Mastering 8D Problem Solving: Crisis to Containment (D1-D4)

1. Welcome & Introduction: Hello. Welcome to Mastering AT problem solving. Chris is the containment D one Dvocus. My name is Meta and I'm thrilled to guide you on this essential journey into mastering the early states of the AT methodology. Before we dial into the course, let me tell you a little about myself. I bring over ten years of experience in quality management and having worked with leading automotive companies in Turkey and in the United Kingdom. My career has been deeply rooted in IATF 16 949 standards, where I gain expertise in, effective problem solving methodologies in high pressure environments. My goal here is to share these insights with you to help you excel in your own journey. Now, let's talk about what makes this course unique. This course focuses specifically on the critical D one and D four states. The first step is defining problems clearly and then assessing risk for similar parts and processes and the next is conducting initial analysis and the last, creating robust containment action plan. These early states are the backbone of successful problem solving because they lay the groundwork for all the actions that follow. If they are overlooked or poorly executed, the entire problem solving process can fail. We will move beyond theory into practice by exploring a detailed real world casetudy from the automotive manufacturing industry. This hands on approach will help you connect the concepts to real challenge you might face in your professional life. One of the standard aspects of this course is our focus on often overlooked arrays. For instance, we will deep dive into how to perform proper similar part and process risk assessments. Highlight key points in initial analysis that are frequently missed even by experienced professionals. By the end of this course, you will not only understand the D one and D four steps, but you will also know how to apply them in a structured and effective way to resolve issues confidently. While the next stage, D five and D eight will be covered in a separate course to be published in the upcoming months. This specific course ensures a solid foundation for structured problem solving. Let's illustrate with an exam. If the fire is the problem, then the extinguishing the fire is the containment action, which should be addressed in D one and D four. Therefore, in this course, we will learn how to put out the fire effective. Whether you are Nia problem solving or looking to refine your skills, this course is designed to provide tools, insight and techniques to set you up for success. Together, we will work step by step to ensure you are ready to tackle challenge, head on, and turn problems into opportunities for growth. If you are ready to dive in, let's get started on mastering A problem solving together and put out the fire against the crises. 2. What is 8D: Before going into the steps, let's understand first what is 80 and why is 80. The term ET stands for eight disciplines which represent the systematic steps required to identify, analyze, and resolve problems effectively. AT is one of the most effective and widely recognized problem solving methodologies. It was first developed by Ford Motor Company several decades ago and quickly became a standard across all the automotive industry. Over time, its application extended to other manufacturing sectors and beyond. Let's take a closer look at these disciplines and understand how they contribute to solving complex challenge. 80 steps can vary depending on the organization. It is normal to encounter slight differences in the steps across various 80 methodologies. However, all of them fundamentally include steps related to problem definition, emmy de action, root cause analysis, permanent corrective and preventive action, and closure with some lesson learned. In our 80 in the D one, we will explore the 5w2h technique, a high effective tool for clearly defining a problem. Additionally, we will look at the bot side of the problem. One is about the detection, which can be done by our customer or next operation in the same plant or ourselves directly. The second is about the origin point of the problem, focusing on occurrence details. We will learn how to approach and ask questions to grasp all necessary details and make a synthesis of the problem. Moving on to D two, we will assess similar parts and processes to evaluate the risk of potential issues in other areas. At this point, we will learn to manage our doubts regarding other similar parts or processes. For example, what if other similar parts have the same failure? Or what if we use the same defective parts in other products? Or what if we ship the same defective parts to another customer plant, for example. All those will be clarified precisely in this step. D three focuses on the initial analysis phase before the containment action. Where we examine the non detection point, it is where the failure should have been detected and why we couldn't catch this. And the basic conditions check by checking the relevant production area to understand if any anomaly. This may sound like a root cause analysis in early stage which is done in the next steps, but it is not. In this initial analysis step, we will understand the basic conditions of our process and the non detection part in order to put effective actions in the containment actions. D four addresses containment actions, emphasizing immediate measures to mitigate the impact of the issue. We will see what actions we should get and which ones are suitable for our problem. Containment actions are at least as important as the corrective actions because in this step, we bring under control decreases, which is extremely critical for us to be able to manage all problem solving study smoothly and confidently. While this course includes D one and D four steps, I would like to also go through the next ones, which will be covered in the next upcoming months. D five serves as the one of the course step where we identify the root cause using three powerful tools. Those are fish bonuses where we make the brainstorming analysis and failure three analysis and the 55 technique. Establish a clear connection between these tools, guiding us to uncover the true root causes of the problem. The six is the action plan and the implementation where the corrective and preventive actions are taken against root causes. Following step, the seven is the validation of these actions to be sure they are effective or not. Reproduction of the defect is the main approach here to demonstrate the corrective actions are effective or not. And finally, the aid, it serves as the concluding step where listeners are documented, conduct closure audits and update some standards and celebrate team's effort and success. Those were the steps of eight disciplines. Now let's see which areas we can use this methodology. 3. Where is 8D used: It is a versatile problem solving methodology that can be applied in almost any context where addressing and resolving issue is necessary. While it is particularly prominent in the automotive industry, where it is often required as part of the IATF standard, its utility extends to a wide range of production and service sectors, including aospace, healthcare, manufacturing, and IT, among others. This makes AT a universal tool for structured problem solving in both technical and non technical fields. 80 can be applied across various industries and in numerous sterations. Let's take a look at some of the most common scenarios with examples. First is the production quality issues. One of the primary areas where ATus is in addressing production related to problems. For example, suppose the dimensions of parts produced in manufacturing process begin to deviate from the required standards, or alternatively, a single defect is identified in an otherwise stable production line. In such points, AT serves as an excellent tool to conttisue and implement listing solutions against the root causes. Second, is customer complaints and warranty claims. Another significant use case for AT is managing customer feedback and product warranty claims. For example, a customer reports an issue where a product malfunctions after usage or a warranty claim highlights a recording defect in a specific batch or products. By employing AD, we can systematically investigate the issue and ensure customer satisfaction. Trees, again, the supplier quality issues. It is invaluable in supplier management to address quality concerns. For example, a supplier delivers raw material or components that fail to meet aggregate specifications, or let's say, there's a recurring issue with delivery timeline or material handling. Here AD helps not just in resolution, but also in strengthening the supply relationship through corrective actions. F is logistic and supply chain problems. Supply chain and logistic chains are another area where AT proves effective. For instance, a shipment arrives late, disrupting production plannings, or there's a repeated issue with inventory mismanagement. Applying AT allows business to pinpoint vulnerabilities and optimize a supply chain. Five is non conformity in quality management system audits. Audit findings and non conformity often require structured problem solving. For example, an internal audit reverse a gap in compliance with a standard operating procedure, for example, AD helps resolve these findings while driving systematic improvements. Next is safety and environmental concerns. It is also widely used to address safety and environmental issues. For example, a workplace accident occurs or an environmental compliance breach is identified. Using AD here, ensures preventive measures are implemented effectively. Next is continuous improvement. Beyond solving the problems, AT can also be applied proactively for process improvements. For example, identifying and eliminating inefficiencies in a production line. This makes AT a versatile tool, not just for firefighting, but also for building a culture for excellence. Next is lastly project management. It is also utilized in project management to handle unforeseen challenges during projects. For example, delays in project milestone due to resource constraints or coordination problems among cross functional teams, for example. With AT, such problems can be systematically addressed to keep projects on track. These are the main areas where we can use, but not limited. 4. D1 Problem Definition - 5W2H Overview: D one, problem definition. This is the first step in the AD process. And here, we will use five WH methodology to create a comprehensive problem definition. So let's explore these steps in detail now. 5w2h methodology is a structured approach to thoroughly analyze and define a problem by addressing seven key questions. First is, what is the problem? It clearly describes the issue, including any relevant details such as defect type or process issue. We should simply understand the issue when we read this. The clarity level should also have as much as details such as required and actual status. This can either be a dimension or some visual curterias let's give an example. A five windshield has a crack in the corner. Here, basically, as part of the question what? We understand the relevant product. It is a five windshield and we understand the problem itself. It is crack in the corner. This could be fairly enough for initial and clearly describe the problem. However, if they have more data like crack level, it is depth or length, we should absolutely identify to improve the city level of the problem. In that case, it would be like a five windshield has a four millimeter long, 1 millimeter deep crack in the corner. So now we understand the frame of the problem correctly. Second is where, where is the problem occurring and where was it first detected. Specify the location or a affected here, whether it's a specific workstation, production line or region is useful. For example, Warwick Plant, second production line, Operation 20. As much as details we'll make this description better. This was the occurrence place, but we should also identify the detection place for this question. Let's say for this London plant car assembly line. This is the detection place. Basically, we have two answers of those questions to describe the problem better. One is focusing on occurrence and the other is focusing on detection or the submissione site. Let's review the next question, this. When did the issue occur and when was it first identified? Here, the provide timeline highlighting whether the issue is consistent or singular and not any relevant trends is useful. In that point, it is important to use some graph to see the trend of the problem if it is not a singular case. This will show all the relevant variation of the problem, which will help us to resolve it properly. But on the other hand, for the singular case, that means if the problem seen just on one product or one case in process, we can identify the exact date and time of the issue without a trend graph. It will be like January 3, 2025 night shift 3:00 P.M. This gives us the data when did the problem occur. But as mentioned above, in addition to this, we should also define the date of detection of the problem. This was the time that the problem occurred. Then when it was detected after a few days or just the same time when for this example, let's say, January 5, 10:00 A.M. It is important to identify both occurrence and detection point in the problem definition. Next is who is involved or impacted. Here are the identified involved stakeholders, such as operators, customers, or suppliers is important. By the way, the problem doesn't have to be related with the operator as sometimes machine can fail due to some issues. But here, the main purpose is to collect accurate data from the key person who involved the problem. Therefore, it is important to identify those. Let's give an example. Problem was detected by JL in customer plans. We can also write his operator identification number, but for the occurrence, let's say relevant operator name is MT. Those identifications matter for us to get prompt and fast data by contacting them. In case the occurrence is not related with the operator. Let's say, even one operator doesn't work in the relevant production line, it is fully automatic. In such cases. We can identify the problem is independent from the operator for such cases. Next is why? Why is this a problem? Here, basically, we need to justify whether it is indeed a problem by comparing the actual state of the standard requirement in order to determine if there's a deviation. This question is often misunderstood and using why this problem happened, but it is not. We don't question here why did it occur because it is part of the root cause analysis. Our focus here is to understand why it is a problem. Let's give an example. We can use the previous windsheld crack example for this. Why is it a problem? Because windsheld quality requirement is no crack. That's why this crack is a problem. In that case, we can just indicate that the part doesn't comply with the quality requirements. Because requirement is no crack and the actual status is crack in the corner. Apparently, this doesn't complies. This is how we explain this question. In case the quality requirements are different by the way for both customer and manufacturer, we can also indicate both separately as well. In this example we just gave as. So our next question is how is the problem manifesting and also how was it detected? What is the story? Here we describe the mechanism, process or sequence of eons that led to the issue. Let's give an example for this. During the windshield assembly to the car, operator notice that there is a small crack in the corner. This clearly explains how the issue was detected. Let's give you an example for the manufacturing site for documents. We need to answer for the question of how did the issue occur in that point. The crack occurred during normal production process when the windshield was being shaped and tempered, resulting in stress at the corner. So we understand how did it occur with this answer. In addition, we ensured whether the process was in normal condition or not. That means we checked if there's any additional operation or rework, the additional control, et cetera. So basically, any difference flow from the normal process, as it is important because as part of the how question, we should also identify whether the defect occurred in a normal process or not. It is important. So another thing in that point, for example, if you are not sure that the occurrence point, if you don't know. For example, maybe it occurred during the transportation due to an external damage, not in production, and we don't know. In those uncertain cases, we can identify this by considering the relevant possibilities and we can update the section letter after collecting accurate data. Last question is how much? Here, we try to understand what is the severity of the problem at initial. Is the singular issue singular part or are there many parts affected quantify impact in terms of the defect rates here. From the same example, we can say one car affected at customer side and six wind shelt detected at crack, and there are 400 windsht are suspect. In this point, we indicate the initial status. It may be all those 400 parts are defective as well, but we don't know currently because we didn't check yet. We just know that one car has already been detected during the assembly and six more wind shots have also detected S not okay, which are stored by the car assemble line. That's why we just addressed all points clearly. So by addressing all these points systematically, we enjoy a comprehensive understanding of the problem, pairing the way for effective analysis. At this point in the course, now we will continue with the case study and we'll have practical wave for all those theories. Let's dial in and see what we have. 5. Exploring the Case Study: So now to illustrate the AT process, we will work with a case study. Now imagine we are a company and our name is MK hinge. Our core business is producing bonnet hints for the cars seen on the screen. And our customer, a car manufacturer company and named liver cars. They use our hints into their vehicles. Basically, we produce bonnet hints, ship them to liver cars, and they use these hints for their car assembly. This simple supply chain sets the base of our case study where we will walk step by step through the AT process. Now we have received a problem notification from our customer. Let's listen careful to what they're saying about the problem. Hello, we have a quality issue on a 49 bonnet hinges. The holes are not matching. We cannot assembly the part to the car. Please solve it immediately because we cannot produce our cars car assembly line just stopped because of this problem. So the customer mentions an assemble issue with our A 49 bonnet hints. They state that the holes are not aligning, it's preventing them from completing the assembly. This has caused their production line to stop. Let's ask ourselves, did we fully understand the problem with that description? Let's read again. We have a quality issue on a 49 bonnet hints. This A 49 is the project code of our hinge and the holes are not matching, and that's why they cannot make the assembly to the cars. Please so immediately because their assembly line just stopped. So at this point, the customer's statement might seem like a tangled ball of iron. While it is clear that there is an assembly issue with the A 49 bonnet ins due to the whole mismatch, we don't have enough specific data to move on because what hole, which hole, how the problem occurred, maybe they didn't assemble the operation well and our parts may be okay. We don't know anything. We just have information of our parts hole are not okay. It is quite limited information. So to proceed effectively, we need to understand by asking all those details to our customer, the problem submission site because they raise the problem. Also, we need to ask question our relevant production area because we produce those parts. So first, let's start by taking a look at how we produce the bonnet hinges. So this is our process to produce bonotinsF we begin by receiving sheet metals from our suppliers, which are then stored in the receiving material warehouse. Then those sheet metals are fed into the press operations. At first, we produce the first semi product at the first press operations, and these semipducts are then stored in an intermediate storage area located by the production line. And after that, we produce the second sempduct like the first, these components are also stored in the intermediate stock area by the line. In the final operation, we assemble these two semi products to produce the complete bonnet hinge and the finished products are initially stored in the intermediate area by the production line and then move to the shipment warehouse where they await for dispatch. So from there, we ship the complete bonnet hinge to our customers. Basically, as a summary, our production process involves creating two sempducts that are assembled to form the final product. It means bonnet hinge. The thing that we noticed here, the part has the horse which was mentioned by the customer during the problem notification is produced in the second operation because only the part has hole is produced in there. So basically, our focus point in the production process seems will be here, the second press operation. It is the key place that we need to focus on during the problem solving. This is our process and now we know how we produce those bonnet hints in our production. But we still don't know about the problem details because initially customer had just said that the whole mismatch about how and which whole, how many parts, a lot of questions in our mind because we don't know yet. Now let's go to D one problem definition where the 5w2h method will help us to identify all those details. 6. 5W2H: Applying the What, Where, When Questions to the Case Study: So since we understand our process, we can begin defining the problem using the 5w2h method in D one problem definition step. So the first question was? What is the problem? In this point, it is crucial to contact the customer or the next operations depending on the problem type. Who reported the issue to gather any missing data and clarify uncertainties. For example, customer head declared that the holes are not matching, but we need to understand which hole is it fully misaligned or assembly is possible with some additional force to define the problem more precisely by clarifying these details, we can absolutely refine the problem definition to ensure it is clear. After reviewing the process and collected the missing data from the customer by contacting them, we can identify that the second hull of the A 49 bonnet hinge is misaligned and making it impossible for the customer to assemble. As you see on the vision, there are three holes on both bonnet and the hinge. But the second hole of the hinge is mislocated, not matching with the bonnets hole. Defining the problem basically from the customer's perspective, it is a 49 bonnet hinges cannot be assembled to the cards because the holes are not matching. This was the description that customer notified. But from our perspective, after reviewing the issue and collected data, we can identify the second hole of the a 49 bonnet hinge is mislocated, making it impossible for the customer to assemble. This problem description will be our guide instead of the customer's one because we refined the problem description after collecting more data. This is what we did. So now we know that, what is the problem, which was the first question of the 5w2h. But in this point, we also need to ask that. Did we encounter this problem before? Is it a new problem or recurrence one? This question helps in leveraging past data because if we had the same problem before, then we can check our pest analysis and the actions and can understand why it didn't work at that time. So we can consider those previous data in our current analysis. But in our current case problem, in our case study, we assume that it is a knee failure, not repetitive. Second question is where, where was the problem detected and where did the problem occur? We need to address both areas as they provide distinct insights. The detection area highlights where the issue was identified and submitted, and the occurrence a focus on where the issue originated. From the customer's perspective, the detection area, we need to gather specific details such as which customer, which plant, which assembly line. After collecting this data from the customer, who raised the problem in our case study, we can identify the detection areas, liver cars, London plant, third assemble line. And additionally, from the supplier perspective, it is occurrence area. We focus on more detailed production data like plant, which production line, which production step once we have this information, we can define occurrence arrays. MK hints Glasgow plant, fourth production line, second press, le piercing operation. By answering these questions, we now clearly defined for both the detection and occurrence areas. What we do is here collecting the missing data from the stakeholders to identify those and this stakeholder can borrow own plant and colleagues or the customers, depending on the problem type, and of course, depending on our organization. We can now proceed to the next question in the five WH. Then in this step, we focus on identifying the timeline of the failure, covering both when the problem was detected and when it occurred. The aim here is to define the time period of potentially affected beds using traceability information. Because this helps us determine if the defect is an isolated, it means a single ratio, or if it impacts multiple beds within a specific time frame. So first we ask, when was the problem detected? According to the customer, the issue was detected on 22nd November during the night shift. We captured this because we collected this data from the customer by contacting them. And next, we investigate when did the problem occur? To answer this, we need to check traceability information for the faulty part reported by the customer. By analyzing this data, we establish the following timeline. The press part, production date is Vk 46. It is semipduct and finished product production date is 18 November, it is the final part, final inch last the shipment date is 19 November. So now we know those specific details of this defective part because when it was detected, it is clear 22nd November at night shift, we got this information from the customer who raised the problem. Now we also know that when we produce this defective part, we check the traceability data, which includes production date and part specific information and seeing that we produce this defective part in November 18 and shipment just in the next day. All those are clear and simple for now. However, it is critical to consider whether other parts produced during the same time period might also be affected. To do this, we need to check the bats produced around the same time and we need to identify the last okay part. This is a crucial threshold as it helps us to determine the point after which defects began occurring. In our case study, last part, production latest 16 November. This means all parts produced in the second press were good up until 16 November. Because after that date, something's change, we don't know yet and resulting in defects. Basically, our suspected time period includes the part produced between 16 November and 22nd November. 22nd is the date the claim was received, so that it is the last date as we didn't continue to produce defective parts again after received the problem notification. So now, simply, we clarify the suspected batch time period and we know in which dates we produce those not okay parts. It seems it is not a singular part, it's a chain, a lot of parts affected because it's a time period. Additionally, we can review the production records to assess to see who trend in case it also occurred before that. However, in this case, it is clearly a new failure, not a occurrence, no past occurrence recorded. It is limited between those specific dates. 7. 5W2H: Applying the Who, Why Questions to the Case Study: This step, we focus on identifying who detected the problem and who was involved in the creation of the problem. First, we determine who detected and reported the issue. According to the customer data, the problem was identified by Libor cars car assembly operator and ID number 87. Clear and simple. We just get this data from the customer or it could be a next operation or directly our plant regarding our organization. Next, we investigate who was involved in the production process, where the issue occurred. Since the defect originated at the second press operation, where the holes are pierced in the semi product, we need to identify the operator, which operator working at that station during the time of the defect. In this case, it was the MK hints press operator, idea number 27. However, it's also important to highlight that. Identifying the operator doesn't mean that they created or contributed to the defect. It just simply means they were working at that station during the issue occurrence. This information is crucial for collecting additional data directly from the source. In some cases, for example, in the station, no operator working, a fully automatic systems. It is okay to note that the issue is independent from the operator. But other than capturing who was working at the time helps establish a point of contact for further investigation. Because our goal here is not to place blame, but to gather accurate information from the appropriate individuals to better understand the situation. Maybe there was some animally during the operation, operator or related supervisor might know what happened. Maybe there was some extra operations or rework at that time. That's why it is very important to identify the relevant person for further investigations if available. Next one is why is this a problem? This question is often mistaken for why did it occur? As we already mentioned in the previous lecture. But the purpose here is not to analyze the root cause. Instead, we aim to determine whether this situation is truly a problem or not by asking why is this a problem? From the customer's perspective, the problem is clear. They cannot assemble the parts and their assemble line has stopped. This downtime results in financial losses, which will likely to be passed on to us because the issue is causing pos. To answer this question for the customers, parts cannot be used and assembly is impossible and the car assembly line has stopped. That's why it is a problem for the customer. So now let's consider this from the perspective of the bonnet hinge manufacturer. It is us in this case study. This question is critical to confirm whether the issue is truly a defect or not. In the other words, customer or the next operations can have some problem by using our products, but maybe it is not about our products, maybe about their process. For instance, if the customer experiences, assembly problem due to variations in the bonnet that are unrelated to our parts. We will inspect and verify that our products meet the specifications, our products are okay, in such a case, our answer for these questions will be no, and we will conclude that there is no problem with the hints. The issue lies elsewhere. So that this question as part of 5w2h is very important to justify whether it's a problem or not. However, based on our current findings for this case study, we know that the whole on our part is misaligned, meaning it is out of tolerance and doesn't comply with the required specifications. This confirms that the problem originates from our site. To answer this here, P dimensions are out of tolerance and don't comply with the specifications. With this justification, we confirm that why this is indeed a problem regarding those answers, we can enhance those by identifying the part dimensions to demonstrate the issue clearly, which is the best way. In our case example what we kept a bit simply to easy understanding. Now we addressed all aspects of the five W. We can proceed to the next to discuss about two H, how and how much. 8. 5W2H: Applying the How, How Much Questions to the Case Study: This stage answers two key questions. How was the problem detected and how was the problem occurred? Essentially, this is where we narrate the story of what happened. From the customer side, the car manufacturer, the sequence of event is straightforward. The operator placed a hinge on the bonnet. Tightened the first hole, but the second hole was misaligned. As a result, the assembly couldn't be completed. The operator then disassembled the part and informed the line manager. This explanation gives a clear picture of how the problem detected at customer site, which is the first part of this how question. Now let's consider our site, the bonnet hinge manufacturer. This needs to answer for second question of how was the problem occurred? At this point, providing a detailed account is challenging without further analysis, as we don't yet know exactly what happened during our production process because the free story will emerge during the analysis phase of 80. However, it is crucial to identify and document as much data as possible here. For instance, we already know that which press and operation were involved. How the operation is typically performed. So that we can identify it is in a normal process. During the whole piercing operation at press, second hole was pierced in a wrong position. In that point, our failure mode is about the hole location. We know that in which operation this happened. Therefore, we identify that it was the whole piercing operation, second hole was pierced in a wrong position, and this description basically tells the story of how it was occurred. In addition to this, we ensure whether any difference from standard process occurred during the production or not. For example, were there any revers or additional operations at that time? These questions allows us to identify whether the operation went in a normal process conditions or not. Input from the operators and spare wsors involved in the operation at that time is also valuable for this purpose. And that is the reason we identify this in a normal process. There was no work and additional operation. Operation steps were same as always. This statement narrates the issue while confirming that operation was in a normal process when the defect occurred. Currently, we have two stories here which tell how was the problem detected and occurred. Now let's see what we have in the next question. How much. The focus is on understanding the scale and severity of the problem. To achieve this, we evaluate each process stage step by step. At customer site, the customer has informed us that the hints are not usable due to assembly issue. Using the information from the VN stage, if remember, we identify the suspected pads. First, we evaluate the customer stocks to determine how many defective parts they have. Then we consider the possibility of problems in subsequent operations like the finish or salt cars. In this case, since the defect prevents assembly, we can confidently state that no defective hinds made it into finished cars. This eliminates the risk of further issues. However, just imagine, in other scenarios, undetected defects could lead to costly inspections of finished cars at the customer's end. But in this problem, it is beside the point because defective parts couldn't be assembled, so it is not possible to have defective cars because of this problem. So that we say the cars are okay, the problem didn't go after the car assembly operation. For this specific problem, we can identify as all bats delivered to the customer are defected, total eight boxes and 320 parts. In the next at our site, we evaluate our internal operations, starting from raw material for the production and transport states. Identifying suspected parts along the way. Let's review together. First is raw material. It is the sheet metal. The issue is not about the raw material, it's about the whole location. So we don't have anything about raw material. This tap is cleared. There's no problem. Second is the first press operation. This operation produced the initial semipduct unrelated to the defect because our defect is about the whole misalignment. In this operation, the parts are clear. It is not relevant. Next is the second press operation. So this is the operation where the defect occurred, misaligned. So all stocks after the states are suspected. Next is the assembly operation. Here, the defective sempducts were assembled into finish bonnet hints. So consequently, all parts at this stage are also suspected because we use the defective sempducts in this assembly operation, which were produced in second press. Regarding the next one, all finished parts in the shipment warehouse, which were produced during the identified period are also suspected. After that, intransit parts are also suspected. So quantifying the defective parts using data from 16 and 22nd November period. As this was the result that we found in van question, we identify 18, 20 suspected parts. This includes both semi products and finished products. If you make a breakdown of this, we have total 360 sempducts in the line, 200 finished parts in the assembly line, 860 finished parts in the warehouse, and 400 finished parts in transit on the way. All those parts are suspected and we will need to isolate them effectively to prevent further risk to the customer as part of the containment action in the next step. So now we completed all the answers of five WH, so we can make a summary and finish this section. Let's go to the next lecture to do this. 9. Summarizing the 5W2H Analysis: Now that we have completed the steps of 5w2h and finalized our D one problem definition. But one of the most important points, part of D one is visualizing the problem in the simplest and clearest way possible. Our issue is straightforward. The second hole of the hinge is not aligned with the hole on the bonnet, making assembly impossible. On the other hand, an okay part aligns perfectly, allowing assemble without any issue. To effectively communicate this problem, the best approach is to use a visual representation that contrasts not okay part with O part. For instance, we can create a side by side comparison like in the picture. In the not okay frame, we clearly highlight the misalignment of the holes in the okay frame. We show the proper alignment. We need to always emphasize the defective point on the visuals to focus attention and ensure the problem is easily understood at a glance. This clarity is crucial when presenting the issue to the teams, customers, or relevant stakeholders. By visualizing the problem in this way, we simplify communication, avoid unnecessary confusion, and set the stage for the next steps in the AT process. So customer site problem definition. Now let's summarize the 5w2h, which we already went through detailed earlier in the presentation. First is, what is the problem for the customer? It is a 49 bonnet hints cannot be assembled to the cars due to whole misalignment. The second is where was it detected? The issue was detected at Liver cars London plant on the third assemblage. Next is when was it detected? The problem was identified and submitted on the 22nd, November 2024. During the night shift in these questions, if we have more specific time, we could add that, but this is the level of detail we have, it is enough. Next is who detected. The person who detected the issue was the car assembly operator and the ID number was 87. Next is why is this a problem for the customer? Because the parts cannot be used. Assembly was impossible, so as a result, the car assembly line just stopped. This is why it is a problem for the customer. Next is how it was detected. The issue was detected when the operator placed the hinge on the bonnet, tightened the first hole, but seeing that the second hole was misaligned, the assembly couldn't be completed. That's why the operator disassembled the part and notified the line manager. This sequence explains how the issue was detected and reported. Next is how much? All hinge beds at the customer site are defective with a total of eight boxes, 320 parts. This concludes the customer site problem definition. Now let's move on to the hinge manufacturer side of the problem definition and see what we have. So before this, we focused on the customer side and ask questions based on their perspective. But now, our focus shifts to the manufacturer, so we will be looking at the occurrence point rather than the detection. First is what is the problem for the supplier for the hinge manufacturer? It is a 49 bonnet hints have a mislocated second hole, making assembly impossible at the customer's end. This description simply but effectively cows situation as demonstrated by the visuals already at before. Next is, where did it occur? The issue occurred in the MK hinge Glasgow plant, specifically on the fourth production line, second press at the Hoy piercing operation. Here we started from the plant address and drill down to the specific operation to describe better. Next is when did it occur? We utilize the not okay part traceability information for this data. Based on that, our semi product press part production date was week 46 with the complete hinge production date on 18 November the shipment was just the next day of it. It's the 19 November. The suspected batch time period is 16-20 second November. This is a dirty batch time period. It is crucial to have this traceability data as sometimes we can only have weekly based dates in this example, it is week 46, which is typical for large volume parts, especially in stampings generally. Next is who was involved or created the failure. The purpose here is to collect more data from the right contact for analysis and action, not to identify the person responsible for the failure as highlighted before, not for the blame. And next is why is this a problem? The problem arises because the hinge dimensions are out of tolerance, meaning they don't comply with the specified requirements. This explanation also clearly identifies why it is a problem. Next is how did it occur to understand how this happened? We needed to check production process condition. Were there any anomalies like additional operations such as rework, ask all those questions. These factors are important as this could be the potential causes of the issue also. Based on the control, we confirmed that in a normal process, there was no rework or additional operation. During the hole piercing operation at press, the second hole was pierced in the wrong position. There was no extra operation or rework. Issue just happened in the relevant press operation. Next is how much? How many parts were affected by this problem? So the answer is here, total, it is 18 20 parts are affected. This includes 360 sempducts in the line and 200 finished parts in the line, and 860 finished parts in the warehouse and 400 finished parts in transport on the way. We can also visualize the production quantities history on time basis to see the entire picture. This is very helpful, especially in chronic problems as we can easily see the trend and any peak point if there is. Another important point is sometimes we may face missing data. It is crucial to gather the missing data, but we shouldn't delay the solution process while waiting for it. The best approach here is to move forward with the available data and in the meanwhile, collect the missing pieces as soon as possible. But of course, if the missing data is the problem itself, it is the important part of the problem, we cannot take more steps without that. We should immediately clarify it in such cases. This concludes the D one problem definition step of the AT process. We defined here all 5w2h questions for both problem raiser and which is the customer in our case and for the supplier, which is the hinge manufacturer in our case. Now we can move on to the quiz to reinforce our learning. 10. D2 Similar Parts & Processes Analysis: Overview: In this step of the AT process, we will assess the risks related to similar parts and processes. So far, in the problem definition phase, we are focused on the defective part itself. However, as process owners, we must now evaluate our entire processes to prevent the production of defective parts. Now let's review the key questions we need to ask in this step. Here, our aim is to detect other effective parts or beds or processes. Before starting those questions, let's give a simple example to better understanding. Imagine you have two chairs and you use the same screws and process to build both them. After a few months, one chair becomes vobly and breaks. The first questions you will ask in that position, will the other chair also become vobly and break? This makes sense because both chairs were made the same way using the same scripts. If there's a problem with one chair, it could mean there is a similar risk for the other chair too. This is the logic of this setup. Now let's look at those guidance questions. First is other models and versions. For example, our part might come in different colors or variations like left and right or up and down, or different parts we can have from the same processes. Those parts can be risky either if the process itself has some variation. Or this could be a carryover part, for example, it can be a different project, but these parts may still share the same sempduct that carries the defect. We will assess all those possibilities in the other models and versions guidance question. The next question is to consider is whether there are any similar processes. We might have similar or identical processes used to produce different parts. If the defect is related to the process itself, such as process parameters or process methodology, then this similar process could also be at risk. We will evaluate these risks as well. Last one, we need to consider other customer plants. We might be sending the same parts or other potentially defective parts to different plants of the same customer maybe or directly to other customers. In these cases, we must ensure that other customer plants are protected from the same failure. Now we reveal those guiding questions and to understand better, let's go to our case study and make this step on that. 11. Applying the risk assessment for Similar Parts & Processes to the Case Study: When assessing the risk of other models and versions, we should ask some specific questions to address it better. The first one is, are there any other models of the parts such as right left versions or color variants or counterparts? In our case study, the defective part had an issue with the second hole location being misaligned. And when we check the other models, in this case study, we find the opposite site hinge, which is almost the same as the defective one. And these two hinges are used on the same bolt as right and left. We need to absolutely ensure that if these parts have the same defect. We are checking the whole locations for the second part, and in this case study, we are confirming this as okay. There is no problem on the opposite side hinge. It is fully functional, so we tick this is as okay then. Let's see the next question. Are there any different parts being produced at the same process? Here we focus on the process itself. We suspect that if there's a problem with our process, it might affect other parts produced by the same process. Let's bring our process here. This was the press where the defective parts were produced. This particular press is used for two different project parts with completely different geometries. Since the dies, which are the key for the geometry are different, it is very unlikely that we see the same issue on those parts, especially since their whole structures are also different. However, since the process are the same, just to enjure, we can check their dimensional conformity to ensure. In this case study, as expected, they're okay, no problem on those parts too. The final question, it's a bit different. Is the part produced from the suspected process used in any other finished product? In the other words, are we using the same part in different finished products, such as carryover or common parts. Let's remember, this was a defective semipduct which has whole location problem. We produce it in the press operation, then assembled it in the next operation. But what if the specific semipduct is used in other finished products, possibly for different projects or even different customers? In this case study, when we check, we are finding that the same semi products is also used in another part. That means we have potentially one more defective part which belongs to a different project. This is identified as risky because we used the same semipduct which was identified as defective. This is crucial to investigate further. Let's cross it. So for the first guidance questions of D two, we checked other models and processes and flagged one more potentially defective part. Now, let's continue with the other questions in D two and see what else we need to assess. In this step, we need to check if we have any other similar processes that might follow the same production way. In our case study, our failure process was the whole piercing operation in press. Let's say our press is A, and the part was this seen on the picture. Now we need to evolate other similar processes we have. One of the similar processes is press B, which produce a different part, another similar process is press C, which also produce a different part. So at this point, the part iometers are completely different, the dies are also different as well. Therefore, we don't suspect any issues with the whole locations for those parts. We can absolutely confirm that there is no risk for those other processes. Sometimes our failure mode can be different. For example, it could be related to the sheet metal like rust problem, let's say. In such cases, we need to evaluate the risk very carefully because if the material is defective, then all other processes using the same material could be at risk. What we are doing here is a risk analysis to ensure if any other parts or processes are affected by the same failure mode. The last question is about the other customer plans. Do we send these or other risky parts to other customer plans? This could be a different customer or it could be a same customer, different locations like different plants. In our example, our customer was Liver cars London plant, and the part we are discussing was this one. Now we have another part here that have the same sempduct as the issued part as we already identified during the other models inversions analysis in the previous page. So we need to find out where we send these parts to ensure other customers are protected. In our case, we ship these parts to Bosa car company, a different customer in a different country. So fundamentally, we now have two different customers to check who could potentially be affected by the whole misalignment failure modes. First is the liver cars London plant, which we ship the issued part and the second is Pusa car company. We ship them the other potentially risky parts. This was the last question in the two similar parts and processes analysis. Now, let's summarize what we've covered so far in this step. I D two, the similar parts and processes step, we answered all the questions. The first question was about other models and versions. We reviewed that process and found another part affected by the same failure mode as the part has the same defective semipduct. I captured the D 38 bonnet hinge as affected. The next question was about other similar processes. We had checked the other processes and confirmed there was no risk. So no parts were suspected from the other similar processes. The final question was about the other customer plans. We checked if we had sent the defective part to any other customer or to different plants of the same customer. And we confirmed that the defective part was only sent to one customer and one plant. So no problem here. However, we did find another potential risky part. It is Dtorte Bonnet hinge, which was shipped to Bursa car company. So that we are also defining this information here. Currently, we have to formalize summary of dt and we know that what parts and processes are affected, therefore, the customers are also affected by the failure mode. Now we have completed the D two similar parts and processes step. Let's do a quick practice with some quiz questions to reinforce our learnings. 12. D3 Initial Analysis: Overview: D three, initial analysis. In this step, we will begin examining our production processes to determine why defective parts were released to the customer. Where should we have detected tos, not okay parts. Alongside this, we will also evaluate the production process itself. Checking the basic conditions for any anomalies or signs that something went wrong. Our primary goal here is twofold. First is to identify where non detection occurred. We can address this gap initially and can take a proper containment action not to make same mistake again at the control activities. The second is about the production check. It is to assess the current status in production, if any anomaly. All those are to ensure we take effective actions in the next containment actions. This is crucial because if we don't understand where detection failed, the containment actions in the next step might not be effective because we can make the same mistake if we don't identify where and why we release the not okay parts. Let's dive into see what kinds questions we have in this initial analysis step. Our first essential point is to identify the non detection point. Where should the part have been detected? What is the control? Is frequency and the control location. In this point, our primary reference is the control plan, which includes all those data. This first question is about detection is quite important in this step because this will provide us an essential input to put effective control activities in the containment action section. And the second question is, why was it not detected? Why the controls were ineffective. There wasn't a control about the failure mode, or it was but not effective due to some reasons. The main purpose of these two questions about the non detection is to identify where the issue should have been detected and why was it not detected. Because without this information, we cannot determine what went wrong and why the defect was relates to the customer. So that we cannot get effective contamination, which we will need. One important point here is the question of why was it not detected, is not for the deep root cause analysis. It is just initial reason of the non detection to understand and address it. Let's give an example. P should have been detected during 100% visual control in operation 20. However, it wasn't detected due to operators lack of training. So this just gives us an initial overview of the non detection, which will be needed to put effective containment action in the next step. After learning this for this example, we can train the operator in the containment action or assign the trained operator and we can check the parts and detect if any not okay parts. In that way, we wouldn't solve the root cause of the problem, but we would have fixed the problem temporary. In other words, we will put out the fire as part of the containment action after this initial analysis. So far, those two questions were about the non detection point as part of the initial analysis. Now, we will focus the production process, basic conditions. The next question is the anomalies in the process. It is to determine if there are any anomalies or differences in the process. This could include production variations higher than usual scrap rates, for example, or machine breakdowns or anything else that deviates from normal operating conditions. Because these small details are significant as they could even point to the root cause of the problem. But in the step as mentioned again, our aim is not to root cause. It is to ensure our process working well because we will set our containment actions in the next step based on this data as mentioned. So the next is process in normal condition. In here, we check if the process was running under normal conditions or if there were any additional operations like rework or extra processes introduced. While these temporary change like rework can sometimes be necessary, they may lead to undesirable outcomes if they are not properly controlled. It is essential to confirm whether the process was operating normally not. The next is about the process conditions. Here, we evaluate various process factors, including machine settings, process control records, operator trainings, maintenance or breakdown recalls, process layout, and any other relevant process parameters. These aspects help us to understand if everything was functioning as intended. For example, maybe the machine was not set up correctly or the operator wasn't adequately trained or there was an unexpected change in the process layout. These checks will help to unfold all those points. The last one is operator or relevant supervisor feedbacks. Here, we gather feedback from the origin point. This can be operators, supervisors, or even the part designers depending on the problem type. These individuals are closest to the process and have the most accurate insights into what might have happened. Engaging with them is crucial for collecting reliable data for our problem. In summary, this step involves identifying anything that deviates from regular process activities or conditions. The first two questions were about the non detection and address the non detection point which is very important to put effective control activities in the next step. The other questions were about the basic conditions check of the process. Here we basically examine any differences, anomalies or unusual occurrences in the process. Now let's move on to the next and explore how we will conduct this initial analysis on our case study. Oh 13. Applying the Non-Detection Initial Analysis to the Case Study: In the initial analysis, our first question is, where should the defective part have been detected, which is identify the non detection point as initially. At this stage, we need to examine our process and identify the key steps related to the failure mode. Let's review the process of our case study. First point is the reception warehouse. This is where the raw material is stored. However, since the failure mode was about the whole position on the sempduct this cannot be a detection point. Next one is the first press operation. While in step we produce a part, the failure is unrelated to this operation because our defective part was the which have holes. Therefore, this also cannot be a detection point. Next is the second press operation and here is where the defective sempduct is produced. This should be a critical detection point as the issue originates here, we cross marks here. Next is the assembly operation. The defective semi product is used here in order to assemble with another part. This is another key point where the issue could have been detected. The last one is dispatch warehouse. Finished products are stored here before shipping. As we store our finished products here, which are defective, this place, again, is a potential detection point as well. Now we have marked the critical points in our process where the failure could have been detected. The cross marks indicate that proper checks at these points could have identified defect. However, we still don't know where we check the failure mode. It is in all these three operations that we cross marks or just in one of them, or maybe in none of them. Let's move on next and examine these points to understand better. So in this point, we are specifically reviewing the controls in place based on the official process documents. The primary document to reference is the control plan. Although we can also check additional documents such as work instructions or risk analysis of process. However, the control plan is the main source to confirm the existing controls. It is, what is the control and its frequency and where it is. From the key points we marked on the previous page, we now refer to the control plan to review the controls against the failure mode, which in this case is the whole position. So let's bring our defective part and process here. The first relevant operation was the press. At the press operation, there is a control fixture check with a frecuen we have five parts per h. We understand this by checking the control plan. So the next at the assembly operation, there is a 100 assembly fixture control. It means all parts whole are checked during this step. And the last in the warehouse, there is no control mechanism in place for this failure mode. The critical question now becomes, why were the defective parts not detected and shipped to the customer despite having two different control mechanism in the production process. Now we need to resolve these questions by addressing those gaps in order to put effective containment actions in the next step. Regarding the control in the press production, there's a control fixture in place. This control requires five parts to be checked every 1 hour during the production process. Basically, the parts are placed on the fixture and the operator confirms whether they are okay or not based on this control. This is the normal process of control. Basically, what we expect normally from the process. However, then what happened here and why wasn't the failure detected during this control? To answer this, we need to gather data from the production records and the relevant team to understand the situation. In our case study, the production records revealed that the control was not performed for the affected parts because control fixture was sent for calibration and the team continued production without this. At this point, we could ask a lot of questions to dig deeper. Why wasn't the fixture returned from the calibration on time? Why did the team proceed with production despite missing the control device? What decisions or approvals led to this situation? While it is valuable to collect as much data as possible during this step, Our purpose here is not to perform a full root cause analysis. This step is part of the initial analysis, focusing on identifying relase points to take effective actions in the next step, containment actions. So we will not deep dive into the underlying reasons at this stage because our priority is to produce good parts and protect customer in those steps. Now we have identified the first release point in our process. Let's move on to the next one. In the assembly control, we use an assembly fixture where two semi products are placed and assembled. During this operation, we also check the whole existence using pins to ensure hoists are present and aligned. While this may not be the most sensitive control, it is certainly capable of detecting if a hole is missing or misaligned, like in our failure mode. So then why wasn't the failure detected here? Because this control always shows that if the parts are assembled, then holes should exist and aligned because parts cannot be assembled without pins passing through the holes. Let's check the actual status during the production of the defective part based on the production records and feedback from the teams. It turns out we had the control fixture, but the middle pin was missing. As a result, when the operator placed the semi products on the assemble fixture, the part passed through the check without detecting the issue. As the middle pin, we checks the second hole is missing. At this point, we should also recall the other suspected part we identified in D two similar part and process analysis. You might remember this part as another risky part because it used the same semipduct as the defective part. In our case study, we use the same assembled fixture for both parts, adjusting the fixture tools accordingly so that the same fixture also checks the D 38 bonnet g. That means there is no different process that we need to investigate. Explains the issue. So now we understand why the defective parts were not detected and released to the custom. Despite having two different controls in the production process, we have controls in place, but those are just ineffective due to some reasons. Before we move on to the next section, I want to consider a different example to better understand. So just imagine what if we didn't have any control for whole locations in the control plan? How would we approach the situation in such case? In such cases, we would refer to the risk analysis where the controls are determined. It is specifically PFMEA, process failure mode and effect analysis. In PFMEA, we can easily see whether this failure mode was evaluated or not. We should also contact the origin point of the issue, whether it production method or quality teams to understand what went wrong. By reviewing the data, we can gain clarification on this issue. Now let's proceed to the next questions of D three initial analysis. 14. Applying the Basic Conditions Check to the Case Study: In the basic conditions check, our focus are a is the production process. We had some guidance questions to identify if there is any anomaly or difference at the current process. Let's bring our questions here and start to review our production process. This is the press operation where we produce the issued semi product. Normally, the process works like raw material comes in and we stump it and three holes are created. But the question is here, why is the middle hole misaligned t in the defective parts. To understand this, the best approach is to go to the actual workstation, run the process with the relevant team and observe. When we check and test the process, we see the issue, the stamping operation produced misaligned hole because the piercing punch in the die had been incorrectly placed at time over. This point, we could ask further questions to understand more details and this will be especially useful during root cause analysis. However, right now, our priority is not to dive deep into the reasons, but to focus on putting out the fire. This means taking immediate actions to produce good parts and supply our stocks for the customer or the next operations regarding the problem. Since this is the initial analysis, we have identified that the problem is clearly related to the die as the issue is not singular, all the defective parts have the same level of misalignment. So the die needs to be fixed immediately in that point. Otherwise, we cannot fit the stocks and ship good parts to customers, which means containment actions will fail. So that our address in this point is clear. In some cases, for example, if the issue was related to a singular part caused by a process variation or any other factors, it would be more challenging to identify the exact occurrence point. So instead, in those scenarios, we will systematically check all the process factors we mentioned in the guidance questions to ensure everything is working properly. But in our case, since the defective parts share the same issue, we can pinpoint the problem to the die and act quickly to recover production and restore stock levels. Before we proceed to next, let's review our findings with these guidance questions. The first is, is there any animal at the process? The answer is yes, we found one animal during the production check, which is the piercing punch location in die. It was wrong located. Next is is the process in normal conditions or any different operations such as rework. And the answer is, no, there is no different operation or rework. Process was in normal condition. Next is are the process conditions okay such as machine settings, process control records, operator trainings, et cetera? Our answer is here, process conditions are not okay. As we also found that the part controls couldn't be conducted by the operators due to problems in the control fixtures. But other than all machine settings, operator training, machine maintenance, breakdown status seems as okay. The last one, any other thing that was reported by operator. The answer is no, for our case example, operator didn't report any different thing that we found during production checks at initial analysis step. Now fundamentally we did a basic conditions check of the current process. Our reference documents were the process standards such as control plan, instructions, and even FMEA. We basically compare the current status with those standards to ensure all is okay or if any gap. While doing this, we also observe the process to ensure if any anomaly or difference. Now let's move to the next and make a summarize for the three initial analysis. 15. Summarizing the D3 Initial Analysis: Now let's summarize the initial analysis step. The first thing we tackled here was non detection point where the defective part should have been detected. It is in the stamping process. We have a control fixture to check the middle hole position at a frequency of five parts per hour. However, the fixture wasn't used because it had been sent to an external company for calibration. This was the first miss detection point that we identified during the initial analysis. The next detection point was in the assembly operation. There is an assembly fixture where all semi products are placed for assembly, and the middle hole is supposed to be checked here as well. However, the middle pin of this fixture was missing, so the defective parts pass through as undetected. Now for the current status analysis, which is the second phase of the D t initial analysis, we had checked the production process to identify if there is any anomaly or difference. We focused on understanding what went wrong in the process. When we check the current status here, we observed the process and we found that the male piercing punch in the die was incorrectly placed, which led to the misalignment of the middle hoy during stamping production. So at this stage, now we completed the initial analysis. We didn't conduct a root cause analysis here yet, but we have identified the main reasons for the non detection and also the occurrence during the production check. This allows us to implement the right controls to produce good parts, which is what we will focus on the four containment actions. Then, that was the end of the three initial analysis. Now let's move on to the quiz and reinforce our learnings. 16. D4 Containment Action: Overview: So now we have reached the step where we actually implement the immediate actions based on all the information that we have gathered so far. This is where we apply contained actions, which are essentially temporary solutions aimed at addressing the issue right away. It is like putting out the fire while we prepare for a deeper investigation and long term solutions in the following steps. By this point, we have done a rigorous analysis of the problem. We have defined it, assess the similar parts, and processes risks, and pinpoint the factors through initial analysis. The containment actions we take here will focus on ensuring that no further defective parts are produced or shipped to the customer. It's a critical step because in this point we extinguish the fire, contain the creases and makes time to carry out a detailed root cause analysis and develop containment actions later. What's the next step to put out the fire? Well, the answer is straightforward. We will take everything we have uncovered in the previous steps and turn those findings into actionable steps. These actions will serve as a temporary solution of the problem. Now let me be clear. This isn't about fully solving the problem just yet. Instead, it's about stopping the production and shipment of defective parts. By doing this, we will ease the pressure and buy ourselves time to be able to work on further step analysis. Containment action is a basic action plan, which have the immediate actions like this one. Here, we describe the actions, assign the owners, and set the dates. But the question is, what actions we will identify on this table? We have some basic types of actions we need to consider in this step. Let's review them. The first one is stopping the production. If our process keeps producing defective parts continuously, we should stop production until we have fixed the issue and ensure the production of good parts. However, if the issue is singular or occasional, we can't just stop everything until it's fixed. Instead, we need to focus on sorting activities to make sure only good parts are shipped for such issues. Next up are sorting and control activities, which are crucial. At this point, we set up control activities across all relevant parts of the supply chain and segregate defective parts for either scrap or rework. These activities aren't limited to current stock. They should also cover any parts that will be produced in the future. Why? Because until the permanent corrective actions are in place and validated by the end of the AT process, we can't be sorting whether the problem will recur or not. That's why it is essential to manage sorting activities across all potentially risky bets. Most of the time issues are isolated or occasional rather than affecting every single batch. In these cases, it is important to secure both current stock and any parts produced going forward. But if the process is producing continuously defective parts because of some reasons that cannot be fixed immediately, then checking every single part might not be necessary. In such cases, it might be more efficient to isolate entire whisky bats instead of individual parts and of course, stop the production to cut it. Another important thing here is about the documentation of these control activities. How we will check, what is the method. All those need an instruction to describe clearly, even for those additional control activities in the containment action. Let's give an example. We can apply 100% visual control with a slightly different way than normal, or we can adapt the quality wall to one more check for all parts. So we should document these kind of activities with some instructions to describe clearly to the control operators. Next is the training activities. Sometimes it can be necessary to train the operators regarding the new control activities that we put in use or the other details of the failure mode. Next is the rework activities and additional operations. These are operations that aren't part of the standard process, but are introduced temporarily to ensure parts meet quality requirements. Sometimes defects can be easily fixed with rework activities that can be done manually by operators or automatically by machines. For instance, if a product has excessive burs, grinding operations could help remove them. Or if a whole diameter is slightly out of tolerance in a part, we may introduce an additional measuring step temporarily to bring the part back into specification. However, this should only be considered after conducting sufficient trials and validations and after obtaining customer approvals. Because these actions are usually only considered when defects are too large to isolate and rework is the only viable solution. But it is a riskier approach because it could create new problems while resolving the existing one. In the automotive industry, we generally avoid rework unless it is required. Another type of action is temporary process adjustment. In some cases, we may need to adjust process parameters temporarily. While this adjustment may deviate from the standard process requirements, they might be necessary to maintain part quality and keep production going. For example, if a welding operation isn't assuring the required strength due to material variations, we might increase the welding temperature beyond the standard parameter temporarily as a containment action. Again, we need to ensure that these adjustments are validated and we have approval for the deviations, as we will be working out of the standard process. The next type of containment action is replacement. This is usually applied to recover defective or suspicious stocks from customer. Let's give an example. If there is doubt about customer stocks and it is difficult to check them all, we can simply swap the suspected parts with the good ones. It is a straightforward approach, just a replacement. Then the next one, we have substitution, which is a bit tricker and should only be used when really required. Substitution involves using an alternative part or material that meets the functional requirements but differs from the original specification. For instance, if a specific sealant isn't available due to supply chain issues, we might use a compatible sealant with similar properties as a temporary solution. However, this action requires significant effort because it typically involves a full validation and approval process. The next is scrapping, scrapping the not okay parts. After sorting and control activities or replacement actions, any defective parts that have been segregated and isolated must be scrapped if they are not going to be revoked. The scrapping is another contained action that should be considered in the contained action section. And the final type of the contempt action is controlled shipments. Up until now, we have focused on segregating defective parts and ensuring that only good parts are produced and shipped to customer. But at this point, we manage the shipment of the good parts and specifically marking them, labeling them with appropriate definitions. And informing the customer as well and the other relevant stakeholders. It is important to clearly identify the status of the parts, whether they are okay or still defective, as this is a traceability necessity for the new parts. In that point, we can mark the products directly after the agreement with the customer. Also we can directly define the batch with some definition label like controlled parts or 100% checked that. One other important point in the content induction is about the timing. The general acceptance timing, especially in the automotive industry is 24 hours. That means we need to define and submit the containment actions within 24 hours, including all previous steps complon depending on both internal and customer standards as well. But general acceptance is 24 hours due to its urgency because each single hour we lose money, sometimes even repetition of the company. It is very critical to be in line with the timing. Now that we went over the overview of containment actions and understand the steps we should take. Now let's dial into our case study and explore what we will apply accordingly. 17. Developing the Defective Parts Statement for the Case Study: Before we dive into preparing the containment action plan, we first need a simple table to see the current status of the defective parts. This table is important as it helps us to manage sorting activities and recover stock efficiently. Let's break it down. In the first column, we'll list the relevant areas of the supply chain. It starts with the supplier then moves to the receiving warehouse, the production area, this page warehouse, and followed by parts in transit, external warehouse, if applicable, and customer stocks, and finally, the end users if relevant for the problem. Across the top row, we have our defective parts. The first part is the semi product, which have three holes. Next, we have the A 49 bonnet hinge, which is the final product that we received quality claim for. After that, we have the Dtortate bonnet hinge, another finished product that we identified as defective doing similar parts and processes analysis. And finally, there is the vehicle itself because in some cases, the defect might not be detected in the next operations with the customer and the cars might be sold with these defective parts. The vehicle can be an important element in this table regarding the problem. While the main control areas are mostly the same, this can change depending on the issue at hand and its specific supply chain structure. Similarly, the parts listed in the first row can also change depending on the problem organization or production steps. Now let's go over the defective parts in this matrix. We will start with the supplier. Since the defect occurs in the press operation, both the supplier and the receiving warehouse are clear. There is no defector. Next is the production area. Here we have defective semi products and a 49 bonnet hinge. Moving to the dispatch warehouse, we have got both the defective a 49 and D 38 bond hints which are waiting for shipment. Then there are parts in transit. Right now, we only have defective a 49 bontins on the way to the customer. And after that, we have the external warehouse. Sometimes there is one or more external warehouses after the main plants warehouses where parts are shipped and stored before final delivered to the customer. This depends on the supplier and customer organization and the agreement as well. But in our case example, there is no external warehouse. Our parts go directly from the production plant to the customer. Next is the customer stock. The customer has a certain number of defective parts as they have notified us of the issue. In this case example, over 300 defective a 49 bonnet hints are at the customer site. These parts were going to be used if there hadn't been a quality issue, but now they have been isolated and are waiting to be shipped back. And finally, we'll look at the end users or varianty areas. Since the defective parts haven't been assembled into the vehicles yet, there are no issues at customer site or with end users. We have a significant amount of defective parts, both semi products and finished products. This raises the question, can we rework these parts to not to vest them? In our case, it is not possible because we can't change the whole locations on stamping parts. However, if the issue or something like the birds that could be reworked with additional operations, we would consider rework activities with the necessary controls and customer approvals. But that's not the case here for this example. The plan is to scrap defective parts and replace the ones in customer stock. So the numbers are 360 defective semi products, 17 80 defective a 49 finished products and 11 20 defective D 38 finished products. How we have received those numbers? If you remember, we had done five WH analysis where we had the data for defective part date period. We know the last part production date. Referencing this K date, we determined the dirty batch and easily took these numbers from the system. In our containment action, the goal is to eliminate the defects from all areas as quickly as possible. Until the containment action is in place, we are still in a crisis, so we need to reduce the tension by putting out the fire. Now let's move on and look at the containment actions we will put into place to address those. 18. Containment Action Planning for the Case Study: First of all, this action plan format in the content action can vary. However, it should basically include the action, the responsible person, due dates, and status if possible. At the previous phases, we addressed the problem and determined that all parts produced after a certain date were defective. The first priority should be fixing this issue to be able to produce and ship good parts. So our first action is to correct the die punching as its location on the die was wrong. As I already mentioned in the earlier stage, it is not always possible to identify the occurrence during the initial analysis, especially in singular cases. This is because the initial analysis is not a deep investigation, but rather a basic conditions check. In such cases, we will start with the sorting activity to segregate the good and bad parts. However, in our case example, we need to immediately begin producing good parts as the only way to recover the stock and meet the customer's needs was to address and fix the occurrence point because our process was producing continuously not okay parts due to wrong placement of the die punch. Next action is we need to put effective controls in place to ensure our shipments to the customers are okay, defect free. To achieve this, we relied on the findings of the initial analysis and address the non detection point identified in step D three. Now to implement an effective control, we fix the control fixture which is used for frequency control and the press production and the assembly fixture which is used for 100% for aligning the whole locations. With these corrections now, we are ready to produce good parts and we are ready to verify their quality to ensure they meet the required standards. After that, we train the operators on the failure mode and the status checks of the control equipment to be sure these control equipments working well. Don't have any problem like before. This training is necessary to ensure they could also recognize similar failures if occur again during the containment action. Our next action is to produce new parts to recover customer stock. After that, once these parts are produced, we verify them using the fixed control fixtures in the process that we just put in use to ensure they are okay. Now we produced good parts, check them, and so that we have good parts. Currently, we can send them to the customer. However, before doing so, it is critical to ensure traceability in case feature identification is needed. This might involve marking the parts themselves, defining the bats or both. In our case example, we do both. We mark the parts and define the shipment bats with labels. If the problem had been internal and not customer related, it will still have been important to identify the parts with special marking or definitions to ensure they performed correctly. Because without traceability, managing further identification would have been very difficult. So afterward, the next action is we inform our customer about the first control batch information, the first shipment date, quantities and other relevant data regarding the first shipment because this allows them to identify the parts and plan their production accordingly. This information submission is important. The final step is scrapping the defective parts since rework is not possible due to the nature of this failure mode. So now the most significant point regarding the containment action is maintaining it until the corrective and preventive actions are validated as mentioned before. This is critical because even if we take the necessary emtd actions to put out the fire, the problem could arise again if the root cause has not been resolved. In the specific case example, we will continue sending the parts with defined label until the final validations are completed as already defined a action here. This will show that we are still checking our parts 100% with the control fixture, even though it is normally done as frequency five parts per hour. In some cases, extra controls can be introduced to the product in contained actions. This might include temporary new controls or temporarily increasing the frequency of existing controls. For example, adapting a quality well for extra 100% check for all parts or adding 100% visual inspection to the current process, or increasing the frequency of the gauge control, let's say. So in some of the case, we should consider those extra control actions as well as the sorting activities. Because if our controls are not effective, we can fail again if we apply the same controls without any changings. But however, in our case example, we didn't consider some quality or extra control actions or control fair cons increase because the current actions were already effective with properly functioned equipment. The problem was those controls were not being applied to some failures. So we sorted out those failures, made those controls effective again as part of the contamductions. In this case example, we didn't make a sorting activity to segregate the defective parts to clean the bets. The reason is because the failure was a chain, it means all parts were defective. The sorting activity was not required for the current stocks. However, if we do some sorting activities when it is necessary, depending on the problem, it is very important to follow the results as daily, to be able to see whole picture of the problem. In such cases, we can follow the results with some simple table as daily, which is easy and useful. So this concludes the containment action, which also marks the end of the lectures in this course. In this step, we identify our immediate actions, resolve the quizzes by controlling our process and ensure that only good parts are shipped to our customers. Now let's make a practice with a quick quiz. 19. Closure & Next Steps: I extend my sensors congratulations on successfully completing the mastering AD problem solving Chris's the containment, D one, D focus. I deeply appreciate you joining on this learning journey. My hope is that the invaluable insights and practical techniques we have explored here will empower you to confidently and effectively navigate challenge that arise in your professional life. This foundation course has equipped you with the essential toolkit and strategic framework for robust problem solving. We have delve into critical areas such as precise problem definition, insightful analysis of similar paths and processes, conducting thorough initial analysis and implementing crucial containment actions and essentially containing decreases in containment action. In the coming months, I'm delighted to announce the upcoming release of the next course, D five and D eight, including the root cause analysis, corrective and preventive actions, action validations, and the closure. This next step builds on the foundation you have established, focusing on advanced problem solving strategies to help you fully master the 18 methodology. Thank you once again for your time, dedication and commitment. I look forward to potentially encountering you in future course, and I really eagerly anticipate the continuation of our collective learning journey. Thank you, Again.