Monday, August 4, 2008
Sunday, August 3, 2008
Littlefield Technologies - Round 1
Round 1 of Littlefield Technologies was quite different from round 2. We started the game with no real plan in mind unlike round 2 where we formulated multiple strategies throughout the duration of the game. Starting off we could right away see that an additional machine was required at station 2 to handle the dual processing load from station one and three. We purchased a machine for station 2 as soon as we gained control over the factory. Looking back now I can see that this could have been a risky move had we purchased the machine for the wrong station. Luckily, station 2 seemed to be right choice as the queues at station 2 and 3 began to flush out and the number of jobs completed per day increased. We then proceeded to watch the game over the next couple of days making no big moves except for switching the priority of station two depending on the load at station 2 and 3.
However, despite the addition of the extra machine we seemed to be in a bad spot. A bottleneck started to build up at station 1 and our revenue per day dropped correspondingly. At that point we felt that it might be a little risky to purchase another machine for station 1 in case we could not recover the investment. So we decided to play it safe by continuing to monitor the queues. This did not work well for us. We continued to lag behind the other teams in last place for a couple of days. At this point we were at wits end trying to get out of the rut that we were stuck in. Finally Mark proposed that we go on the offensive and purchase another machine at station 1. The rest of us agreed since we were already in last place and could do no worse.
So we crossed our fingers and purchased another machine for station 1. The change was instantaneous. The bottleneck at station 1 was flushed out and the utilization dropped down. We slowly started to make more money than the other two teams. We continued to keep a very close watch on the factory and even more on the other teams. Mark and I had the game running at work all day long and I would watch the game for an hour each night before bed. All of us were addicted to the game and could not stop obsessing about it. In the end our obsession paid off when we finished in first place. It was extremely rewarding to finish first especially since we went through a frustrating period thinking we were going to finish last.
Wednesday, July 30, 2008
Tools Used
We used an excel sheet to track our progress over the duration of the game. Having all data located in one place in a usable format enabled us to make fairly accurate predictions on demand and inventory consumption. You can find our inventory management tool here.
Tuesday, July 29, 2008
LittleField Round 2
The second time through playing the Littlefield game was much more planned and the attention we gave the game much more deliberate. From the very get go we knew that our output was lower than demand.
Stage 1- We knew that the hang up was at both station 1 and station 3. The problem for us was deciding which we would purchase first, an addtional machine at station 1 or station 3. We chose station 3. Looking back I feel like this first move may have been our first mistake. In hindsight I probably would have bought an extra machine at station 1 first for the following reasons, 1) It was $10,000 cheaper, 2) utilization was slightly higher there and 3) it only makes sense because all orders have to start at station 1 so what's the use of increasing capacity at station 3 when we aren't producing enough kits further up the line.
Stage 2- Within a cycle or two of spending lots of money on kits and then making lots of money selling the kits under contract 1 we realized that contract 1 made very little money and that we would have to rely solely on contract 3 if we wanted to start tasking control of the game and maximizing or profitability. With this realization came the strategy to risk running out of inventory for the sake of getting another machine at station 3 so that we could produce a finished product in less than half a day and switch our contract over to contract 3. We set our reorder point on kits to 0 and calculated that we would have just enough to buy another machine at station 3 and just enough inventory left to get back into a rythm of buying more kits and making more money. Stage 2 was very successful and was a huge turning point for making money and making it quickly.
Stage 3- While the first two stages focused on doing what was best for us and making the most money we could in the least amount of time stage 3 was about planning how to keep our lead and our strategy working against the other two teams. One of the easy things to see within the game is when teams buy kits and buy machines. We had a good idea of how many machines and kits the other two teams had throughout the game. Our strategy was to stay within the same working capital as the other two teams so that we all looked like we had a shot of winning the game. But behind the scenes start stashing inventory away in small increments so that the other teams couldn't get a good sense of how we were using our money and that we had $100,000 to $200,000 worth of kits stashed away for a rainy day. The problem we encountered in stage 3 is that all of our numbers were shown to the other groups during class just 4 days into the ten-day game. Our whole strategy here was to keep our reorder point and amount of kits hidden. Soon after our numbers were revealed we saw that the other two teams started increasing the amount of kits they were ordering.
Stage 4- We realized on day 125 out of 390 days that we could plan how and when we would purchase kits for the rest of the entire game and then sit back the entire last week and wait for the results. Stage 4 had our team running hundreds of numbers and equations to find the average amount of kits moved per day, the average amount of money we made on contract 3, how many days we could run before reordering etc..... These numbers gave us a pretty precise schedule for ordering kits, keeping up inventory levels, slowly increasing our inventory levels and allowed us to purchase all the inventory we wold need for the rest fo the netire game by day 275. Satge 4 was probably our most successful stage in the game because we had good and meaningful numbers to work with that helped us to calculate within 3,000 kits exactly how much we would use over 265 days. The order points and order levels were accurate enough that we could set back for the entire last half of the game and still feel confident in a victory.
Stage 1- We knew that the hang up was at both station 1 and station 3. The problem for us was deciding which we would purchase first, an addtional machine at station 1 or station 3. We chose station 3. Looking back I feel like this first move may have been our first mistake. In hindsight I probably would have bought an extra machine at station 1 first for the following reasons, 1) It was $10,000 cheaper, 2) utilization was slightly higher there and 3) it only makes sense because all orders have to start at station 1 so what's the use of increasing capacity at station 3 when we aren't producing enough kits further up the line.
Stage 2- Within a cycle or two of spending lots of money on kits and then making lots of money selling the kits under contract 1 we realized that contract 1 made very little money and that we would have to rely solely on contract 3 if we wanted to start tasking control of the game and maximizing or profitability. With this realization came the strategy to risk running out of inventory for the sake of getting another machine at station 3 so that we could produce a finished product in less than half a day and switch our contract over to contract 3. We set our reorder point on kits to 0 and calculated that we would have just enough to buy another machine at station 3 and just enough inventory left to get back into a rythm of buying more kits and making more money. Stage 2 was very successful and was a huge turning point for making money and making it quickly.
Stage 3- While the first two stages focused on doing what was best for us and making the most money we could in the least amount of time stage 3 was about planning how to keep our lead and our strategy working against the other two teams. One of the easy things to see within the game is when teams buy kits and buy machines. We had a good idea of how many machines and kits the other two teams had throughout the game. Our strategy was to stay within the same working capital as the other two teams so that we all looked like we had a shot of winning the game. But behind the scenes start stashing inventory away in small increments so that the other teams couldn't get a good sense of how we were using our money and that we had $100,000 to $200,000 worth of kits stashed away for a rainy day. The problem we encountered in stage 3 is that all of our numbers were shown to the other groups during class just 4 days into the ten-day game. Our whole strategy here was to keep our reorder point and amount of kits hidden. Soon after our numbers were revealed we saw that the other two teams started increasing the amount of kits they were ordering.
Stage 4- We realized on day 125 out of 390 days that we could plan how and when we would purchase kits for the rest of the entire game and then sit back the entire last week and wait for the results. Stage 4 had our team running hundreds of numbers and equations to find the average amount of kits moved per day, the average amount of money we made on contract 3, how many days we could run before reordering etc..... These numbers gave us a pretty precise schedule for ordering kits, keeping up inventory levels, slowly increasing our inventory levels and allowed us to purchase all the inventory we wold need for the rest fo the netire game by day 275. Satge 4 was probably our most successful stage in the game because we had good and meaningful numbers to work with that helped us to calculate within 3,000 kits exactly how much we would use over 265 days. The order points and order levels were accurate enough that we could set back for the entire last half of the game and still feel confident in a victory.
Tuesday, August 7, 2007
Reflections of Week 2 Supply Chain (Area 51)
Reflections of Week 1 Supply Chain Game (Area 51)Supply Chain Game: Round 1Team Area 51 - Mark Barany and Wick Newell
When we started the second week of the supply chain game we were surprised to learn that the two other teams we were competing against came to class with their strategy already mapped out. I am sure Wick and I would have did the same had we actually understood the way in which the order quantity and order shipped parameters changed the game. We experimented throughout the first week with changing parameters, but I think never got the hang of how they worked since we were playing catchup the whole game and shipping every piece of inventory we had.
Since we didn’t have a strategy mapped out, we created one on the fly. One of the very first things Wick and I agreed upon was that we wanted to have at least enough inventory to cover median demand throughout the game. We looked at the plot demand history and found the median of all the countries and then estimated the demand of Enthworpe and came up with a median demand of 135 units spread between the continental countries and a demand of about 20 units for the island country of Fardo. Ramping up production to meet median demand was one of the smartest moves we made in the game.
The second thing we had to look at was where we wanted to build factories and where we wanted to build warehouses. Since we had to make a spur of the moment decision we decided to build a factory and a warehouse on the island of Fardo. We quickly calculated that there was enough demand on Fardo to warrant the decision. It would have been interesting to have enough time to time to actually calculate if the move was worth our time. We decided that since factories are so expensive that we were only going to build one for all the mainland countries. I think this strategy also worked well because the additional costs to ship outside the country were less than the cost of building additional factories.
The last strategy that worked out well for us was that we decided to make sure we had inventory on hand at all costs. This means that even if during low demand times we still had our production at full blast so that we could stock up on inventory for high demand times. This strategy also saved us some headaches because it allowed us to just keep producing and shipping to all our warehouses while we were sleeping. One of our biggest worries was completley running out of inventory in the middle of the night and not being able to make adjustments until the next morning. By just ramping up production overnight and then letting the inventory burn off throughout the day we were able to manage our inventory with greater precision.
In hindsight, there are a few different approaches that I would have changed. One of those things is the quantity we set for our factory to produce. We chose 135 units a day purely because it seemed like that was the median demand. We forgot to factor in that trucks make the most money when they ship 200 units and the truck is full. Next time I would have chosen to produce 133 or 150 units so I could ship in batches of 400 or 600 without waste.
The other big thing I would have done differently is I would have stopped production sooner. I assumed that demand was going to be steady throughout the end of the game. But by the last 14-20 days demand continiously went down until it almost reached zero.
One thing that I learned from playing this game is that it is important to add capacity as soon as possible so that you have more time to watch your investment make a return. But one of the most important things I learned was that it was beneficial to always have inventory on hand because the times we did the worst is when we lost sales due to a lack of supply. One of our main objectives throughout the game was to monitor the plot lost demand table to look at where we couldn't keep up with demand and try to change that so that we could end up having no lost demand.
When we started the second week of the supply chain game we were surprised to learn that the two other teams we were competing against came to class with their strategy already mapped out. I am sure Wick and I would have did the same had we actually understood the way in which the order quantity and order shipped parameters changed the game. We experimented throughout the first week with changing parameters, but I think never got the hang of how they worked since we were playing catchup the whole game and shipping every piece of inventory we had.
Since we didn’t have a strategy mapped out, we created one on the fly. One of the very first things Wick and I agreed upon was that we wanted to have at least enough inventory to cover median demand throughout the game. We looked at the plot demand history and found the median of all the countries and then estimated the demand of Enthworpe and came up with a median demand of 135 units spread between the continental countries and a demand of about 20 units for the island country of Fardo. Ramping up production to meet median demand was one of the smartest moves we made in the game.
The second thing we had to look at was where we wanted to build factories and where we wanted to build warehouses. Since we had to make a spur of the moment decision we decided to build a factory and a warehouse on the island of Fardo. We quickly calculated that there was enough demand on Fardo to warrant the decision. It would have been interesting to have enough time to time to actually calculate if the move was worth our time. We decided that since factories are so expensive that we were only going to build one for all the mainland countries. I think this strategy also worked well because the additional costs to ship outside the country were less than the cost of building additional factories.
The last strategy that worked out well for us was that we decided to make sure we had inventory on hand at all costs. This means that even if during low demand times we still had our production at full blast so that we could stock up on inventory for high demand times. This strategy also saved us some headaches because it allowed us to just keep producing and shipping to all our warehouses while we were sleeping. One of our biggest worries was completley running out of inventory in the middle of the night and not being able to make adjustments until the next morning. By just ramping up production overnight and then letting the inventory burn off throughout the day we were able to manage our inventory with greater precision.
In hindsight, there are a few different approaches that I would have changed. One of those things is the quantity we set for our factory to produce. We chose 135 units a day purely because it seemed like that was the median demand. We forgot to factor in that trucks make the most money when they ship 200 units and the truck is full. Next time I would have chosen to produce 133 or 150 units so I could ship in batches of 400 or 600 without waste.
The other big thing I would have done differently is I would have stopped production sooner. I assumed that demand was going to be steady throughout the end of the game. But by the last 14-20 days demand continiously went down until it almost reached zero.
One thing that I learned from playing this game is that it is important to add capacity as soon as possible so that you have more time to watch your investment make a return. But one of the most important things I learned was that it was beneficial to always have inventory on hand because the times we did the worst is when we lost sales due to a lack of supply. One of our main objectives throughout the game was to monitor the plot lost demand table to look at where we couldn't keep up with demand and try to change that so that we could end up having no lost demand.
Wednesday, July 11, 2007
Week 3- June 27th
I must admit that I was just downright excited about our week 3 class (which is no small feat for someone who has pretty much been in school for eight out of the last ten years). This was the class where we started an even more complex supply management game. We used the first fifteen minutes of the class to go over rules and to get a better understanding of which variables we could control. We then spent the next 15 minutes looking through the graphs, charts and text within the game to better understand our objective and to evaluate the history of transactions that happened before we joined up with the game.
With the push of a mouse button Professor Hughes started the game and the clock ticked forward. Every real-time 15 minutes equaled one day in the game. To be quite honest my teammate Wick and I felt like we had very little idea of what we were supposed to be doing by the time the game was underway. And as the first few days past and we became accustomed to the game's interface it felt like we had even less of an idea of our strategy or even what variables we could change.
By the time one hour passed (or 4 game days) there still had been very little movement on the board amongst any of the teams. One thing I did learn (from the beer game) is that it is important to start ramping up supply for the long haul early so that we don't have to start the game in a huge hole and try to make up for supply throughout the whole game. By day 5 within the game we made a push to improve our factory production from 20 barrels a day to 40 barrels at the cost of 1 million dollars.
The felt like the move to build an aditional factory producing 20 additional barrels was the best move we could make to start the game and get a leg up on the teams we were competing against: donothing, 3 guys and scam. The only thing I was questioning was should we build even another factory on top of that for a total yield of 60 barrels a day?
After buying an additional factory we instantly dropped to last place in overall capital. At that point all we could do was pack up, go home and watch for 90 (in game) days to unfold and watch the effect it has on our profit.
With the push of a mouse button Professor Hughes started the game and the clock ticked forward. Every real-time 15 minutes equaled one day in the game. To be quite honest my teammate Wick and I felt like we had very little idea of what we were supposed to be doing by the time the game was underway. And as the first few days past and we became accustomed to the game's interface it felt like we had even less of an idea of our strategy or even what variables we could change.
By the time one hour passed (or 4 game days) there still had been very little movement on the board amongst any of the teams. One thing I did learn (from the beer game) is that it is important to start ramping up supply for the long haul early so that we don't have to start the game in a huge hole and try to make up for supply throughout the whole game. By day 5 within the game we made a push to improve our factory production from 20 barrels a day to 40 barrels at the cost of 1 million dollars.
The felt like the move to build an aditional factory producing 20 additional barrels was the best move we could make to start the game and get a leg up on the teams we were competing against: donothing, 3 guys and scam. The only thing I was questioning was should we build even another factory on top of that for a total yield of 60 barrels a day?
After buying an additional factory we instantly dropped to last place in overall capital. At that point all we could do was pack up, go home and watch for 90 (in game) days to unfold and watch the effect it has on our profit.
Week 2- June 20th
During week 2 of our supply chain management class we played the "beer game." The beer game is a classic learning experience for those involved in the supply chain and operation management world. The game is comprised of a factory, distributor, wholesaler and retailer. The object of the game is to keep the supply flowing evenly throughout your given timespan, but without keeping surplus stock by the end of the timespan.
I think most grad students within the class had played the game before and had a pretty good idea of what strategy they would use from the very get go. The only problem is that I think we all didn't learn from our past mistakes and relied to much on the "old" information coming from the other teams. When I think back on how I played this game last time (In Jim Rand's Operations Management class) I remember how throughout the whole game we always struggled to keep up with supply. So the big question is, How come we were unable to see the same predicament happen all over again?
I think one of the obvious answers is that we weren't allowed to talk with one another. Some of us who had been exposed to this game could have clued in other players as to the expected outcome. But I think an even bigger factor is that we actually trusted the relayed information being sent from down the pipeline. We were oblivious of the fact that we were adjusting allocations based on news that was considered to be 2 weeks old and even more oblivious to the fact that the supplies we were sending along wouldn't reach the retailer until 8 weeks after their request.
In hindsight I think the strategy we should have used was to have the factory send out as much stock as they could for the first 8 weeks and push the product as quickly as possible to retailers to make sure that they could keep up with initial demand. Once those first 8 weeks were over then I would start sending quantities out based on the average demand you saw over the first 8 weeks. The thing you absolutely want to avoid is being stuck without product early in the game because most likely you are going to spend the whole game trying to recover from the loss.
I think most grad students within the class had played the game before and had a pretty good idea of what strategy they would use from the very get go. The only problem is that I think we all didn't learn from our past mistakes and relied to much on the "old" information coming from the other teams. When I think back on how I played this game last time (In Jim Rand's Operations Management class) I remember how throughout the whole game we always struggled to keep up with supply. So the big question is, How come we were unable to see the same predicament happen all over again?
I think one of the obvious answers is that we weren't allowed to talk with one another. Some of us who had been exposed to this game could have clued in other players as to the expected outcome. But I think an even bigger factor is that we actually trusted the relayed information being sent from down the pipeline. We were oblivious of the fact that we were adjusting allocations based on news that was considered to be 2 weeks old and even more oblivious to the fact that the supplies we were sending along wouldn't reach the retailer until 8 weeks after their request.
In hindsight I think the strategy we should have used was to have the factory send out as much stock as they could for the first 8 weeks and push the product as quickly as possible to retailers to make sure that they could keep up with initial demand. Once those first 8 weeks were over then I would start sending quantities out based on the average demand you saw over the first 8 weeks. The thing you absolutely want to avoid is being stuck without product early in the game because most likely you are going to spend the whole game trying to recover from the loss.
Subscribe to:
Posts (Atom)