Speaker 1:	Thank you for attending this session. Here is the game interface. Customer orders come in, and if you click on this icon, it will give you the orders that are coming. When an order arrives, you need to put 60 pieces of raw material on it. In the first game, don't worry about this because inventory is managed by the game automatically. You don't need to worry about anything, but cost of each one is $10. Therefore, overall, for each order, you have a cost of $600, and that's it. Ignore any other costs. When you receive orders, there is a lower limit and there is a upper limit to deliver that order. For example, it may be 1, and 3, and 1,000. What does it mean? It means if you deliver the order in one day, you get full $1,000, if it is delivered in one day or less. But if it passes three days, you get zero. If you deliver it in two days, you will get 500. You get 500, but the cost of raw material is 600, so while you are delivering the products, you have negative $100 profit or $100 loss. If you deliver after three or more days, you get zero and the cost is 600, so for each unit of product that you deliver, you lose $600. The whole story of this simple game is to match supply with demand, production capacity with demand. You do it in two different ways. On one hand, you try to increase capacity, increase capacity, increase capacity. But if by increasing capacity, you cannot improve the situation, then you immediately need to go and reduce the demand. How to reduce the demand? You create a block here. You don't allow the orders to come in. How do you do that? Here, it will allow you to fix the work in process, the maximum number of orders that come in. In this game, you have two important decisions and one not very important decision. Two important decision is this. Whenever you find it out that the flow time in your company is going to be greater than or equal to one day, you first try to reduce incoming orders, not to allow them to come in until you reach the balance of delivering everything in less than one day. You go over there, based on the capacity that you have understood here, you know what the flow time is under the current capacity. If you want to reduce the flow time, the only way that you have it over there is to reduce inventory. Throughput is your capacity. Indeed, it's a little bit less than your capacity because systems which want to set their production equal to capacity, they will observe a large waiting line. I will explain it by the end of the semester, but for the time being, you accept that no system can perform at full capacity. Whenever someone says, we are working at 120% capacity, that is not correct. Orders come in, they get the required raw material. They go through Station 1, then they go to Station 2. From Station 2, they come to Station 3. They go back to Station 2, and then they go out. What are the two important decisions? If you think that flow time is longer than one day, go and don't let products to come in, reduce your maximum work in process. Alternatively, find the bottleneck. Suppose this one has capacity of five per day, this one has the capacity of 20 per day, and this one has capacity of 10 per day. 5, 10, and 20. You have a chain, which has three links. Capacity of one link is fine. Capability of the second link, for example, is 10, and capability of the third one is 20, and you want to lift a load. What is the maximum size of this load? This one can carry 20, this one can carry 10, this one can carry five. A chain is as strong as its weakest link, and the weakest link here is five. Capacity, which we usually show it by R_P, can never be equal to throughput. It is always greater than equal to throughput. If capacity of the bottleneck, which is capacity of the system, is five, suppose you decide to think about four units of throughput, four orders per day. If that is the case, therefore, you go and look at the Little's law; throughput times flow time is equal to inventory. Throughput, based on your capacity, you know that it is four. You may purchase another machine and increase the capacity to eight or even nine. Flow time must be one because if it goes beyond one, you will not earn $1,000 per order. You want it to keep it one or less. This should be equal to inventory or work in process. Cannot exceed four. If your current work in process maximum is equal to 100, you need to go over there and quickly reduce it to four to make sure that your income per order goes back to no. Two important decisions. Block here. Do not allow new orders to come in until you send whatever you have it here out to reach to a point that everything else which is inside here is delivered in less than one. The second alternative, increase the capacity of the bottleneck, either reduce demand or increase capacity. One other thing that you can do in this game, but it does not have that much impact on your performance, incoming product come here, they go over there, then they come here, then they go over there. Station 2 gets product from this station or from this station. You may set priority equal to first in, first out, first come, first served, FIFO. First in, first out, or first come, first served. Whatever comes first, goes first. Or you may set priority to those which come from step 1 and those which come from step 3. This is the basics of the game. Now I take you to different window. Now let's go through different tables. I first click on incoming orders. This window opens. There are some information. You may look at all of them. But I plot job or I may also copy this, come here and paste it. Then that graph is also over there. You may try different Alpha; 0.4, 0.3, 0.2, 0.1. The reason that graphs looks like this because we have rounded it. If I remove that round, if I go back and remove the round part, things look like this. Say, like that. Now if I change this to 0.9, it looks like that. You can practice whatever you have learned in exponential smoothing, find the best Alpha using math and so on and so forth. Or you may come here and type intercept and then apply a regression line on it. These are my Ys. These are my Xs. You may also lock this if you want to copy it here or there. F4, Enter. That is intercept. Now I go here, copy to the right. Still, because I have locked it, it refers to those cells. Slope. Now I have slope and intercept. Or I can alternatively go here, Add Trendline, Linear trendline, Display Equation. Equation is also displayed over there. You can really play with whatever you have learned in forecasting part of this course and implement it, not just in theory, but in reality, or at least in virtual reality. The first part of the semester, we were dealing with forecasting. You can implement all those forecasting here, and data is revolving, so you may do it for Period 1-50 and 60, 70, 80, 90. If the pattern of data is changed, for example, if in Day 150, you see that the data is different than what it was before, you may throw away those parts which are different and just deal with what is more recent. I'll go back to the site of the game. I'll go to completed order and plot completed job counts. I download it, open, and that is my completed orders. Control C, Control V, and the data is here. That is input and that is output. We may assume each of them or average of them as throughput. Usually, we assume output as throughput. Equal to average, Tab. Average of what? Input. Enter. Average of input is three or 2.5, and then I copy to the right. Average of output is 2.46. We assume this one as throughput. Next step, I can do inventory. Inventory at the end of first day is input minus output. Therefore, at the end of first day, you have one unit in inventory. You will start the second day with this inventory plus input minus output, and that is inventory at the end of Day 2, inventory at the end of Day 3, inventory at the end of Day 2 plus incoming minus outgoing in Day 3. Inventory at Day 4, inventory at Day 3 plus incoming Day 4 minus outgoing Day 4. Then click, click, copy to the right. Therefore, now we have throughput R, and we have inventory. According to the Little's law, inventory and throughput have a relationship like this. Inventory is like a reservoir. Throughput is like a flow. Suppose we have inventory of 10, and suppose we empty it at rate of two per hour, for example. We have 10 gallons of dense liquid here, and we empty it at rate of two per hour. Then next molecule which comes here, we want to know how long does it take to go from this point to this point? In order for this molecule to go from this point to this point, all other molecules should go out. Everything else should go out. We have 10 over here. We empty it at rate of two per hour, so it takes five hours for them to go out, and then it takes five hours for this molecule to go from here to here. Of course, when it reaches here, already other molecules have come and are there behind it. Therefore, if a new molecule arrives here, again, 10 gallons are ahead of it and they are emptied at a rate of two, so it takes 5 hours. Then for this one to go from this point to this point, it takes 5 hours. Of course, when this comes to this point, all other molecules behind it have also come here. Again, over there, there are 10. I hope this explanation can be sufficient to know why in the Little's law, flow time is equal to inventory divided by throughput, and therefore throughput times flow time is equal to inventory. Now we have inventory, and we have throughput. Therefore, flow time is equal to 1.42/2.46, and that is equal to equal to this one divided by this one, and that is 0.57 or 0.58. Therefore, at the current situation, it takes another 0.58 days to go through the system and we set as long as it leaves before one day, we collect $1,000 and we spend 600 on raw material, so we make 400. But if it takes two days, we only make $500, and we need to spend $600 on raw materials, so we lose 100. If it takes three days, we make zero, we need to spend 600, we lose 600 per order that we satisfy. If you are in this situation, it is better to create a wall in front of orders and do not allow them to come in. One thing in this game we do is to stop the demand, to bring the demand to where we can help. Now, let's talk about supply because on the other side, we can increase supply. I go to Station 1, I click on it. I download utilization. This is utilization of Station 1, download it. Yes. B. That is utilization of Station 1. Let me just take this flow time down here in order not to mix things together. Go to Station 2. Plot utilization. Download, open. Yes. Control down, Control C, Control V, and that is utilization in Station 2. Then we'll go for utilization in Station 3. Plot utilization. Download, open. Yes. See control. These are the utilization of different stations, and then we copy to the right. Utilization of the first station is about 0.55, around 0.2, 0.2. What does it mean utilization. If a utilization is 0.5, which is almost the case here, that means whatever the capacity was, 50% of it was utilized. The rest was left over it. Station 2, if this is the capacity, with this production of 2.46 per day, only 20% of it was utilized. Utilization is equal to production. Divided by capacity. If it is equal to 0.2, that means 20% of time, the server was busy and 80% of time, it was idle. In this course, I use R to show throughput. I use R_p to show capacity. U is equal to R divided by R_p, and therefore R_p is equal to R divided by U. We do have R, which is 2.46 per day. We have utilizations so in each case, we should take R, and I lock it because I'm going to copy to the right. Divided by utilization. This is the capacity of Station 1. This is the capacity of Station 2, and this is the capacity of Station 3. Let's for this one, just for simplicity, assume it is five. For this one, assume it is 13. For this one, assume it is 12, just for simplicity. Now, we have a chain. Capacity of the first link is five. Capacity of the second link is 13. Let's say this is five, this is six, almost twice of it. Capacity of the third link is 12. These are associated with two cables. Now, capacity of this one was five, capacity of this one was 13, capacity of this one was 12. Now I want to put a load here, and I expect this chain to take that load up. What would be the maximum capacity, maximum weight of this load? 12, 13, 25, 30, what it is. A chain is as strong as its weakest link. Therefore, this chain, no matter this is 13 or 13,000, the capacity of this chain is five. Because this chain is not strong as 12, it is not strong as 13. It is only strong as five. Capacity of the process is equal to the capacity of bottleneck. This system right now, can produce five product per day. And if inventory inside this system is greater than five, then it cannot finish it in one day, according to the ladles law. Now, suppose we are in this system for a minute. Suppose we go here and compute inventory and we come with the inventory of the last day, which is accumulated of all days, suppose it is 20. Therefore, 20 loads are inside of our system. Capacity of our system is 4.5. This is capacity, but capacity should be always greater than throughput. Throughput should be always less than capacity. When you are in a highway, all cars cannot drive very close to each other at speed of 65. As soon as one of them makes a mistake, the whole system is destroyed. We can never produce at full capacity because the smallest mistake if operator is not there, if too many products are coming together, the waiting line can significantly goes up. Therefore, if my capacity is 4.5, I assume my throughput is equal to four. Therefore, I can expect to produce four per day, and that is my throughput. Inventory is 20, throughput times slow time is equal to inventory, four times T is equal to 20. T is equal to five days. Any new product which comes in, it takes five days to send the product out. If we send a product out every five days or revenue is equal to zero or cost, is equal to 600 or profit. Is negative 600. We should make some changes. The first thing I will do, I immediately go to WIP and I reduce it. At the same time, I may decide to increase this capacity also. It is 4.5. I may decide to buy two new machines. If I buy two new machines, that is 4.5 + 4.5 + 4.5. That is equal to 13.5. Therefore, the capacity of this station goes up to 13.5. But I cannot go ahead and set WIP. Equal to say 13 because this one is no longer the bottleneck. The bottleneck shifts to this point, and capacity of this point is 11.77, and therefore, I may assume that I can have a throughput of 11 per day. If throughput is 11 per day, then I should go and set my WIP = 11. After asked how I make these changes, I will let you know. I'll go here to WIP, edit data, and I change 100 to 11. Click Okay. It asked me for password. And then I go here to Station 1. Capacity was 4.5. I want to buy two new machines. I change 1, 2, 3. Okay. Ask for password, and that's it. As soon as you buy a machine immediately, because each machine is 90,000, 90,000 is subtracted from your case because you buy two, that is 180,000, but the machine will be immediately available. The capacity of Station 1 quickly goes from 4.5 to 13.5. Thank you, my friends. Don't forget. In my class for day 2, I have said this is a pool learning, not push teaching. Furthermore, this game is a competition. Therefore, the team which is more alert, the team which makes better decision will stay first. I should also tell you when you buy a machine, you may buy it time. You may buy it early, you may buy it late. If it is time, you will be fine. If it is late, you lose profit. If you buy it early, you don't make interest on your money that you have because as long as your money is over there, it's not turned into equipment, it earns 10% interest. The game starts Saturday October 17 at 1:00 PM. It ends exactly seven days later at the same time, but the game will continue. For 50 more days. At 12:00, the next Saturday before the game ends, you need to decide whether you want to keep your machinery or not. You may sell your machinery and suppose earn $10,000 or $9,000 for each and put it in your pocket. But in the last 50 days, you will not produce anything. But you may keep your machines for that last 50 days, and they may make more profit than their sales price. But don't forget if you keep them at the very end of the game, they sales price is zero. The salvage value is zero. You cannot sell them. There is a trade off. Should I sell the machine before the last 50 days and collect the sales price or keep them? Know the sale price because at the end of the game, the value is zero but benefit from the income of the last 50 days. Thank you very much. I think I have explained much more than enough. I let you to drive the rest yourself with your teammates. But if you have a specific question, you can come and ask. Nevertheless, we will not tell you how to play, how not to play. That is something that every team should work on it. Thank you and best wishes.