 So, welcome. Hopefully people can hear me. Good afternoon. Good evening. Welcome to our second food risk predictions webinar. I'm delighted to again to be joined by two very esteemed members who will take us through some interesting discussions today. And it's a follow on from a previous webinar. We did approximately one on six weeks ago. My name is Neil Marshall, and I'm the moderator for today's session. And with me, I have the pleasure to introduce Dr Chris Elliott, who is the Professor of Food Safety at Queens University Belfast, who is going to give us some more very insightful information and maybe some breaking news today during his presentation. I also have Janis Stortis, who is the CTO and partner at Agrinom, who are a technology company who are going to give us some of the details around the risk prevention exercise today. Apologize, Mrs slide. Yeah, so there's the pictures of our panelists and hopefully you can see down the right hand side, the people. I'm sure most people know Chris for his work in food fraud and in the UK and Ireland, and also across Europe and globally now as a very esteemed part of academia. And Janis who can also provide subject matter expertise from the technology and solutions to support the predictive analytics. So, next slide please. What is our goal today so our goal today as I said is try to be interactive. We're going to have some questions and some data shared by Chris. And then we're going to go into a demo and some insights from Janis to talk about the food archive platform and how that can inform some of the technology and data, because data is critical. But for our session today, I mean the key two points are to get those industry and academic insights from Chris and there are some real new breaking news as I mentioned. And to see Chris discuss and debate some of the data that's in the platform and some of the information he's bringing forward. And also with Janis to use that technology and try to use the technology to connect those insight between data and reality, and where does that convergence occur. So we can use that in our day to day work across the food industry, not only in a theoretical context but also in a practical example, so trying to give you some data and insights that you can use in your day to day roles. So without further ado and much debate. I just want to just again re-emphasize really the industry need for that data and why do we need it. We shared some of this previously on the previous webinar, but it's really more and more important than ever before now to avoid major incidents. We've been through a year and a longer now with the pandemic. And as Chris mentioned last time, maybe we're storing up some incidents or issues for the future. And I know food fraud is one of those topics that's growing. And we'll probably see more incidents because of the pandemic. And we need to be more proactive and less reactive using data insights and the future insights to give us that information to do things ahead and plan and change our testing programs to make sure we're being aware of what's around the corner and to prevent those issues happening by being more informed. So there's a whole piece around informing and being prepared. But there's also the fact around data and multiple data points out there and where do you see the wood from the trees because there's so much information. But one of those things and again, I like the quote that Chris mentioned last time, you know, use technology to inform us to save money on the recalls to invest in the future and put people like Chris out of a job. Yeah, so it's trying to be informed, trying to use that data and predict the future using AI and technology so you can spend the money in a different way and not invest it afterwards on cleaning up the recalls in the crisis. So with that, I'd like to pass to the next slide and hand it over to Dr Chris Elliott to give us insights for today please Chris. Many thanks Neil and it's great to be part of this second webinar. I think we had a lot of fun on the first one. And for those of you who don't know me, I really do investigate serious issues about food safety contamination food fraud locally nationally and internationally. And also as Neil mentioned, I think the risks associated with both food safety and fraud have only heightened over the last 18 months and most likely will continue to for some time to come. So maybe if we can flick on to the first slide, data is complex. And I think all of us use data in some way to make decisions. And I think in terms of our food supply system or supply chains or networks as I prefer to call them is they generate a huge amount of data. And I think we want to really start to get into a position where we can exploit that data and collect lots and lots of disparate data together to actually make sense out of it. And I talked a lot to people from from the food industry and they are rich in data as well. And the problem is not quite sure what to do with it, because it's layers. It's a layers of spreadsheet. It's another spreadsheet. How do you bring all of that together? And I guess my really strong interest in connecting with agronome over over the last year or so is about that connectivity of data, the interpretation of the data. I just came off literally 30 minutes ago and part of the UN summit on our food system and it was about food safety. And I was asked to present what I thought were the five biggest risks in terms of food safety going forward. And I put climate change number one, and I will not deviate from that. And I think the outcome of the discussions was absolutely climate change is going to have a massive, massive impact on the level of risks, the number of risks, emerging risks. And it will be old risks in new places and new risks in new places. And there's going to be so many things that are going to happen over the next relatively short space of time because I can already see climate change having an impact on food safety. I spent quite a lot of time, 2019, part of 2020, investigating a particular food incident that caused hundreds of food poisoning incidents, killed five people, four of them young children, and it was directly related to climate change. So what we want to do now is, yes, looking back over data is really important, but that allows you to be, you know, a little bit more reactive. And I think what we need to do is not only look at the data retrospectively but in real time, and that will give us so much more information. So we could move to the next slide please. Again, expressions that we've all heard about that, information is king, knowledge is par. Well, I think in the area of the arena of food safety and food fraud, that is absolutely true. I spent a lot of my time collecting information from multiple sources. Some of it I would call open source, some of it close source, a lot of it one-to-one conversations with different informed people, and it's about collecting that information and interpreting that information. So access to the right information at the right time is really very, very important. From that, we can understand risks in terms of what geographic region is now presenting us with the biggest problems and do we source from those regions? If we source from those regions, should we introduce new measures of auditing, testing and so forth, or actually if the risk is so high, can we move that sourcing to somewhere else? But it's not only country specific, you can get it down to commodity specific, ingredient specific, or even company specific, and that is hugely important and insightful information for the food industry. So moving on please. This is a term that I came up with a few months ago actually interacting with agronome, and I call it my digital crystal ball. It's looking forward and looking forward based on data and data analytics. And I think in terms of predicting what the problems will be, not today, tomorrow, next week, but next month the month after that, or even a year after that is incredibly important. And to reiterate what Neal said was, if you as an organization can get away from being reactive from having to deal with food recalls, from having to deal with having your name on in a national newspaper, because often companies contact me to say, I've been implicated in this food safety scandal and this food fraud scandal, what do I do, I will tell you it's a challenge, it's a massive challenge, challenge financially, reputationally, so many different ways. Now, over the past, yes, perfect move on, that was great. Over the past week or so, I've had some, this is how I entertain myself, working with agrino and I've been trying to say here based on my different sources of data, I think some of the big risks are, and then we've been using the agrino way of collating and analyzing information to say it's the same. So here's the first thing is, going to ask you a question, you've got 20 or 30 seconds to answer, then we're going to hear what your thoughts are, and then we will find out what collectively the agrino food archive system has said, and then I will give you my comments on whether I think it's correct or not. So one of the following ingredients have the highest increase in incidents over the next 12 months, and you think it's poultry, oregano or olives, so if you could vote for one of those now, just to think about what might be really on the radar screens and causing risks and just give it some thought and then give us your vote now. Chris, maybe I'll just take advantage while we're just going to get for the questions. I was also, as we spoke earlier, I was on the call with you this morning as well, the UN summit, I thought some of your insights around the protein, the GMO and the plant base were really interesting as well. I don't know whether you're going to cover that later, but I think the whole allergen piece is also very interesting, the fact of new emerging allergens that you mentioned as well. Yeah, thanks. So we've got the results of our poll. This is a bit like the Eurovision song context, except we didn't put no new points, because we know we know the UK doesn't get any points. 50% oregano, 38%, sorry, poultry, then oregano, olives. If we move on to the next slide, perfect. Thank you very much. So here are one, two, three, four, five, six, eight different ingredients that we looked at on the food, Akai database. If you look at the last column on the right hand side, and this is the prediction in terms of trends, what are likely to be the biggest issues. Flip down through that and you can see absolutely top of this league table is oregano. poultry actually where the majority of people voted for, we don't see or certainly food Akai doesn't see any big issues at all. Oregano and the food Akai said 400%. But independently, and when I, when I looked at that list, top of my list was oregano. And I think probably in the next few weeks, there will be some breaking news about oregano and which will absolutely vindicate this piece of data. So I think I give the 10 out of 10 to to the food Akai for this one. So let's move forward to our next question or next poll. So three more commodities three three particular issues that might crop up mineral oil is present in wine lead and ground turmeric or fin bendazole in beef. Again, have a think about which of those you think might be the biggest problem that's going to emerge. Take a few seconds about this. Just to let you know that, but whenever these slides were being put together, one of these was already on the list, and it was already top of my list even before we started to discuss which I thought was was very very insightful. So do we have a vote. And this time we've got lead ground turmeric, followed by and bendazole and mineral oil. So thank you very much for that. So if we close the poll and then go on to our next slide. So again, here we've got six or seven different types of ingredients and commodities. And what we look at in terms of herbs and spices and you look at them lead chromium and mercury in ground fennel in ground turmeric coming from India. Other issues with different types of herbs and spices as well. So, I think if we could flick on to our next slide. Another question is in terms of increased predicted risk for salmonella in pork, the stereo in beef or heavy metals in herbs and spices. Again, and what what do you think might be the biggest issue the biggest risk that we're going to face over the next 12 months. A few more seconds to think about that. What are you going to pick. What would you like for dinner tonight. Salmonella list area or heavy metals. So we're going to get the results of our third poll very soon. Here it comes. And wow, 84% for heavy metals and herbs and spices. I think 84% people voted for that because of the information that was on the previous slide, you know, there was a warning there that there is an issue. So this is the power of information the power of data. So if we close the polling now and just look at these ingredients again. Thanks. Moving down through the list. Absolutely heavy metals is and going to be an issue in herbs and spices. Now, what I'd like to just and very quickly said why is there issues with heavy metals and herbs and spices. Is it about accidental contamination or is it deliberate contamination. My view is it's a combination of both of those. You know, you can get added particularly lead can get added to some spices to increase the color of it on the perceived value of that. And from my information sources is and there has been quite a lot of that going on, because the prices of particular herbs and spices is absolutely soaring now. I have an issue about supply and demand, which I think the food archive databases picking up very nicely. Also interesting to see that the wine is coming up at a very high predicted value there. So it would be interesting to understand and why the food archive databases picked us up. It was just, I think, quite interesting yesterday. I picked up an article to say that there's likely going to be massive fraud, not only in wine but actually in the beverage industries, due to pandemic related issues. So, I think, and this, this data that's coming is really, really very powerful. Okay, so if we could move on to our next slide. So, moving away from ingredients moving to commodities. Here we have your choice. What would worry you the most in terms of sourcing sourcing from Egypt sourcing from China or sourcing from Turkey. What do you think that they would be in terms of risk, where you source your materials from, you know, they're all countries that have a lot of food exports, they are countries that do a lot of food processing as well. And so it might not be the country of origin of the particular commodities or ingredients and but are passing through particularly in processing. So again, we'll see what the participants which I say we'll have more than 100 people on this webinar which is fantastic. So here we have China, absolutely, as our number one worry. And again, I would, when I saw this data, I picked China, and but I also had Turkey quite high up on my and my risk register as well, mainly because of the amount of processing they do there, mainly because the amount of processing of herbs and spices, and that they do in Turkey. So if we could close that poll and then just move on to the next slide please. Actually, here, here we can see the three biggest hits in terms of trending countries, I guess this this is the, the, the graph the map that you really don't want to appear on, but we see that the China, absolutely very darkly colored. We see Brazil and we see Mexico. And one of those surprise me, I've heard quite a lot of issues coming out of China. Many, many different issues arising there, and some of it is about fraud, some of it's about safety, some of it's pandemic related, some of it's about water shortages and particular parts of the country. There's a complex business going on interested to see Mexico and because recently and some of my lectures on food fraud I've been warning about Mexico, and I've been warning about Mexico because there has been more and more reports about organized crime drug cartels, taking parts of the industry there, particularly avocados, and if you, if you want to read something to scare you Google blood avocados and you'll see what's going on in Mexico. Brazil is a country absolutely massively impacted by the pandemic. And I think that might be why Brazil is trending highly there as well. Now, one thing is getting the data that the other part is the interpretation. So thank you for that. So we'll maybe move on. I think this is the final question or the final poll in terms of an emerging food fraud issue. You've got saffron, olive oil and milk, three very very different commodities three very very different ingredients. Some quite bespoke in terms of the saffron, olive oil globally traded, but milk, one of the true commodities that that is global in nature. So I'm interested to see what you're going to pick for saffron or sorry. I mean, there was a little clue there sorry and gave it away I hope you've already voted so let's see what the audience thinks about this saffron olive oil and milk. So it's not surprising that olive oil is appearing front and central and saffron and milk following up in quite even numbers. So if we go on to our next slide and see what what food archive tells us about those three different commodities. Now we could just close the poll for some time, probably most of this year, my own sources of information intelligence gathering has said saffron is going to be a major issue about supply and demand. So here we have. This is was breaking news very recently is a massive fraud was uncovered by your poll, the European police force, looking at organized crime, working in saffron, and the backdrop to this is Iran is the world's biggest producer of saffron but actually saffron is higher in value. And it looks like a lot of the Iranian saffron was was being smuggled into Spain, being re labeled and called Spanish, and you know, the value of this particular fraud was was over 10 million euros. This is big business really big business. So next slide please. So, that was just my overview my little, you know, few insights about what I've tried to predict using my sources of information what the food archive does. There is very good overlap there's good harmonization, but sometimes the food archive and databases will throw up things that I were not on my radar screen before, and I think that's the big value of it. And my research group now we are constantly mining this this food archive database for so many so much information now. So what I'm going to do is stop talking, close my mouth, open up my ears, start to listen. I'm going to pass you on to Janice now who's going to talk a lot more about the what sits beneath the the bonnet of agronome and food archive and how the system actually works. So Janice, over to you. Thank you so much Chris. That was so interesting. Thank you very much for sharing this breaking news. So, first of all, I would like to introduce if we go to the next slide please I would like to introduce our company briefly. We started in 2008 and we specialized in data management problems for food and agriculture throughout all these years. We have been working for the food industry. We have been working with academia in international organizations to deliver some of the most important projects in the area of agriculture and food like the scientific, the very important scientific search engine. And that is run by a file the data discovery and data sharing a project in the global food safety and private public partnership work working with many of the food companies that would like to transform the risk assessment to a more digital format and to a more digital methodology that can be used to drive the decisions, the very important decisions. Today, we are particularly focusing on big data processing and artificial intelligence to extract the food safety insides but also predictions related to the food to the global supply chain. Slide please what we do is that we use our data and our technology to support the food safety professionals make effective and timely decisions, making sense out of the millions of data records that exist out there. And how we actually do this is that we have developed a technology that provides easy access to all this very important insights that food safety expert needs. We have developed a technology that uses the power of the artificial intelligence but it uses also the knowledge and expertise of the domain expert of the food safety experts in order to deliver to provide food risk assessment and prediction insights. It's also an easy to understand way and actually to deliver something that can be actionable that can be used to take very important mitigation actions and to finally prevent risks. Slide please what is behind all this is millions of data records so we have a continuously growing database that is scanning and processing information from 97 currently data sources, different types of information, including recalls, border rejections, inspection results, results of the laboratory testing and monitoring programs that are run by the authorities, suppliers information and all this information is processed and used in order to deliver and to produce the insights and the predictions. The prediction models are continuously updated with all the is with the latest information. So it's not a static version of the models but we ensure that we have that our models have the knowledge, using the most frequent information. So our prediction is based on the robust methodology. We have developed something that we call the intelligence which equation. In this equation in this approach, we always start with the business question. It's very important for us to understand what is really the problem that the food safety experts need to solve. What is exactly the question that we need to answer, and to do so. First of all, we identify together with the food safety experts, which are the parameters, the drivers that change the risk that affect the risks, and behind these parameters, which are the data sources that we can use. So this is the data collection part. After doing that, we are selecting the best artificial intelligence method, the best prediction method that fits to the nature of this problem, and that can be used to predict a very important indicator that will give us the answer to the business question that will give us that will address the problem that we are trying to solve. Next slide please. So our approach, our risk prediction approach is based on four steps. The first step is the risk identification, using all the scanning, all this scanning process that I described. The second step is to identify to monitor and to identify the risks for any ingredients. The second step is to perform a risk assessment for all the ingredients for thousands of the ingredients that we have in the food data. The next step is to perform a risk assessment for the suppliers. So using the information of all the recalls of the inspections of the border rejections, we are applying risk assessment algorithms, and we are delivering a risk score for the suppliers that can be used to prioritize audits, but also certificates of analysis, laboratory testing results and the corresponding plans in order to have the best control of the risk. And the final step is using all this information, of course, to have timely risk prediction. Next slide please. So let's try to see all the things that Chris described, but also the methodology that I described. Let's see how this works in action. Let's see these things in action. So I will share my screen in order to show you how these predictions, how these insights can be accessed and can be provided by a platform like Fudakai. So as I mentioned, Fudakai is collecting information from many different data sources and can be customized because it's very difficult to mine the information, all this information. The platform can be customized to specific ingredients that is of interest for your supply chain. It can be customized and adapted to any, to all the ingredients that you are using, and this will give you the ability, will give us the ability to monitor the trends for this ingredients. For instance, using analytics techniques, we can identify which are the ingredients from the ones that I have in my supply chain that have increasing issues. Here we can see on the left side of the screen the increasing issues block of the dashboard that the Fudakai platform provides, where you can see which are the ingredients that have a high increasing trend, sorry. So we see for instance that Ground Turmeric is increasing the trend based on the reports that have been provided by the national authorities. We have a high increase during the last months, the same stands for sesame seeds and for cumin and also other ingredients. This is a live information, so every time that we have a new information announced by a national authority, these trends are updated. On the other side, we can see the emerging risks, we can see increasing and new issues for all these ingredients. But this is the risk estimation, this is the risk assessment view. So we are using the information of the hazard severity, but also the frequency, the information of the frequency of the incidents and of specific hazards happening for specific ingredients to estimate a risk score. So this block can provide you the information for all the ingredients that you have in your supply chain, the information about which risks are at a high level, which risks are increasing, and which risks may be new and I need to take them into account. So when we see this information, this information we can go deeper and study more about this information. So if we select, assume that we are a company that is sourcing poultry meat from a region like Brazil, which Chris mentioned that is one of the trending countries for the next month. So I can click to view more data and to perform a very good hazard analysis for this specific product ingredients category. So we can see which is the trend of the incidents, we can select the specific region so we can focus only on the issues coming from this specific regions, from the specific region from Brazil. So we see that the majority of the issues that is causing this increase in the risk is are the biological hazards and specifically you can drill down by selecting the category and see which subcategory of the hazards are affecting very much poultry meat coming from Brazil and specifically you can still going even become more specific and going deeper by selecting pathogen, which what types of pathogens this is that it's pathogenic bacteria. And finally, to see which are which bacterias are the ones that are causing this increase in risk, and also to go behind the diagrams and study all this data. And we can also perform a risk assessment, a full risk assessment of this specific category, and this risk assessment will help us very in a very easy way, and very fast to identify which are the increasing hazards which are the increasing risks. For instance, we see that Salmonella is increasing by 25% and I am repeating that all this data, all this, sorry, all these insights and assessments are based on historical data that we have. We can see which of the risks are decreasing, but we can see also if there are any new risk, and I can be focused very much and see specifically what is happening in each case. So, for instance, we can see if we go down, we can see a case of labelling with description like declaring that this less fat but the product actually had more fat. We can see also cases like Fipronil that is decreasing right now, so this may help you to perform a deep risk analysis. But what if we want to go one step ahead to go to the prediction part. If we want to predict which will be the hazards or which will be the trends based on all this historical data for a category like poultry meat. So here we provide an approach that is based on four steps, a predictive approach that is based on four steps. The first step is actually what we see on this blog, where we can, we have all our ingredients and the system automatically checks and the prediction models provide us the information for which ingredients we will have more incidents during the next process. So we see here that one of these category of ingredients is also the poultry meat that we were analyzing in terms of hazards but also in terms of risk just a few minutes before. And the second step that I can do is that I can check the hazards, I can check specifically using the predictive analytics, what will be the trend of the incident. So I see that it is predicted that there will be a slight increase of the incidents, but not something that is very, very big in terms of the trend. But still, it is very important to see which are the hazards that are likely to increase. And applying the models, we can see here that based on the information that we have and using the predictions that the models are providing to us. We see that Salmonella is one hazard that will increase in an important, has an important increasing trend for the next month. And the same stands for different stereotypes of Salmonella but also for cases like Listeria. This trend of the hazards can also help us to provide a prediction for the risk for poultry meat. So we see here that we see a risk, a hit map for risk for the main risks for poultry meat and we see that Salmonella is at a medium level. And the predicted based on the prediction models, the Salmonella will go at a higher, at a high level of risk. And this is further analyzed using the evolution of the risk for Salmonella in poultry meat. And you can see specifically how this risk will involve during the next months. A very important thing is that we are predicting also the emerging risks based on the latest information that we are collecting. So we see here that in the case of poultry meat, we have some cases with misdescription that will be increasing and emerging within the next months, but also a new stereotype of Salmonella that was identified and will also affect some products. And this kind of analysis, can this kind of predictions can help us to see which are the products or the suppliers that will be affected because the products or the suppliers are providing to us this ingredient, this type of ingredient. So we can see here that one product and one supplier would be affected. The models are providing also other products or ingredients that may be affected by the increase of Salmonella. So you can see also other products that based on the knowledge of the models may be affected by this increase of Salmonella. And a very important thing is that you can see also the trend for the country that based on the predictions, it seems that for Brazil, the trend of the incidents will increase during the next months. So we use this four step analysis that helps us to identify which are the ingredients that have and will have increasing trend for the incidents, which are the hazards that will likely increase for this ingredient. We also check the risks, but we also are sure and we can immediately identify how your finished product and suppliers are affected. So I will give some more examples in terms of prediction. There's just a lot about ground tourmeric Chris mentioned the case of ground tourmeric. So this is, this is a specific ingredient that it seems that will have an increase of incident during the next 12 months. As you can see one of the hazards that are likely to increase is the presence of lead, one of the heavy metals, but also Salmonella it seems that will be increased so again you can see how the risk profile is affected and will change. And also how the Salmonella in this case will increase sorry for such an ingredient like ground tourmeric, and in the same way as I also presented for the case of poultry meat coming from Brazil. If you get your ground tourmeric from India, you can also see the trend of the incidents coming of the ingredients that are searched from India, so we can see also the trending, the trend for this country. And again, you can also see other cases like for instance if you are sourcing chili peppers from Mexico, which is again a country that Chris highlighted based on the predictions it seems that we will have more incidents within the next few months. And as you can see here that chili peppers will have more incidents during the next months. The hazards that are likely to increase are several chemical issues, like Fibronil, which was present in the past in the supply chain but it seems that the models based on the recent information that they predict that we will have more issues in the future. We can see which are emerging hazards that are a lot of emerging issues, mainly a link to chemicals. There are several chemicals like Oxamil, but also ethylene oxide that seems that will affect also these chili peppers. And again, you can see the trend of the Mexico for the incidents for the next few months. So I can stay here for hours and present you all the information, all these predictions, but I think that we need to switch back to the presentation. Thank you very much for your attention and I will be very happy using all this data but also clarifying how we are producing these predictions to answer these questions during the Q&A. Just Janice, before you go back, I just fantastic overview there of the platform, but when I see it I think, well, Janice is obviously an expert, could I interrogate and use the platform like that? How difficult is it for people to use and do what you just did for the people on the audience? As you saw, it's one of the main goals when we were developing all these insights and developing all these services for our platform is to provide a really easy access to all this information. So it's super easy. I can confirm that because most of the people that are starting using the system within a few minutes are very familiar with the system. They can find with few clicks all the information that they need. Of course, there is always a live chat that the people can use and we can provide support or any clarification. And we know that what changing from the current way of estimating risk with the robust methodologies that are already used to a more digital way is something that needs change. So we are here for the people, for the companies that want to make this step to move to a more digital risk assessment and risk prediction. And we can help them by starting the current process by redesigning the process and helping in the change management. And this will enable the, we will help to automate the process. So it's very user friendly, Neil, but what I want to for you to keep is that we are here to facilitate the process of changing the way that risk assessment is performed today with the way that you want to the risk assessment to be data driven to integrate all the knowledge and the experience that you have right now, and to optimize the process. Next slide. Okay, so thank you for the demo, I think it's transitioning back to me again there so you got the benefit there of the expertise of Chris for the science and the academia and the fantastic insights is giving us already there to combine. The platform that Janice has just described, and even for someone who's not so technically advanced like me I think you could see click click click click, and you go down and you get all the data in one place which is a fantastic way of doing your risk assessment process. And really the benefits you know out of that you know that the real bonus for me is getting the insights and showing Chris's example out he's leading his team and their researchers are using that data. And they're adding insights that they're finding from industry and research, and then also from the technology in the platform. It's allowing you to use that data to make decisions. And this is the critical piece for people who in industry want to use the tool on the platform. It's also being aware of what your supplies are doing. As Chris mentioned some of those supply chains get complex, particularly like avocado example, the turmeric example, the oregano, you need to be looking at your suppliers, you can potentially then monitor how you assess and check on the you can change your testing and inspection programs based on what you see from the data from the platform, and the insights you get from Chris. And the main thing really is trying to avoid recalls we want to say for food supply system, and we want the food to be safe for everybody. And that really goes affect everybody and I think the other watch out that Chris mentioned apart from the climate change is the almost hopefully not but probably a backlog of issues around wine beverages food from the pandemic, because we've just stuck in docs and transportation that's probably going to get tried to be reused and re resold elsewhere so I think that's a big opportunity for using technology and looking at the data to predict. All right, please. I think what what we're just quite trying to show here is really the agronome slide to show that different models cover all these areas, particularly useful around hazard analysis can supply checks, but there's a basic, there's a premium, and there's a diamond package. Obviously would vary, but the main thing is the company's agile and they can adjust the offerings and the support, depending on the size of the company, the amount of people that you have, and you wish to use the platform. But the main thing is how you automate your processes and build it into the solution, but the company are really agile and can be adapted to your request. Next slide please. What we're also saying scan the trying to move again with technology use use the code scan it with your phone or your camera phone to use the scan. If you get in touch with Anna and the team, they can work out a program to give you insights on your specific ingredients from your company. They can do a demo they can set that up for you and give you some of these live predictions around what could happen for you for your production from your products. So I think you know that's a good step to take, following this webinar. So please investigate that further. All right, I think we're going to try to go to some Q&A and some questions. I don't know if we've got some already. Just looking at the chat. Yeah, we have several, several questions, Neil, some, some are already answered but I can start answering some things and you can complement. Sure, sure. So which one do you want to use from the chat room. I will start from the last one and we'll go. That's a question. Yeah, so for the audience. Yeah, what is the role of traceability here. That's the question. Yeah, so there's another one by I will answer both. So what is the role of traceability. We rely on the location information that we have from the reports. So if they are mentioning a country as an origin of specific ingredient, we will use this information, and we don't have any further information to be more specific in terms of the region. Okay, if they don't give us this information in the reports. So what we have in addition to that is in some cases we have the UPC numbers. So we have some more information that can be used to identify if this is an incident that is connected with a specific UPCs and also a lot of numbers so you can check if this is affecting a specific lot, a specific batch of an ingredient that you are getting so I'm saying this because this has happened in the past we have companies that we are working together that they were notified with an alarm that was an incident. We checked we provided the lot number or the information they checked so they made sure if this will be affecting. But in any case, we always suggesting that having the trends at the level of the country, specifically for risk is something very important because the traceability issue is not solved throughout the supply chain so it's better to have more information and to be more sensitive than trying to be more accurate and lose some very critical information. The next question is about the risk score based on the number of the incidents. If we are using only the previous year or all the incidents that we have in the database, we are using the maximum period that we have used is 20 years of data. In order to select the optimum period of time, we apply validation methodologies and we select the period that gives us the best accuracy for the prediction models. Just try to multitask and look at the question at the same time. I was trying to find one for Chris to answer that is one that you saw Chris that you want to give a response to. The question that you asked Neil was how user friendly is the system. Yeah. I think it's important question because I don't know a very kindly given access access to the database for quite a while. But I've been asking over the last couple of months to get access, you know, can this student do a piece of work and this student do a piece of work. The first time I asked you was about two days ago for one of my students and today I got an email from that student saying this is unbelievable. Look what I was able to find because she was searching for particular problems with with illegal pesticides. So that was, I think that's a clear indication that it is unbelievably user friendly because often it takes you a while to find how to navigate systems. Absolutely Chris. And I think, you know, your other comments earlier around the herbs and spices and the analogy between what you've seen and what you were hearing to what was in the tool is also a good testament to the data, you know, you can always question the data, but it's the intersection I think that you're providing with your insights is really useful. Yes, and also, you know, there are some things that the food act kind of databases throwing up that is not on our radar screen and as you can imagine those are the ones that we're really interested to drill down in terms of what is really going on here. Yeah, absolutely. Okay. Any other questions that's just a look at some compliments to you Yanis on the insightfulness of the score. If anyone's asked about you use any kind of BI or Bayesian algorithm. That's one for you Yanis it's too complicated for me that one. It's a very good question thank you for asking. Yes, yes we are using. We are using several artificial intelligence and machine learning algorithms, specifically Bayesian network we are not using Bayesian networks but we know that this is one of the methodology that can very well integrate the knowledge of the experts with the data for each parameter with with many different parameters that may be affecting risk so this is one of the things that we would like definitely to try. And, and don't just. Sorry, I just see another one here that maybe this is a question for your team as well for after the webinar, but as a use case example where you can share in terms of using saving cost savings. I think that you have some of those case studies already maybe Anna can share afterwards or people contact can contact Anna for information to some case studies already I remember on herbs and spices that you did previously. I'm not sure you could that can be shared. What, what, yeah what I can share Neil this is an excellent question what I can share right now is that there are two ways, two very important ways of saving cost, and based on the experience working with the companies. The first is that the ultimate goal is to prevent a recall. So, having early the notification about another iteration in honey was something that he previously one of the companies that we are working together to save to prevent a recall the same was with the ethylene oxide. Another very important cost saving factor is that using this information you can optimize the laboratory testing plan that you want to apply the specific ingredients. And of course, it's always the matter of time that you need to spend to at least monitor all this information. Whereas here with one click or doing just a two minutes visit on the platform, you can have a very good horizon monitoring of all the risks. Absolutely I think that's the other bit we didn't touch on today because we were, you know, basically focusing on on the predictive analytics but the dashboard inside the tool gives you all that in one place so you only need to go one place. Every day you get the information. I know we're about at time now and Anna will slap my hand if we don't do the webinar post webinar question so we'd like to I think one more poll just to give people a rating on how good or bad we were today. I think that would be useful so if you can finally complete before you log off the recording will be shared and available. Thanks again tremendous insights again from Chris as always, and thank you also to yourself Yanis for giving a deep dive demo on the platform on the tool. Please complete the webinar evaluation and contact Anna and you can see the contacts there for further information. I would like to analyze your suppliers if you'd like to do a demo, contact the team, the more than happy to set up sessions and help you protect your food supply. So thank you everybody thank you again Chris, thank you Yanis, and it's been a pleasure as always doing this session. Thank you. Thank you. Thank you Chris. Thank you all for your attention for participating. Thank you. Thank you everybody. Bye.