had uh fresh ingredients 60% uh 60° uh 18 in not colorful uh good in terms of aroma and we got 29% of the clients happy. Do did you guys 60:09 Speaker A understand that I ran a series of experiments? Just type experiments. No, I did not use any formula. No formula here. I did not use any formula. I [snorts] am just organizing a series of tests. Just that type experiment. If you understood that 60:34 Speaker A I'm just organizing a series of experiments so I can test if you have a table like this. If you have a table like this. 60:55 Speaker A Okay. If you have a table like this [clears throat] nowadays, it's relatively easy for you to do this. Take a look. 61:11 Speaker A You take a snapshot of that and you ask for your favorite AI app. Okay. 61:23 Speaker A Please generate a mathematical model for the CTQ1 based on the five factors presented. Um and also based on the results of please generate you know based on the five factors and then also based on the results of the 62:04 Speaker A experiment. Okay. [clears throat] Are you guys following me? So, let me copy let me copy this to my to my template. Yeah. So, this is model process for CTQ1. 62:41 Speaker A Yeah. And I want to see if AI gives to us a mathematical model. Okay. So using multiple linear regression. Interesting. Okay. So let's see. That's the mathematical model. 62:59 Speaker A That's the mathematical model. Okay. And with this mathematical model, with this equation, and again you don't need to be a master in math, you don't need to be a master in math or you can even do something like because this is the most 63:20 Speaker A important thing for us. Can you guys interpret that out of ingredients, temperature size appearance and aroma, which one seems to be the most important factor? You guys tell me which one seems to be the most important factor. 63:39 Speaker A Which one seems to be the most important factor? The size. [laughter] So as long as as long as your pizza is big enough, the students will be so happy. So if we are talking about the whoa you know our experiment showed that 64:05 Speaker A it has to do with size and guys you you know that it happens in real life type here real life if this is mimicking you know real life how many times in real life we have something in mind it's because of the fresh 64:19 Speaker A ingredients when we collect data when we use science then science shows wow it has nothing to do with fresh ingredients A student cannot differentiate a fresh broccoli to an expired broccoli from last year. You know, as long as there is 64:37 Speaker A no there is no stomach cake. [laughter] This is this is beautiful or what? So [clears throat] if this experiment is correct, we should pay full attention, full attention in the size of the pizza. Do you understand that we should pay full 65:02 Speaker A attention? It's kind of nonnegotiable. It's non-negotiable. You're you you must have a pizza of you know of uh 18 in. 65:15 Speaker A Do you guys understand that? Let me ask create a cartoon image um for the results of the experiment. I don't know if you guys agree, but this is beautiful. Type type tears. Type tears if you understand that. This is beautiful. This is 65:42 Speaker A emotional. This is emotional. This is beautiful. This is beautiful. [snorts] This is beautiful. Yeah. 66:10 Speaker A Then I need to do the same for cost. Okay. I just think for cost just because of time constraint. I'm going to reduce the number of uh factors. Okay guys. So I'm going to work with three factors only. 66:28 Speaker A Let me know if this is okay for you and I will repeat I will repeat the I will repeat the experiment. 66:40 Speaker A Okay. Um so the first one is for CTQ1 CTQ1 [snorts] and the second one is for CTQ2. 66:54 Speaker A CTQ2 Yeah. So, CTQ is um pizza cost. CityQu is pizza cost. Pizza cost. And we have cost of the ingredients. 67:20 Speaker A cost cost of the ingredients. We have uh baking temperature, baking temperature and uh baking time. 67:38 Speaker A And we don't have these two columns. Baking time. Baking time. And then uh when you have three factors, it's two power three. 67:59 Speaker A 2 \* 2 \* 2. So it's only eight. Only eight experiments. Yeah. So cost only eight. 68:11 Speaker A Yeah. cost of ingredients. It's high high low low. Yeah. Then um baking temperature. Baking temperature. 68:34 Speaker A Uh one second. Let's have a reference here. uh area baking sure. So we can run an experiment an experiment with um with 250 uh or Napoleonian would be almost 500. Wow. So 450 it's an Aolitania Napolitanian 450 um 69:13 Speaker A Celsius degrees. Okay. And uh 250 250 and baking time [clears throat] we can have from uh five minutes uh five minutes. Ah, I'm so sorry. Here it's uh 250. 69:38 Speaker A 250. Yeah. 2 and 2. And baking time 5 minutes or 15 minutes. 5 minutes or 15 minutes. 69:56 Speaker A Yeah. And so here it's in terms of cost. It's in terms of cost. Okay, in terms of cost and then let's suppose let's suppose we have here $20 $21 $22, $19, and here we have $7, $6, $5, and $6. Okay, 70:32 Speaker A beautiful. So let me take and again guys I'm sorry if I am being repetitive but [snorts] do do you understand that we would need to run a kind of experiment buying expensive ingredients baking for 450° 5 minutes and then we evaluate the cost. 70:57 Speaker A We have a finance finance expert close to us and I know it takes time. I know it costs you know a lot of money depending on the experiment but you will be learning so much. 71:12 Speaker A [snorts] Yeah. And then let's suppose okay we saw that this pizza the the total real cost was $20. 71:24 Speaker A Yeah. Why two? Why two? Mariza, I don't know if I got What's your question? 71:42 Speaker A I thought the last column should be number of people. No, no, no. This is for city Q2. CityQ2. 71:50 Speaker A For city Q1, it's percentage of people that reacted. Like, wow. From 0 to 10, how much did you guys understand what I did so far? From 0 to 10, I totally agree with you. If we do for trial and error, we would spend so much 72:12 Speaker A more time and probably not learning anything. Yeah. So [clears throat] again, okay. Okay. [snorts] Okay. Okay. I see many t but I'm going to explain again. 72:36 Speaker A How how much focused are you on me right now? Attention ladies and gentlemen. How much how much focused are you on me from 0 to 10? This is very important what I'm showing here. Very much important. 72:55 Speaker A How much focused from 0 to 10? Okay, remember that we had a list of CTQs. 73:06 Speaker A Remember that we had a list of CTQs [snorts] that we obtained in quality function deployment and we used in Q matrix to select our best concept. Type remember if you remember that we have a list of CTQs in design phase where we are right now 73:39 Speaker A we should run a proper analysis we should finetune fine-tune We should detailed design the [clears throat] concept for each CTQ. 73:59 Speaker A I selected only two because of time just that and so you can understand how to model the process for all the other CTQs. 74:12 Speaker A Type okay if we are on the same page right now. So we identified things that can potentially impact you know the percentage of pizzas that got a wall feedback like condition of the ingredients, pizza temperature, size of the pizza appearance, pizza aroma. We 74:39 Speaker A also got factors that would [clears throat] impact cost. Yeah. So far so good. Yeah. From the experts. And now I am trying to model the process. 74:53 Speaker A How do I model? Via an experiment. It's a kind of DOE design of experiments. 75:02 Speaker A So I am checking what would be the percentage of people that would celebrate if the ingredient is fresh, pizza temperature is 60, size of the pizza is 18, uh it's colorful, the appearance is colorful and the the aroma 75:16 Speaker A is good. And so we did a series of experiments and we [clears throat] virtually collected the percentage of people that got this wall feedback. And then based on that we found the mathematical model that showed us that the most important 75:33 Speaker A thing is not the condition of the ingredients, it's not the temperature, it's not the appearance, it's not the aroma, it's the size of the pizza. Size of the pizza drives that reaction of wow. Size of the pizza. 75:51 Speaker A Size of the pizza. Again from 0 to 10. How much did you understand that this is a very rich information? 76:02 Speaker A This is PI process intelligence. This is very rich because we have so many things to control. If you can rank the things to control and say I I I may make mistakes in the freshness of the ingredients but I cannot make mistakes 76:20 Speaker A in the size of the pizza. I now now I like the explosion of dance [snorts] and that's one of the most exciting things for me. I love to find a different way to teach you guys. Do you feel that I care about you guys 76:42 Speaker A understanding this? Type love. Type love. Type love. If you understand that it's important for me that you if you feel that it is important for me that you understand that And I hope you understand that this is powerful guys. We are speaking serious 77:18 Speaker A things here. Yeah. I hope I am promoting a revolution in your brain in your heart. Okay. Then now I will repeat I'll just repeat the same mechanism. 77:36 Speaker A [snorts] Take a look here. So we checked like fresh ingredients not fresh pizza temperature pizza size appearance pizza aroma but by far the size was the most pizza size is the game changer. 77:52 Speaker A [laughter] Yeah, it's a game changer. So I will just repeat this for cost for city Q2. Yeah. So for CTQ2 I have Yeah, guys, just one thing that you must keep in mind. 78:37 Speaker A Many times when we run these experiments, it's like you used a f sometimes a super fresh ingredient and and a okay a fresh ingredient. So customers didn't notice that. But if you understand that, so in the ranges that 78:54 Speaker A we have operated here, the conclusion is size is the most important thing. Does it mean that we should not care about how fresh the ingredients are? Not necessarily in the range of your experiment. It doesn't make any difference. Yeah. 79:16 Speaker A [snorts] Yeah, beautiful. Um, so let me take this and model. I need a model. 79:43 Speaker A I need the same Mathematical model modeling you did before please and then I'll just take the equation I'll just take the equation And I'll just take the See, in terms of cost, what is by far the most important 80:37 Speaker A the most important factor? You guys tell me. In terms of cost, what is the most important factor? 80:46 Speaker A What is the most important factor? Cost of ingredients. And guys, does it happen in real life? 81:07 Speaker A Does it happen in real life? Yes. And what's the importance of you understanding of you understanding the critical factors because now you can um decide on which aspects of the process you need to focus to give more attention. 81:33 Speaker A Yeah. Isn't it beautiful? Isn't it beautiful? And so uh beautiful. What do we have? 81:43 Speaker A What do we have now? This is for city Q for city Q2. Yeah. for CTQ2. I'm going to show I'm just asking AI to generate an image for us because now we need to optimize optimize process process. Yeah. And again imagine that 82:25 Speaker A you are repeating this exercise for all CTQs. type here all CTQs. If you understand that in a real full DFSS project, you would repeat what I just did here for all CTQs. Okay. 82:44 Speaker A [snorts] For all CTQs and then question for you, is it possible that one CTQ says I want expensive ingredients and another CTQ says I want cheap ingredients. Do you guys understand that? Type trade off. 83:21 Speaker A Trade off. If you if you if you understand that is it possible in real life that one CTQ says I want fresh ingredients expensive expensive ingredients like quality and another CTQ says I want cheap ingredients. Yes, it is 83:41 Speaker A possible. And [snorts] how do we solve that? How do we solve that? Now we need to run what we call a multi-objective optimization. 83:57 Speaker A Multi-objective optimization and you need to inform the software or AI the importance the importance. 84:13 Speaker A Okay. So take a look here of each the importance of each CTQ. So it would be something like this. Okay. Um something like this. 84:27 Speaker A This is for city QU. Okay. So this is for cityQu. So take a look. 84:42 Speaker A Considering that CTQ1 quality is twice more important than CTQ2. Let's suppose this is the case. Okay. 85:04 Speaker A Let's suppose no. Let's suppose no. our our our CTQ matrix, our CTQ list here already indicates that already indicates that quality is pizza delicious is more important than price. Okay? And by the way, if you want, you can use these 85:26 Speaker A numbers 15 and 9 if you want. Okay. Uh just to keep it simple considering that CTQ1 is twice more important than CTQ2 price um optimize this process and indicate the best settings. 85:56 Speaker A this process and indicate the best settings considering both CTQs sim simultaneously simultaneously. Do you guys understand that I am asking AI or the statistical software to optimize my process considering both CTQ simultaneously and I know that they have that they have 86:24 Speaker A different importances and and then you should do this for all the CTQs and then you let the software let the software and let the AI find the best settings the best detailed design for the process or for the product. 86:44 Speaker A You know with that and indicate indicate how much we have how much we have um met the CTQ targets. Huh. Now considering that cost should be less than $10 US and percentage of wall reactions should be more than 20%. Do you guys 87:29 Speaker A understand that you are setting now targets limits for the city kills? [snorts] Yeah. Gold. Is this gold or what? [snorts] Multiobjective optimization. 87:52 Speaker A Yeah. And there are many softes. Many many softares. many softares that can do that for us. Okay. Uh and including AI. I am using here chat GPT but you can use any AI app that you like. Um it's like I think it's $20 $20 88:14 Speaker A per month. Yeah. And so see this is very powerful guys guys please please please please please please please this is very powerful using the two previously developed main effects regression model the following combination satisfies both targets. 88:37 Speaker A Yeah both targets. So let me copy this to our template. This is very valuable type type mindblowing if you fully understood that. This is mindblowing. 88:55 Speaker A If you fully understood that this is mindblowing. This is mind-blowing because now I know that uh create a image please. You know that you have the campus harvest you know pizza but more than that now you know that your 89:26 Speaker A ingredients should be fresh. Pizza serving temperatures should be 60°. Pizza size must be 18. Appearance must be colorful. Aroma good. Ingredient uh cost can must be low. Baking temperature 250. Baking time 15 minutes. So you have more details 89:50 Speaker A more details about your concept. Now you have what we call detailed design. This is the detailed design. And imagine that you repeat that for all the other CTQs. 90:03 Speaker A Then you are getting um an instruction instruction that is so rich. It's like the Coca-Cola formula. 90:18 Speaker A No more trial and error. [laughter] I love that. Yeah. It's guys, it's a way of living. And I don't know if you agree with me but it's a way of living and you apply this to so many things you 90:33 Speaker A know you can apply this to so many things if you are building your house you can use that you know how to minimize cost and maximize what's important and then when you apply this please listen to the VW 90:48 Speaker A have I talked about Voice of wife listen to your life. [laughter] If you are applying this to design a vacation, to design your new house, to design to decide on the next car that you are buying. Listen to the voice of 91:10 Speaker A the wife. The voice of the wife is more important than the voice of the business, more important than the voice of the customer, more important than anything. 91:27 Speaker A Guys, this is applicable for anything. If you want to create, you know, anything in terms of process, product service. 91:49 Speaker A Yeah. So at the end of this design phase, I hope we can get what we call a detailed design. 92:03 Speaker A A detailed design. Yeah. So take a look here. [snorts] Take a look here. Yeah, this is a little bit too technical desiraability. This is uh uh this is very helpful for multiobjective optimization. 92:27 Speaker A Yeah, if you join the black belt program, you will fully understand that. But the predicted result for CTQ1 is 32% you know and the target is 20%. The predicted result for cost is $5.5. So below $10. Yeah. The key takeaway is 92:49 Speaker A this is so beautiful. Then we have the quality factors. This is beautiful. Right? Type beautiful. If you understand that this is beautiful, this is science. This is and you may think Marcelo but then I need to spend a lot of time guys. How 93:06 Speaker A much time do you spend with trial and error? How how much how how much real work do you have today? How much real [snorts] work? 93:18 Speaker A and trial when error I mean when you find the solution quote you don't know why you don't know why so and then I'll just ask to create uh create the final image create a final cartoon Um, is it helpful the cartoon cartoon 93:49 Speaker A images? I think it's very it's it's friendly. It's colorful like these images. Yeah. [snorts] Final cartoon image of the detailed design. The final campus harvest pizza. 94:15 Speaker A Yeah. And now we are going to the last phase. The last phase is extremely quick and easy to be honest. Yeah. Cuz now we need to check if we have designed for six sigma. If we have designed for six sigma. So we need to 94:39 Speaker A uh produce a certain number of pizzas and collect the data of percentage of people that had the reaction you know the reaction of um the reaction of wow this is amazing. Yeah. And so there is a concept that is you need to check if the process 95:07 Speaker A is varying [snorts] uh and all processes vary and then you calculate something that we call sigma level and for green belts for green belts I want to keep it as simple as possible. Okay, sigma level varies from minus infinity to plus 95:28 Speaker A infinity and the greater the better. Six sigma is 2 ppb parts per billion problems per billion. 95:39 Speaker A Okay. So once again let me know if you understood that now we should there is a risk analysis that I'll be talking about in a second but um what we would be doing is to collect data for all CTQs. 95:56 Speaker A Yeah. Ideally we should meet six sigma for all CTQs. [snorts] Yeah. And then if we have a three sigma, it means that we have 99.73% of good pizzas or pizzas in the right cost. 96:22 Speaker A If we have sigma level equals two, it means 95%. Okay, at this point I would not worry about that too much. Again, if you want to fully understand to fully understand everything and know more tools, I strongly recommend you joining 96:43 Speaker A the black belt program. Okay guys, um [clears throat] the design for six sigma black belt program is very powerful and you know that now that I am talking to the best of the best of the best of the best. Black belt is for you. You 97:03 Speaker A know that all of you guys are fully ready for black belt. you will become a certified green belt in design for six sigma and your next step is black belt and I [clears throat] don't know if this is good news or bad 97:24 Speaker A news but technically speaking the BFSS is lighter than lin sigma so the statistical tools are much simpler we don't depend on one certain software Okay. It's it's more qualitative and in my humble point of view nowadays it's more relevant. 97:53 Speaker A So if you want to join black belt you can't miss the registration offer next Monday. Do you [clears throat] guys understand that? Next Monday I will be opening up the registration process for black belt. Green belt is free. 98:14 Speaker A Black belt is paid. Type honesty if you understand that. Green belt is fully free. Black belt we have an investment. 98:25 Speaker A Okay. But on Monday [clears throat] the investment will be unbelievable. Unbelievable. Unbelievable. the special price, the special fee, the special offer for black belt in design for six sigma that you will see on Monday is unbelievable. Unbelievable and 98:54 Speaker A simple, simple, simple. You become a certified green belt in design for six sigma and you will become a certified black belt in design for six sigma before Christmas. [laughter] Do you receive it? Do you receive it? I see you guys becoming a black belt in 99:15 Speaker A design for six sigma a powerful simpler faster you know method before Christmas this is for you this is for you and and on top of that regardless certificate and PDF and verifiable certificate that you guys will have it's the knowledge 99:41 Speaker A The knowledge the knowledge that you will be gaining the knowledge that you'll be gaining is transformative transformative. Transformative I want to break the way you think and I want to redesign the way you think. Oh my god. I want to apply design for six 100:03 Speaker A sigma in your heart and in your mind. And I am doing that because you are already a different person. 100:11 Speaker A You are already a different person and I want to keep redesigning, reshaping. I am reshape. I am no we are we are we are together reshaping [clears throat] yourself. Okay [snorts] guys, verify is kind of quick. So I'll 100:41 Speaker A go a little bit fast. Okay, there is a an amazing tool named FMEA. But for green belts, I don't present FMEA. I present for black belts. Okay. So I will just ask I'll just ask you know um please 101:02 Speaker A create image for risk assessment for this design. Do you guys understand that? It's like you calling the most negative people. 101:21 Speaker A Call the most negative people to see your detail design and indicate places where we can have problems. Do you understand that? Do you understand that? Simple like this. I need you to call your negative people, [laughter] not negative, the most critical people, 101:53 Speaker A you know, and say please criticize. Please, I am asking you criticize. Go ahead and criticize, criticize, criticize. 102:08 Speaker A What can go wrong? Because then evaluate risk. We can anticipate problems. Do you guys understand that? 102:19 Speaker A That's our idea. We want to anticipate problems. And that's it. Very simple. Very simple. 102:26 Speaker A So, let's suppose and this is something that you can have a first version with AI as well. You can have a first version with AI as well. 102:42 Speaker A Okay. So, take a look here. Uh, ingredients quality and supply risk inconsistent freshness impact medium affects flavor appearance and wall likelihood. Medium mitigating use multiple suppliers incoming inspection. 103:02 Speaker A Define freshness criteria. type powerful if you understand the importance of running a risk assessment. 103:12 Speaker A Cost variation. Whoa. Whoa. So you need to have approved supplier list. So if you cannot buy from one, you buy from another. 103:23 Speaker A The risk of in the baking process, incorrect temperature. Yeah. Food safety. operational execution, aroma and flavor, appearance, consistency. So at this point we understand that the overall risk is low to medium. Wonderful. And we always always work to mitigate risk. 103:53 Speaker A Always always this is a continuous task. You guys understand that, right? This is not a one-time event. You need to constantly constantly uh Okay. Is that clear? Is that clear? And now I'm going to ask um generate random data for CTQ 104:23 Speaker A 1 and CTQ. Um plot a just a graph. I'll not be talking about control charts nor capability analysis plot a graph um and calculate sigma levels uh considering the targets for both CTQ. So what what am I doing 104:53 Speaker A here? I [snorts] am collecting I'm I'm building pizzas campus harvest pizzas and I'm collecting data for these two CTQs and calculating signal level simple like that simple like that uh generate image simple like And yeah, simple like that. uh second 105:40 Speaker A and then I hope that for all CTQs I get the sigma level that is greater than six. So we have designed for six sigma but in reality if you have more than three sigma you would be very good 106:03 Speaker A you would be very good very good okay so you don't it's interesting you don't need necessarily depending on your process to design for six sigma for all CTQs Not necessarily to be honest in real life. [snorts] To be honest, you don't 106:32 Speaker A you don't you don't. So, uh this is more real life, you know. So you may argue like oh Marcelo but I would love to see you know six sigma let me explain this let me explain this for you six 106:59 Speaker A sigma is two parts per billion two parts per billion you know uh one second one second one second guys let me just do a little bit different uh and this is a little bit technical But um don't consider [clears throat] 107:21 Speaker A uh 1.5 sigma shift and um improve this performance improve this performance um reducing variation umation level cost more than six sigma and quality around 4.5 sigma. Yeah. Uh there is this is a little bit too technical that sometimes they add up 1.5 sigma you 108:06 Speaker A know and uh this is this is not good. This is not good. uh and I'm asking also for CHP to reduce variation so we can have a a better performance um the performance can be improved by mean the the shift of the mean and or 108:27 Speaker A the reduction of the variation in terms of mean we are wonderful but in terms of variation since I asked Chachi to generate random data I found the variation uh too high so I I'm just asking to reduce variation. Okay. 108:47 Speaker A And then we are getting to the end. I don't know if this is good or bad. I I am not very good with goodbyes. I hope this is not a goodbye. 108:58 Speaker A Yeah, we are getting to the end of the technical training content. Let's say we have just to create an SOP create a standard operating procedure. So verify final um performance verify final performance. 109:34 Speaker A [snorts] Final performance. Let me see if AI generated something cool for us. Yeah, I believe they are still using 1.5 sigma shift. Okay. But considering our learning objectives, this is perfect. 110:06 Speaker A This is beautiful. So, take a look here. Take a look here. And please don't don't get don't get overwhelmed with the technicalities. Okay? Let's keep it simple. Let's keep it simple. Keep it simple. For CTQ1, I have collected. 110:24 Speaker A See the percentage of people that had that reaction. Our target is 20%. And for green belts, the only thing I want you to know is that after you have your risk assessment, you should collect data and evaluate against target. Just 110:45 Speaker A that just that just that okay the only thing I want you to understand as a green belt in design for six sigma now is that we need to plot our data over time and compare against target just that 111:04 Speaker A okay so from 0 to 10 how much did you understand the principle how much did you understand the principle of risk assessment the principle of final verification. 111:20 Speaker A Yeah. And once again in real life three sigma many times is amazing. Amazing. Amazing. 111:29 Speaker A And then the very final step is to create the SOP. Is to create SOP. 111:43 Speaker A create SOP standard operating procedure and again we can have standard operating procedures. We have people from governance here we can have a standard operating procedure of 100 pages. So here I will just create a simple representation. Okay, simple 112:03 Speaker A representation standard operating procedure is really the reference that we'll be using to train people for training for auditing. 112:17 Speaker A Okay, for all this is the three pillars for standardization. Create SOP training based on SOP and audit. Check if um how adherent we are to the SOP. create um um for SOP. 112:51 Speaker A Yeah, create SOP training and audit. Yes, realistic SOP. Guys, I have something to tell you. your engagement in this chat. I feel like you guys are here with me. 113:07 Speaker A October 22, 2026. I feel like you guys are here in Brazil with me. Yes, I I feel I feel you guys here close to me. It's amazing. Thank you. And um did you guys enjoy Did you guys enjoy this simulated 113:27 Speaker A project? Was that helpful? Yeah. Was that helpful? Was that helpful? Yeah. So, let's take a look on the representation of the standard operating procedure. Okay. 113:56 Speaker A Let's take a look here. Uh, prepare ingredients. Use fresh, good quality vegetables. Wash and sanitize. 114:09 Speaker A Slice consistently. Keep ingredients refrigerated. Prepare the dog. Assemble the pizza. Prepare the dog. Portion to desired size. Beautiful. Ah, it's not centimeters, it's inches. That's okay. 114:25 Speaker A Um, let's change that. [laughter] Assemble the pizza. Yeah. Spread a uniform layer of tomato sauce. Add mozzarella cheese. Distribute vegetables evenly. Beautiful. Preheat the oven. 114:38 Speaker A Bake the pizza. Beautiful. This is maybe this is OCD, but let me let me do that. 114:51 Speaker A Um pizza size should be um 18 inches not cm. Is that correct? [whistles] Just 18 in. 115:36 Speaker A 18 in. How amazing is this pizza? 18 in. Yeah. 18 cm for a pizza. 115:47 Speaker A No no no. Then the students will beat us up. [laughter] Yes guys. So, what do we have in terms of test? Do you guys understand that this is a multiplechoice test? 116:12 Speaker A So you will be receiving a link [clears throat] on Sunday. Sunday. Next Sunday 22nd, 34th, 5th. 116:24 Speaker A Yeah. 25th. Okay. Next Sunday at uh 6:00 p.m. [clears throat] 6:00 p.m. Philippines Singapore time. you'll be receiving via email and also via WhatsApp. Remember that email the deliverability is tricky. Sometimes you get the emails, sometimes you don't. So 116:53 Speaker A the deliverability of email um is tricky. So what I would recommend for you is to join our WhatsApp group. Okay. 117:03 Speaker A So if you have not registered yet, please do so. So you have until Sunday to review the sessions and take notes. 117:14 Speaker A And even if you have people that would enjoy obviously that live is much better. Live is much better. But if you have people that would enjoy this session, they can register and take the recorded sessions and wait for the 117:32 Speaker A certification test on Sunday. Whenever you receive the link, you click, you put your name in the way you want your name to be presented in the certificate. 117:47 Speaker A Write down your name in the way you want your certificate to be presented. You'll be taking these 60 questions and if you score 60% or more, so 36, if you score 60% or more, you will be receiving immediately 118:03 Speaker A immediately the certificate. Okay? If I fail, don't think in failing. Don't think in failing. You can retake as many times as you want, but don't don't think in retaking. 118:18 Speaker A No, don't send your name to me now. Don't I don't need your name. Don't send you will put your name when you take the test. Do you guys understand that? 118:30 Speaker A And then on Monday, I will be correcting the test. I'll be reviewing the topics according to according to your results and then we'll be together. 118:44 Speaker A Is that clear? Wonderful. Amazing. So from Sunday 6 PM, you have until Monday 8:00 PM. So you have 26 hours to take the test. On Monday, Monday 8:00 p.m. everything goes offline. Type here offline. If you understand that on 119:07 Speaker A Monday everything goes offline. Everything, the sessions, the test, everything goes offline. [snorts] And then I just run the final session celebrating uh correcting the test and also opening up the registration for black belt. 119:27 Speaker A Okay. Excited. Yeah. Excited. So, so Alissi gave us a mission to design the best pizza ever. So, we had a chance to identify the proper segment and then we collected the voice of the customer in this segment. Then we were able to 119:58 Speaker A classify this voice. It must be linear and the lighter. Then we prioritized using pair wise comparison. 120:09 Speaker A We found the most important. We were able to rank the voices. We translated the voices into CTQs and then we found the best concepts. We generated different concepts, different pizzas to meet the CTQs. Then we found the best concept. We improved the best 120:32 Speaker A concept. And then we understood the factors that drive the very best concept using a mathematical model and optimization multiobjective optimization and then we had a detailed detailed design on the best concept. Then we ran a risk analysis. We validated the performance 120:57 Speaker A and also created an SOP standard operating procedure to guarantee that we are maintaining the positive results and handing it over to the process owner. 121:10 Speaker A And now it's time to celebrate [laughter] guys. Thank you so so so so very much. 121:18 Speaker A I'm looking forward to seeing you very soon. Yeah. And let's finish with [music] I can't sing. I can't sing, guys. You guys know that. Yeah. [music] [music] [music] [music] [music] [music] [music] [music] [music] [music] [laughter] Guys, thank you so so so so so so so so 123:28 Speaker A very much. I will see you soon. I feel blessed to have you guys as my students. 123:34 Speaker A Bye-bye. Topics: Design for Six Sigma DFSS Lean Six Sigma Pugh matrix CTQ voice of the customer concept selection process design certification exam sigma level Study this video - [Flashcards from this video](https://sozai.app/tools/ai-flashcard-generator/?from=transcript&id=82952) - [Quiz on this video](https://sozai.app/tools/quiz-generator/?from=transcript&id=82952) Free, made by AI from this transcript. No signup. --- This is the markdown twin of https://sozai.app/transcript/i-can-t-create-a-meaningful-slug-from-session-04-alone-since-the-title-doesn-t/ — the same content, without the markup. Published by SozAI (https://sozai.app). Reuse and quotation are allowed with attribution and a link back. Machine-readable index: https://sozai.app/llms.txt · data API: https://sozai.app/api/