 Hi, global supply chainers, good morning, good afternoon, and good evening for those that are far away from here. We have students from more than 180 countries, which is very, very exciting. So welcome to this fourth Hangout event. My name is Eva Ponce. I'm the Executive Director of the MicroMaster Program, Supply Chain Management. We are very lucky to have today with us Dr. Josue Velakev. Dr. Velakev, he is the Director of the Latin American Global Network Scale, and he's also a research associate here at the MIT Center for Transportation and Logistics. He has prepared the screen game. This is an innovative role-playing game, an interactive tool that I hope you have enjoyed to play the game. And he has also prepared for today a debrief in order to discuss with you the conclusions and to highlight the main results of this game. So thanks a lot, Dr. Velakev, to be with us today and to prepare this material for our SCX still. We are approaching the end of SC3X. So I know you are working hard. You just need one last push, the final stretch, take the final exam, and then you are going to be done with SC3X. So please keep up the good work. Today, in addition to have Dr. Velakev, as I mentioned before, he's going to discuss the first 15 minutes, the debrief of the screen game, we are going to have Professor Sheffi. Professor Sheffi is the Director of the Center for Transportation and Logistics, and he's an expert in risk and resilience. He has been working in supply chain management during the last 40 years. So I think it's a great opportunity and an honor for us to have him today in this Hangout and to give you the opportunity to discuss with Professor Sheffi those questions you might have on risk and resilience. The idea is as in previous Hangout, we are going to start with this debrief, then you are going to move to the breakout session for 15 minutes. This time, you know that we always are trying to include an improvement and to do things a little bit better. So this time we have with us our postdocs in the team, they are going to join the breakout session in order to facilitate and moderate the discussion. So the idea is that all of you have the opportunity to share your opinion and help you also to identify these relevant questions before to come back to the lobby and share with us your questions and discuss with Professor Sheffi. So I hope you enjoyed even today. Please join me in welcoming Dr. Velázquez and thanks again. Now my pleasure, my pleasure Eva. So well, good morning. It's really a great pleasure to be to be here this morning. And if I understand correctly, you already have the slides and the materials and also the results of the game. So we have this game for I think it was one week or two weeks actually open. And I have all the ideas to just introduce a little bit about the game and then to show the results and some interesting insights and conclusions that you can take with this. So as Eva said, I'm a research associate and I lead the initiatives of the Center for Latin America. And if you can actually have the presentation open, like showing the slides with the animations or not, just let me know, I will try to lead you with the number of the slides. So we are now actually in the slide number two. And this is actually the main, let's say, assumption of the game. When we deal actually with general variability in supply chain, usually we deal with the type of solutions that I related more into the concept of safety stock or trying to work with lead times or to try to work with forecast tools. When we deal actually with the disruptions and these resilience in supply chain, these things actually have very low probability of happening. But once they happen, actually this can destroy the supply chain the way that we are used to manage. It's much more than just general volatility, it's something that is really hard to predict. And that's why we deal with this with a slightly different type of mitigation strategy that I will try to elaborate a little bit more. So just to make it very clear for you, we didn't give you any information about the type of disruptions that you could have in this screen game. But this is part of the tools or the conditions context of the game. You don't know this. The idea is that we are going to play the role of a dictator. We are going to create just disruptions as scenarios that we may believe this can happen. And then the idea is we are going to test your policies, how good they were under these conditions. So if we go very fast to slide number three, we have the description of the screen game. So you were giving a very basic simple supply chain with supplier plan at distribution center. And the idea is that we are going to have disruptions in either supplier plan distribution centers. And the idea is that you should decide robust mitigation strategy that aims at minimizing the total cost, and at the same time, and maximizing the order of filling. And the idea was to give a five-digit code in which you have the inventory that you plan to keep in working process, in finished goods, and also the configuration of the type of strategy of backup plan distribution center or supplier you expect to get. Anyway, so if we move just to slide four and five, we have some of the questions that we actually were expecting you will be asking before deciding your policy. You could actually just decide to go with a safety stock, which means not to do anything, right? Because clearly, safety stock and backup inventory are different. The safety stock is for the general volatility of your supply chain. And the backup is specifically for the cases of this disruptions in supply chain. You could actually decide to have some backup inventory, either in working process or finished goods, or you could have a combination, right? Like backup in facility plan distribution centers, suppliers, or combinations of all of these. So the idea is what is actually the best mix to be more resilient in case you have a disruption in the supply chain. Again, you also have different alternatives. You can either decide to go for a strategy that protects you more in the supply, or they're more in the plant, and they're more in the distribution center. And the question is also what could be the best way to have a better backup in case of disruptions in supply chain. Anyway, so if we go to slide number six, this is just a general discussion on how we generally decide policies or solutions in logistics or supply chain management. And usually what we do is that we follow the common approach of minimizing costs or maximizing the investment or general approaches that are very single objective. So in this case, I'm showing just a dimension of cost. And then we have two mitigation strategies, one that is in the green circle and the other in the red circle. And then we say, well, what is the best mitigation strategy? And following this common approach in operations management, we can always decide. I mean, hopefully you are also driving to the same conclusion of selecting the green circle, which is the one that has the minimum cost. So this is a very basic approach on supply chain management, but actually it's a tool that we use very commonly for many logistics decisions. But if we go now to slide number seven, and now imagine that we have exactly the same policies, but now we have disruption in the supply chain. And if we have a disruption, now it turns out that this green circle solution is actually the optimal cost, but it still is providing the lowest service level, while the other solution that was much more expensive is providing a higher service level. So we have these two extremes, and this happens because we have a disruption in the supply chain. And the interesting part of this is that if we actually have a situation of not having any disruption in the supply chain, both solutions are providing exactly the same service level. But once we have a disruption, then it's when we start assessing the difference between these two strategies. And the idea of this is that in case we don't have any disruption, then actually that one that was very expensive can put us out of business. And in the case we have a disruption, actually the other one that didn't have any investment in cost could put us out of business as well. And this is the main reflection of this. When we make a decision on supply chain management, particularly in the case of disruptions, we are actually expecting to do an investment that maybe we will not get back never. It's like thinking in the return investment of a health insurance. We pay for the health insurance, but we expect not to use it never. So if we go now to the slide number eight, we actually come out with a general idea of this trade-off between service level and cost, which means we can actually get more solution that just these extreme points of trying to find the best solution from the case of service level and the best solution in the case of cost. So getting all these other solutions can actually solve the problem. And among these solutions, we can actually find those solutions that are actually in this front, the front of the efficient set of solutions that are dominating the rest of the solutions. So if we see the solutions that are actually in the red circle, we observe that these solutions are dominating the rest, but at the same time we cannot distinguish which one is better than the other. While one, for instance, the green one is the best in terms of cost, I can find another one that is maybe not as good in cost like the green one, but this is better in service level. And then we can find the other that actually is better in the same terms until we go completely to the other side of the front, which is a solution that is the best in service level. And to decide which one is the best between them is not an easy task, right? But just finding solutions that are in that set, then we are getting closer to finding a better mitigation strategy. So, well, in general, if we go just to slide nine and 10, we get the concept of robustness in a system. So we are trying to find a mitigation strategy that can actually cope with different potential disruptions or scenarios, right? And if we find a solution that most of the cases belongs very close to the front for different scenarios, this is a robust solution. This is able to actually deal with a variety of potential disruptions and those has a high ability to tolerate these perturbations, right? And in slide number 10, we have an example of this. We have, this is just example. So imagine that we have three different disruptions that are in these scenarios. And then in the first scenario, we get a pariet of front with solutions A, B, C and D, right? So all of them are the best solutions we found for this case. And then we go to scenario two and then we get A, B, G and H. And then in the third scenario, A, F, G, H. So by watching these different results for these hypothetical scenarios, we can argue at the end or conclude that this policy A is a robust solution because actually in these three scenarios belong in the pariet of front. So then if we go to slide number 11, I can show you the scenarios that we use for the screen game. So these scenarios were created based on previous disruptions we observed in supply chain. And in general, if you observe, for instance, scenario one in this table, is the scenario in which we don't have any disruption, which means we start a disruption in week one with a duration of zero and we are ready to be online in the first week. So nothing happened in the distribution center plan and the supplier. But if we go, for instance, to maybe the scenario five, in scenario five, we can observe, for instance, that we have a disruption distribution center that will start in the week 12. We last for 36 weeks and then we will be ready in the week 48. And then we have scenarios like this for disruptions distribution centers in plants and suppliers. And then we created this new concept of, we name it also like scenarios, right? For instance, the sunny day. So sunny day scenarios, which we have the probability of 100% having the scenario number one, which means nothing will happen. And we have, for instance, another scenario that is even probability. Then this considers 10% for all the previous scenarios that I'm just showing in that slide. And then we created combinations like this and using these new scenarios, we assess the best policy, looking for those policies that belong most of the time in the predator front. So I hope this is quite clear. So if we go now to the slide number 12, then actually we build this framework to make the assessment, right? So if we take, again, the red circle in the left, we get the optimal cost solution with actually the lowest service level. And in the other extreme, we get the highest cost solution with the maximum amount of service level. And then if this is just a line, just a curve that is actually showing an estimation of the predator front. Although we know that in this case, this area or this curve is actually non-converged, this is just an approximation just to show how it might look. And then we can actually draw a line, you know, in the vertical axis when we have the service level, we can draw a line of saying, well, you know, I will not accept any mitigation policy that provides less than 90 or 95% of feel rate, for instance. And we can actually later draw another line that is in the horizontal axis for the cost to say, well, this is a line of the average cost. So if I have a policy that provides higher cost than this, then I'm not interested in that problem. And then later, I can also draw another line, you know, a line that goes between the minimum cost and then the average cost, you know, this red line. And this one just distinguishes those solutions that are gonna be closer to the predator front, right? So those solutions are actually very good solutions because are providing for a fair amount of cost, good level of service for this case. And then we can actually distinguish the different areas now. We have those that are in the red circles, it is like a clear red circle in which we are not interested in those policies, right? For instance, the one that is in the left-hand side is more like the greedy approach, right? So we don't want to invest a lot of money because we are following the traditional approach of operations management, of looking for solutions that minimize cost, and then we get very low service level. And if we go to the other square, there is in the right-hand side, then actually the other approach like saying, well, you know, I care a lot about service level that I don't care about investing all the money that I need. You know, I put a lot of money and I'm protecting myself, you know, with very conservative approach. And then of course the other, that is the darker red circle, square, I'm sorry. If you are in this area, you may be looking maybe for another profession, right? Because you are investing a lot of money and you're getting very low service level, right? But then the other solutions that are in the blue and green area, are actually smart solutions, right? Those that are in the green area are very close to the priority front and those that are in the blue are also fair, you know, for the amount of money you invest getting sufficient amount of service level. So what we did actually with your policies was to count how many times you were either in the red area, in the blue area, or in the green area. And we give, well, if you are in the green area, we give you one point, I'm sorry, two points. If you are in the blue area, we give you one point and if you are in the red circle, you're zero points, right? Because there are not solutions that we are interested in. So let me just go very fast for the results, right? So if you go in slides 14 and 15, you can see the list of participants. So I expect to have all your names or the EDX ID names there, so you can find yourself and we are assigning a specific number so you can actually use this number to find in the charts, how do you do, right? So let me just show you some charts as an example. So we have in slide number 16, there's like a scenario and then in the blue line, you can see the priority front. So actually this priority front actually is not complex as I mentioned before and then we can see the rest of the participants, you know, their policies, how they did, you know, and in which area they like. And well, in this case, we see some solutions very close to the green area, the blue area and others that are in the red areas, right? If we go for instance to the other, the one that I just mentioned before, even probability. In even probability again, we get some solutions and we have one solution that actually belongs to the circle, right? But it's interesting that in these 16 and 17 slides, we observe for instance, the circle, the blue circle that is actually the number 12 participant is very conservative, right? So he's investing or she's investing a lot of money to keep very high service level, right? And then on the other side, we have the other optimal call solution, which is like a pentagon or what is it? Yeah, I think so, which is a 47, right solution. So anyway, just to show you another one, when we get DC down long-term, so this scenario considers getting for like 36 weeks of shutdown of the distribution center. And in this case, again, we get solutions that actually are not very resilient, let's say it, and we get more solutions that are now in the red circle, right? So anyway, I will go now to the slide number 19 and just to show you the winners. So in the winners, we have all those teams that actually got very good scores. Practically in all the scenarios, their solutions were in the green set. So they got actually all the possible points and interesting also to observe the differences. For instance, if we go for the 36 team or participant, actually the backup policy, just check out the configuration, it's 666, right? But if we go for instance to the number 30, actually it's 537, and it's quite interesting how by having different configurations, you make it actually a similar results in terms of how robust is your solution for a mitigation problem, for a supply chain risk management problem. So anyway, so let me just conclude by showing some observation of the screen game, and I'm sure that you'll learn a lot following the course of Professor Sheffi on resilience and general, we know that actually different policies perform better in different scenarios, and it's always better to get a combination of redundancy and flexibility, right? Which means to have also inventory, but at the same time also to have backup facilities and distribution centers. And in this case, we just created the scenarios, as I said, playing the role of a dictator, but this actually is an informed process, and well now that Professor Sheffi will take over the hangout, he's gonna talk also of potential things that may happen in different supply chains and in the world, right? And it's important to be informed on this and to be prepared for this. Some of the key learnings come in the slide number 21, and just keep in mind that nobody gets credit for solving problems that didn't happen, right? So this is also part of resilience to be prepared for something that has very low probability of happening, but actually they happen, right? And just keep in mind also that it's always better to, in general, right, to be prepared and to have backups for the entity of the supply chain that is closer to your customers, right? Because if that disruption occurs in the distribution center, usually you are affected immediately and usually you should be careful in that. Anyway, so with that, I say goodbye. Emma, so it was very nice being with you and now I leave you with Professor Sheffi and if you have any questions, at the end of the slide, you can find my email, is the toughest email ever in MIT, hosue.mit.edu, it's been a pleasure to be with you. Dr. Velazquez, thank you so much, hosue. It's a pleasure. I'm very happy to offer our SCX student this innovative interactive tool. I think this is something that they really much appreciate, some tool that they can interact, they can include the decisions and things also for this debrief and to analyze the results with them. Thanks a lot. My pleasure. And now we are going to start with the second part of this hangout. So it's my pleasure to have here today Professor Sheffi with us. Professor Sheffi is the director of the Center for Transportation and Logistics and he's an expert in risk and resilience. So he has a lot of experience with different industry, many companies. I hope you are going to take advantage to have today Professor Sheffi with us in order to ask him those questions, those question regarding to risk, to resilience that you want him to discuss with you. Thanks, Professor Sheffi, to be with us today. Nice to be here. Perfect. So first thing we are going to do is we already shared with you three articles. One is related to Samsung Galaxy, Note 7. The second one was the Hanging Case and the third one is the Brexit. So this is just three ideas to start the discussion. So we want you now to move to the breakout session to discuss with your peers. We are also to have our postdoc moderating the breakout session and after 15 minutes, please come back to the lobby, come back to the chat, share with us your questions and those topics that you want to have Professor Sheffi thoughts on these topics and we will be very happy to discuss with you. Thank you. Okay.