Name: From data to competitive advantage: Renesas transforms semiconductor manufacturing with Spotfire®
Uploaded: 2025-10-23T08:16:07.450Z
Duration: 53 min 15 s
Description: From data to competitive advantage: Renesas transforms semiconductor manufacturing with Spotfire®

Transcript for "From data to competitive advantage: Renesas transforms semiconductor manufacturing with Spotfire®": Good morning, good afternoon, and good evening to everyone wherever you may be in the world today. Welcome to today's webinar, from data to competitive advantage, Renesas transforms inductive manufacturing with Spotfire. We're thrilled to have you with us today. I'm JP Richard Charman, and I'll be your host for today's session. Now before we get started, I wanted to cover a few housekeeping items to ensure you have the best experience during today's webinar. Now the webinar will last for up to around forty five minutes with a ten to fifteen minute, question and answer segment that will be held at the end of the session. If you have any questions during the presentation, please do not hesitate to use the q and a panel, which is located on the right sidease of your screen, and we'll address as many questions as we can during the q and a segment. We've also made a few assets available linked to today's webinar in the docs section of our webinar platform, so please feel free to access these. And the docs section is located right next to the q and a section on the right hand sidease of the webinar platform. After today's session, a recording of today's webinar will be made available on demand, and we'll email you a link to the on demand version shortly after the event. Now with that, let's dive straight in. I'm excited to introduce our presenters today. We have Moumita Bardhan from Renesas Electronics, Alessandro Chimera, our industry solution lead here at Spotfire, and Jingchuan Lin, our principal data scientist here at Spotfire. Now without further ado, I'll hand over to Namita from Renesas to talk through the Renesas story and how they've been using Spotfire and how Spotfire has helped them in their operations. Over to you, Moumita. Thank you, JP. Hello, everyone. My name is Moumita Bardhan. I serve as the director in manufacturing applications area at Renesas Electronics. Thank you very much to the support team for giving me the opportunity to talk about Renesas' journey with Spotfire. So renaissance shares a long standing association with Spotfire, having continually embraced its innovative enhancements across multiple versions to drive efficiency and productivity in data analysis and in yield improvement. So our journey with Spotfire dates back to its early days. Even before it was officially named Spotfire, when it was still known as a decision site, Over time, various Renesas factories in Japan were onboarded to the platform, beginning with Naka in 02/2005, followed by Takasaki in 02/2010, and subsequently Yamaguchi, Kawashiri, and Nishiki in 02/1113, and '14 respectively. Around 02/2015, Renesas began liberating Spotfire for internal engineering purposes, marking a new phase in our data driven operations. While usage in Japan expanded through our manufacturing sites, non Japan regions have primarily adopted software, within engineering groups. Over the years, our utilization of Spotfire has evolved in line with each version of great, reflecting when it starts to commit to continuous improvement, innovation, and effective use of advanced analytics to support business excellence. To the next slidease, please. So in the semiconductor manufacturing industry, as most of you already know, yield is a critical determinant of profitability and time to market. However, yield improvement efforts are continually challenged by increasing design complexity and process variability. At Renesas, we recognize that our existing tools, manual, static reports, and spreadsheets were no longer sufficient to manage the scale and complexity of our data. These traditional methods lead to delays in root cause ideasentification as semiconductor analysis relies on massive dataset, often encompassing billions of data points. Establishing, a scalable, efficient, and accurate analytical process under such conditions proved extremely challenging. Furthermore, in a large globally distributed organization like Genesys, cross team collaboration plays a vital role in problem solving and decision making. However, the lack of unified analytical platform makes seamless collaboration difficult, slowing down communication and alignment across teams. Collectively, these challenges resulted in delayed decision making, business inefficiencies, underscoring the need for a more powerful, collaborative, and scalable data analytics solution. Now next, I would like to talk about how Spotfire helped us to overcome these challenges. So with the adoption of Spotfire, we integrated heterogeneous data sources from fab, test, and assembly operations in real time, creating a unified and dynamic analytics environment. User gained the ability to design and, interact with customized dashboards, enabling data filtering and visualization at various levels of granularity by load, wafer, run, to suit their specific analytical needs. Spotfire also facilitate, advanced analytical techniques such as, you know, pattern recognition, correlation analysis, clustering, allowing engineers and analysts to uncover deeper insight with greater efficiency. These capabilities were delivered within a secure and shareable and collaborative environment, promoting cross functional engagement and transparency across teams. As a result, yield analysis accelerated significantly, leading to measurable improvements in material utilization, reduced retail cost, and enhanced process stability and product quality, of course. Decision making became faster, more consistent, and data driven, enabling teams to act proactively rather than reactive. Importantly, the adoption of Spotfire also laideas the groundwork for predictive data analytics, setting the state years in advance for the full scale predictive capabilities now integral to NSR's data strategy. Moving on to the next slidease, please. So here, I I want to talk about some real life use cases. It ranges from, a very simple ones to some really, deeper analysis of by our engineers. So if we move on to the next slidease, please. So here what we see, it's called, special zone handling, which refers to the ideasentification and targeted analysis of, specific waveform ranges such as age zones, center zones, exclusive, zones, or custom defined areas that may exhibit, distinct process or performance characteristics. By isolating and analyzing these zones, engineers can detect localized process variations to induce defects or systematic issues, for example, that impact heat and quality. So together, wafer map modification and special zone handling enable more precise defect globalization, improved root cause analysis, better process control, ultimately contributing to, enhanced cleaning, performance and, manufacturing efficiency. Okay. Now to the next, example. What we can see here, it's a it's a very common, terminology in the semiconductor industry, just like the statistical wind limit and the, statistical yield limit. Let me start by giving you a very brief definition of it. So statistical beam limit is the threshold used, during testing to classify semiconductor devices into performance beams based on statistical analysis of this data. It helps ideasentify devices that fail to meet desired, performance criteria, enabling detection of systematic issues and ill loss associated with specific needs. SYL, statistical yield limit, whereas represents the expected or theoretical maximum yield calculated using statistical models based on process variation and device performance distribution. AseWire provideases a benchmark to evaluate actual e performance and ideasentify areas, for processing. Now how we use it, we we have, you know, templates to see that whether, you know, some of the material is actually violating these limits or not. So by use use of this, we can we can engage, in a lot, improve lead improvement processes. And then, moving on to the next slidease, please. Yeah. So this one is, is a limit management. No. This is a process of setting, monitoring, and adjusting specification limits, USL or LSL, the upper limit and the lower limit, and control limits used in manufacturing or quality processes. So why do we do that? What is the goal? So we ensure by doing this, we ensure that product specifications are achievable, aligned with customer requirements, and reflective of actual process capability. What I mean, what are the impacts in all these limits? So tighter spec limits reduce yield, which means more rejects, while loser limits increase yield but may risk product performance or reliability. So this template gives our engineer very good insight that to take a decision, so they can take good decision based on, you know, an analysis by these templates. Now moving on to the next slidease, please. So this one, what we see is a a wafer map reconstruction for customer return material, which involves recreating the original spatial layout and test data of silicon laptop devices returned by customer. So this process is, essentially, when devices are sent back due to failures, warranty claims, or any quality investigation, the engineers are basically then reconstructing the maps and really pointing out the problematic zones. As you see here, the wafer map is, is laideas out, but only pointing out the problematic zones. So yeah. So these are the few examples I I picked up, to show you how we are using, Spotfire, in in real life analysis. Now I will conclude with the next slides. So to the next slidease, please. Yeah. So what is there in future? You know, how we are, looking at it? So at the moment, we are really on the verge of, consolideasating the e tools globally. In a big company like Renesas, we have several, tools in different areas, but we are really looking into consolideasating this space. We want to break silos, deliver rapideas insights, and we also want to get, the stakeholder trust by provideasing them a trustworthy scalable, highly available solution. And, move we really want to move from exploratory dashboards to repeatable analytics, automated reporting, which will save us a lot of time and money, of course. We also want to train engineers and the operations to use and and use, and share Spotfire assets. It's more like promoting the self-service, culture. This is one of the, features, you know, I personally like about Spotfire the most, that it really gives the users a lot of space to create their own dashboard, their own templates, and they can really do their analysis, in their in their on their pretty much on their own. Also, of course, we want to engage in operationalize ML models, for lead prediction, early anomaly detection, explain explainability. We we definitely want to engage more with Spotfire, you know, new, yeah, new things like, you know, the the e I e I, part of, Spotfire, the new, ML, functions of Spotfire. Yeah. So so with that, I will conclude, my, presentation, my our journey with Spotfire, and I will hand over to Alexander. Thank you so much. So thank you very much, Moumita. So certainly, manufacturing, semiconductor manufacturing is is a challenge in the industry that requires engineers to be empowered with the best tools. So let me move on the next slides. So I just want to take a few minutes to reflect on some of the pressures that high-tech manufacturing is currently facing. First, the high, capital equipment costs. Today, wafer fab can cost $10,000,000,000 and even more. And with that kind of capital investment, it's clear you make sure that the facility and all of the equipment insidease is producing the highest possible quantity at maximum truth put so that the investment is paideas off before the technology is no longer competitive and needs to be updated. And then second, short and complex production life cycles. Electronic product life cycles can be as short as six months. So we have to characterize and optimize the behavior of complex processes and products as quickly as possible so we don't miss the market window. This means that we need to accelerate the learning rate of new products and optimize the production performance every single day. And then third, aggressive competition and demanding customers. Of course, the competition isn't standing still, so we have to create a unique and and differentiated product that can fund the investments into the tomorrow's next generation of products. But what is common across all three challenges is the amount of feature agendas technical data being produced. And the only way we can do all of this is to be better than anyone else in pulling insights from that data so that our people, our teams can make faster and better informed decisions. So we've seen the business, dynamics and the challenges. But in many industries, not just in manufacturing, we are now entering a new phase driven by the fast rise of AI, generative AI, and now also with the power of AI agents. And across the industry, there's a real urgency to unlock value from data. Companies are now looking into adopting AI into their day to day activities, but where the human factor is still relevant and is part of the decision process. And then the second picture we are seeing is the rise of tech safety engineers entering the field. So engineers who are fluent in Python or are native to open source and used to experiment fast. And they expect a very different experience from the tools that they're using. They demand more flexibility in doing their work. And then the third trend is about taking advantage of all sorts of data to get value from the unification of company widease data. No whether no matter whether these kind of data are. And then another trend, we see is that many projects involve larger, more, diverse team than ever before. So yield and defect engineers, equipment engineers, product and data scientists, they all need to work together seamlessly. And collaboration and sharing insights have become a real success factor because, no single expert has the whole picture. But together, they can move decisions forward much quicker. So the question really becomes, are your engineers equipped with the right tools? Do they have all the need to analyze company widease data widease data to go fast, ideasentify root causes quickly, and take the right decisions? So think about that. Well, we know that every day brings a new challenge with many different use cases. In short, it takes a team to act on them. And the output of one team is the input for another team. And for that reason, it is important, to have an industry ready tool that is flexible enough and able to support engineers in their day to day job. So but what does the market currently offer? So many manufacturers will always have, these two categories of software. On the left, we can see the specialist tools and then on the right, the statistics tools. And regarding the first category, the statistical tools, most engineers are not statistics experts. And today, well, we call them data scientists. And they are highly specialized in snow in knowing which algorithm to use and how to tweak them for a specific purpose. But when it comes to specialist tools, the second category of software, you have to be a specialist to operate them. Unfortunately, most manufacturing rare problems aren't standard. It's not easy to extend specialist tools to incorporate other data that you need. They're just not flexible enough and, it becomes also, quickly expensive. And here, Spotfire is different. It is designed to be the indispensable data analysis technology for engineers. By combining agile exploratory visualization with advanced manufacturing specific algorithms, Spotfire puts the data at your fingertips and helps you to bring together data from anywhere and display it however you like. And this means that engineers can wrangle data, find and clean clean outliers, and create new variables or metrics on the fly. And then Spotfire natively supports Python, R, and even our own, obviously, Spotfire Statistica to create usable data functions so that they become new tools in your menu. And we call this, unique and powerful combination of capabilities, this for data science. So and with that, I hand it over to Jingchuan Lin to deep dive into the capabilities. Great. Thanks, Moumita. Thanks, Alessandro. Wonderful presentation. Let me start to share my screen. So in my section, I will try to show some examples of Spotfire data science and let you know how you can really do the usual data science with Spotfire too. Before I jump to the examples and demos, I think it's quite good to emphasize our vision again to the audience today. First of all, we want to build mutual data science to support all advanced use cases that involve the machine learning or even deep learning use cases. And we focus in specific industry because we realize those industry like high cap manufacturing, organic gas, like SARS. We're really involved with a lot of deep research and scientific work that Spotfire can really bring the benefits. And we do support enterprise scale. That's how, our customers is using and deploying Spotfire across the factory and across the regions to collaborate and drive more values. When we offer Spotfire data science, we focus with different prospects, including, like, how to do the modeling, how to support some coders, and how to resolve the, specific industrial questions like doing some wave map analytics. So that's why just now, Moumita presented a lot of use cases to see Spotfire users are creating the label maps and do the analytics, including the yield, failure, and even the beam, regular map, pattern, classification, own spot by tool. Please providease my specific building data functions for now, so including, to handle the topics about time series data and geospatial data and also some basic functions to handle the missing data from your dataset. And these examples, we already embed in the Spotfire tool, and you can start to use it without extra steps to install. And those data functions are developed by our in house data scientists and have been tested a lot with not just small dataset, but also the large volume dataset to make sure it works very well for different type of volume of the data. So for examples, when users want to do some time series data analytics, usually, they have to do time series data smoothing and time series data resendments. These two steps are very important to prepare the time series data for any use case for any, further analytics as well. So this is the screenshot to show you how Spotfire users can easily do time stamp smoothing and resampling. They just need to import the data and define which column represent the time stamps and then what are the values we need to analyze. And then we can start to see, like, what is the time series re centering rules and the number of the samples, and do you want to apply some aggregation method to, to fill in the data. And in the end, you can see some charts like this, that can, specifically showcase you the pattern, the outliers. Then you can quickly understand the values from your time series data. And, again, this is the screenshot to showcase, how you can handle the missing data from any dataset. After you run this building missing data summary, we will get some tables and visualizations to clearly show you the number of the missing values from each columns and what is the suggest way to impure the missing values. So that will be very smart and very fast, even for the large dataset. So we put everything bundled together, and we call them add ons for Spotfire data science package. For the users, when they open Spotfire, they will find those visualizations, actions, in their tool already. That's how we call it. We already integrate everything with Spotfire product. And, also, we providease some demos that can really guidease users to quickly learn how to start using those advanced visualizations and advanced actions. And now I want to, okay. Here is another example about the volume plot. That is where key examples about additional data science visualizations because we have been working for this volume plot, over the past few years. Why it took so long? Because we keep adding new features to this volume plot. So volume plot is like an enhanced version of the box plot. Compared to the traditional box plot, you only see the box, but the volume plot will show you the distribution of the data points of each group. Then you can get more insights when you compare the data between each groups. Okay. Now I want to show you some real demos, and I decidease to, show the demos in line because in the recent release, we add those demos to guide our users how to quickly learn how to use those advanced use cases. So, for example, if I want to create a wafer map, I just need to click here, and it will load you the example data set we prepared for this demo. For those audience who are not very familiar with the wave map use case, let me quickly explain the data columns here. So we have the dye x, dye y. Those are the values represent the spatial coordinates of each dye on the wafer. And also we have the log column. That is a batch of wafers processed together through the same production step. Also we have the waiver column that ideasentifies a specific waiver number within the log and usually they can number from one to 25, it depends on how many waivers in one slot And this is the bin column. It usually represents the test result category for each type. For example, bin one could represent the pass is a good byte, but bin two or three, four other values could represent the different reasons for the failures. And values could be any other values, like when you measure from the, beam light. It could be the temperature or pressure or some test values. So comes back to the front end. In this guidease, we just follow the instructions here. You can find these actions called create wafer map. And I just click this long button and make sure I'm selecting this example over data. Just need to match the exact columns this action requires. And make sure since I'm selecting the bin column, remember, just now I explained this is the category called data column. Yeah. Click one button, and I will get this regular map immediately on my dashboard. And this is not just some image because it's it's highly interactive that that you can still drill down, zoom in, zoom out, select multiple waivers to do the further analytics. But I would say this is the very basic step to compare your data to convert the data to the vapor mass. And in this tab, it shows you the example output. Alright? Let me show you the other example. I need to close this one first. Yeah. We have another demo. It's called waiver zone analysis. In Moumita's presentation just now, she also pictured, how Renesas is doing dual analytics. So it is a very useful, analytics tool and method to further drill down and analyze the performance by different zooms. So we have a similar example data with x y lot vapor beam, and now I want to create some zooms on the vapor map. Let me follow this instruction. I have to create the zoom values first. And, similarly, I need to define the x y and the number of the radio zone and the angular zone. And I click run. So after I run this first action, I will get two columns that splits my records by the zoom, and I have very specific number to represent the zoom, the the different zoom values. Okay. So the next step, I want to create some visualizations. We prepared another action here, analyze zoom profile. Okay. I have to make sure this is the right dataset. This is the right x y column and the bin value. Okay. But here, we have to select some columns because we need a log ideas. Same. Like, we have to let the action know what columns represent my Zoom numbers. Okay. And I click the wrong button. Yeah. So after that, I will get another tab that showcase you the the values, five different zoom. So those numbers will represent the zoom we created just now and just describe you one example, one quick insight we can learn from here. So this green color, like, being four, it could represent some good values, and we see the high values at the beginning. And we have another group of red color line tracked here. Then we see some hikes in around this area. So for the engineers, they can go further to analyze where those zones are located on my wafer map, and they can try to understand the pattern. And if, let's say, this green color represents some good, but the red color represent some bad result, then the engineers can quickly understand what is the reason it could come from each specific zone of the data map. So that's how we can support the zone analytics from using Spotfire. Okay. So back to the slides. Just now I demoed how we can use those, building actions to create a wafer map and to create a wafer zoom. And indeed, we have more examples similarly as the actions I demoed just now. So for example, if you want to, apply several steps to analyze your time series data, we have this time series preprocessing action for you. When you import the data, it will do not just resampling or smoothing. It combines several steps including normalizing the time series data, input missing values, resampling, smoothing all everything together, then you can get some good output with the deep analytics over the time series data. And, again, like the missing data summarization action, that is the action. When you just click the button, you can get some dashboard looks like screenshot here, can give you a very good summary in KPI charts, in the bar charts, to let you understand how you want to handle those missing values. And then you can further decidease if you need to refresh some columns or you really need to, find some better dataset for the specific, analytics or further analytics. Okay. That's their showcase of the Spotfire capability. I will let Alex to introduce more customer success stories for now. Okay. Thank you very much, Jingchuan. So besideases our customer Renesas, we have also some more awesome customer stories and manufacturing to share. So a great example is is Brembo. Brembo is a global leader in high performance braking systems, and I guess you all, have seen those red brakes on sports cars. So Brembo faced a classical challenge. Quality issues and process deviation across their global plants were not that great. Their engineers were familiar with the pro processes, but they were lacking the tools to analyze the data independently. Mostly they were depending on a central IT team and had access to a small, data science team. And guess what? The process was too slow. So by deploying Spotfire, they put the power of analytics directly in the hands of their process engineers, and they connected Spotfire to their, MES data and credit assistance and enabled engineers to investigate everything from production defects to machine parameters, inconsistencies. And the result, they accelerated their root cause analysis from days down to hours. And this led to a significant reduction in scrap, improved product quality, and empowered the best people to solve problems much faster than in the past. Then we have two more customer stories, Hemlock and, Kaff. Hemlock Semiconductor is the largest polysilicon producer in US and produces the material used in high-tech electronics and also solar panels. And they've used, Spotfire to analyze data from every step of the manufacturing process to better understand the impact of temperature, pressure, and energy usage in their reactor process. Hemlock also use, Spotfire to optimize their energy consumption, and they were able to obtain savings of $300,000 per month. Then last but not least, CAF, a Spanish company that manufactures and maintains and operate trains. CAF introduced Spotify to optimize the maintenance of its moving assets. And by analyzing various complex data, the solution led to a 10 to 40 times a reduction in breakdown rates and a 10% reduction in maintenance costs. So again, let me hand over to Jingchuan, to share our latest product announcements, Spotify 14.6, with specific industry cover release. So, Jingchuan, again, over to you. Great. Thanks, Alessandro. We got a few more couple of slideases to shout out to the ladies in Spotify because just in this week, we released, Spotify 14.6. So that is our recent long term support version, and we do have a lot of new features over 100. And, actually, we are very proud to say we implemented those features based on the ideaseas voiced from the customers. So looking ahead, we want to drive our product to get more innovations from, data science functions, from the actions, and from the industry specific use cases. So that will definitely drive our product direction. And, of course, AI, gen AI durability, we will integrate those new trending topics, technologies insidease for the insidease Spotfire product. And we do have the agenda for the, what's new in Spotfire in in future, and we do have some, several follow-up webinars to showcase their, what is included in the Spotfire inferences. Alright. Thank you very much for that. And thank you to Alessandro and Moumita for those insightful presentations. We just wanted to present some key takeaways for you, Kennel. And a lot can be taken away with today's presentation, and we've narrowed that down to three main takeaways. Number one, really being turning complexity into clarity. And as you saw, Renaissance unified data from fab to test with Spotfire accelerating yield analysis and uncovering insights that were previously hideasden in spreadsheets and static reports. Our second takeaway, empowering engineers through visual data science. Now Spotfire enables every engineer to explore, analyze, and collaborate, combining human expertise with AI powered analytics for faster data driven decisions. And our third takeaway from today's session, building a foundation for the future. With predictive analytics, automation, and global scalability, Renesas is shaping a smarter, more resilient manufacturing ecosystem powered by Spotfire. Now as we've seen with Renesas in today's session, the combination of engineering expertise and Spotfire Visual Data Science is transforming semiconductor manufacturing, turning data into decisions and decisions into lasting competitive advantage. Renesas' journey shows what's possible when people, data, and technology come together. And with Spotfire, every organization can accelerate innovation and make smarter decisions faster. Now with that, this brings today's session to an end. Now before we get onto our question and answer segment, a few things that we wanted to share. So in terms of our upcoming webinar schedule, so as Jingchuan mentioned, just now, we've got some great webinars coming down the line. We actually presented a webinar last week around Spotfire 14.6, which is available on demand. So if you are interested in the latest enhancements, in this first long term support release since 2023, Please do not hesitate, to watch that on demand webinar. Additionally, if you're looking to find out more about webinar or just looking to learn about the latest in terms of what's new, we do have two series available, so please don't hesitate to register to the full series. As mentioned earlier in today's session, a recording of today's webinar will be made available soon. So please do keep an eye on your inbox for the link. If you wouldn't mind moving to the next slidease, Alessandro. Now if you're interested in learning more, please feel free to visit our website at spotfire.com or contact us directly. There are lots of ways to interact with us, whether it is via our socials, through our community. Additionally, our blog site has lots of great content where we share the latest on visual data science, dive into Spotfire data science in more detail. And last but not least, if there are any enhancements that you would like to see or have ideaseas that you would like to share with us, please don't hesitate to visit our ideaseas portal. And as Jingchuan just mentioned, since our last long term support release, we implemented over 90 user ideas ideaseas into, our latest release of 14 dot six. Now without further ado, I'd like to move over to our questions and answer segment. Now before we get on to the questions, if we don't get to your question by the end of today's session, we will follow-up with answers via email. Now in terms of the first question that we have, how do you see generative AI evolving in high-tech manufacturing, especially for day to day engineering tasks? I will take that question. Yeah. That's a that's a really great great question. So, generative AI is going to play a big role, not just in research and development projects, but I think in the daily work of engineers. Just imagine this. Instead of digging through analysis or writing complex code from scratch, an engineer can ask generative AI for help like show me wafer, with, with yield drops in the last twenty four hours or write me a Python function to filter and visualize, failing test bins. I think that generative AI becomes, like a smart assistance helping with data exploration, documentation, analysis, even creating visualizations or data functions. And this speeds up things, up and helps engineers to stay focused on solving problems, not just processing, data. In tools like Spotfire, Generative, AI is already part of the platform. It's not replacing engineers. It's amplifying their expertise, making it easier to move from ideasea to analysis, faster than ever before. So in short, yes, I believe, Generative AI will be part of the engineers' daily toolkit, just like Excel used to be in the past, but, way more powerful, let's say. Beautiful. Thank you very much for that, Alessandro. Perfect. We've got a couple of questions here for, yourself, Moumita. How does Spotfire align with Renesas' requirements compared to other analytics tools in the same domain? So so to be honest, you know, these tools in the in the same space are comparable. You know? I'll be just honest. But the the biggest advantage of Spotfire, what we have, that the the collaboration part of, the engineers along with the analysts and the and the, IT, colleagues, with what we have. We can really leverage, you know, the data. We can providease the data to the engineers in the in their, meeting their requirements, and they can really play around with it, with that data. They can create their own templates. They can really, do whatever they want, actually, you know, with data to meet their requirements. So so that's that's where, you know, the requirements of the engineers kind of, you know, meeting meeting at that point where we can really collaborate and work together. And, of course, with, with, different, added functionalities that Spotfire is coming up with, with each version and their add ons, it is it is making things more exciting for the engineers. To even if they do not have something in mind, specific requirements, it is kind of, you know, directing their thought process into certain direction what, you know, what they can achieve and how they can achieve more with, in Spotfire. Beautiful. Thank you very much for that insight, Moumita. We've got another question for yourself here, based around data points. What best practices would you follow to improve Spotfire dashboard performance when handling millions of wafer level data points? It really starts with a smart, data modeling, you know, in the background. So we really continuously working, to, to have a data model which is intuitive and can really deal, with, with this huge amount of data. And and and then also it also deals, it also, a point where we really need to, create this information links very smartly, you know, for this for the Spotfire to to use. So it is a combination of multiple it it it's a it's a combination of multiple factors, a very good data model, and then, you know, creating the the data links for Spotfire to use, smartly. And then, of course, Spotfire is pretty much able to handle, this this huge amount of data with its capabilities. Also, another thing, sometimes I mean, you know, I will just give you some practical scenarios. Sometimes we really falter on that. Sometimes we really see, you know, the performance is not up to the expectation. But this we we always kind of, you know, make use of different functions of, of Spotfire to use the optic I mean, to use it in an optimal way so that we can get, better performance, alongsidease, you know, some performance tuning with the underlying database. So yeah. So that's how we are, we are reaching. But it's a it's a it's the most challenging part, you know, to be honest, to deal with this huge amount of data, optimally. Yeah. But we I mean, we we take off, the Spotfire, kind of, you know, credibility. We we so far, we we are we are good with that. Yeah. Beautiful. Thank you very much. On that same topic of handling large datasets, I think this would be for yourself, Jingchuan. But, with most of the platforms, question that came in, with most of the platforms that struggle in preparing data, how does Spotfire prepare data efficiently, especially dealing with large, unstructured datasets? Great. That's a great follow-up question. So first of all, Spotfire provideases very dynamic and native data wrangling capabilities to prepare data. Just now you saw the back end of their, dashboard. It looks like some data canvas. It's local drag and drop. Users can prepare their data, combine, or calculate new columns. That's how easy to do the data wrangling part. And just normal, Mina used the word I really like is like smart modeling, to connect the data. So I would also use smart loading for the for handling the bit, volume of the data because we do have a lot of building algorithms, for example, to do the incremental loading if you are connecting to some specific database or dynamic data retrieving only when users scroll down or drill in to specific data, then we will load the data to the the catching areas. So we do have a lot of technologies in Spotfire to make sure this enterprise level demand can be produced for the all the users from the HYDRA manufacturer. Yeah. Fantastic. Thank you for that, Jingchuan Lin. So based on the other customer success stories that you presented, Alessandro, specifically the the Brembo case study, what's the typical time frame for a manufacturer to start seeing real return on investment after implementing a solution like Spotfire? Is it a lengthy deployment, or can we get started quickly? So that's another great question. So what we have typically seen across most manufacturing clients is that the time to value with Spotfire is pretty fast. You don't need really a long project, to get results. Specifically in the Brembo, case, the team started visualizing and analyzing production data just a few weeks. And the first insights on process variability and quality drivers came almost immediately. The thing is that once data connections and dashboards built by engineers, are there, the line of businesses could start making better decision within the first month. And then also with the vast amount of available connectors, it's easy to connect to existing data sources from historians to, MES generated data without, let's say, complex integration steps. And this means that you can start exploring data and sharing insights. And the good thing is that those insights can be can be easily shared across different teams. So essentially, instead of waiting months, most manufacturers feel free to add anything, of your thoughts as well. Beautiful. Thank you very much. Looks like we just have one last question that came through. So whilst we're answering that question, if you do have any final questions to ask to our speakers, please do feel free to add them in the q and a tab. But the last question that we currently have is we have several different use cases where we analyze data and take decisions. Not all of them are related to semiconductor manufacturing. So is it therefore possible to use Spotfire for other use cases, like in sales or energy consumption or even for DRV? I think this question will be more for yourself, Jingchuan. Yeah. Sure. That sounds like a great data science question. I think we started to view the useful options and data science functions for specific industry like what we presented today with high-tech manufacturing. We have a lot of waiver map related use cases. But on the other sidease, we also had pipelines to create the data science functions or actions to support use cases from other industries. I will say from other industries, we can try to gather the common use cases. Let's take time series data as one example. So we start to offer time series data proof processing. But after you prepare time series data, the next step will definitely be the prediction questions, like, what can you predict from your time series data? So this time series data analysis, it doesn't just belong to high manufacturing because for other industries, as long as you have time related data, we do have the good opportunities to try this time series to processing and prediction in future. So that's how we want to drive our offerings in future. And perhaps you will say for the sales, you want to do more prediction, you want to try more algorithm, you want to, even try some deep learning models. This can all be, you know, supported by Spotfire, but it takes some time for our data science team the data scientist to have to think about, what is the best way to create a UI, to create action, to fulfill those common use cases, but we do have the plan to work on that as well. Fantastic. Thank you very much, Jingchuan. As previously mentioned, that was the last question. We haven't received any new questions in the meantime. So with that being saideas, I would like to bring today's session to a close. Once again, thank you very much to our fantastic presenters today, and thank you very much to Moumita from Renesas, for taking time out with a busy schedule and presenting the Renesas use case today. As mentioned, if you if any questions do arise after the session, please do not hesitate to get in touch with us, and we'll, of course, answer your questions by reaching out to you directly. With that being saideas, as mentioned, we're closing today's session, and we hope to see you at one of our future webinars. With that, have a great day, and thank you very much for joining us today. Thank you again. Take care. Thank you very much.