How to win erpsim game

SAP CLOUD Question

Team 6 B. Yang-Vaernet | C. McCulloch | J. Manuel M. Hunt | R. Langowski

ERPSIM Visualization & Analysis Project HEC Montreal Advanced Game

2

TABLE OF CONTENTS:

EXECUTIVE SUMMARY ................................................................... 2

INTRODUCTION ............................................................................... 3

STRATEGIC PROBLEM ................................................................... 3

RECOMMENDATIONS ................................................................... 10

OPERATIONAL PROBLEM ............................................................ 12

RECOMMENDATIONS ................................................................... 21

CONCLUSION ................................................................................. 22

3

Executive Summary

1. Purpose The purpose of this report is to evaluate our team’s performance through our nine rounds of the HEC Montreal ERPsim Advanced Manufacturing game. Our goal was to maximize production and sales in order to gain the highest net income and company valuation, not only to win the game but also to challenge ourselves and put into practice the skills we gained while studying cost accounting. This report aims to identify the problems we encountered that hindered our overall performance and further understand the causes of those problems.

2. Problems Our team identified two main problems we suffered during the nine-round game, one strategic and one operational. Strategically, we failed to identify the most opportune markets in which to produce and sell products. Operationally, we struggled to hold a sufficient amount of inventory as illustrated by the rate our inventory was leaving the warehouses. These problems were chosen as the main issues we wanted to tackle because after analyzing our data, we believe that they had the most significant impact on our overall performance

3. Findings We found that a potential reason for our strategic problem was the inopportune timing of market entry and exit in certain markets. This caused our company to lose out on potential profits and was an opportunity cost to producing other products. Additionally, we placed relatively large material requisition orders which essentially gave us a fixed production schedule, leaving little room for flexibility if we wanted to switch between products. For our operational problem, we believe the root of our inventory depletion problem came from setting the price below the market average and not shipping enough goods from the main warehouse to the regional ones. These two causes resulted in our inability to optimally sell all of the products we produced. 4. Recommendations To improve our ability to identify the most propitious markets to sell products, we should not only enter successful markets based on the ZMARKET data, but also hypothesize where we might be able to gain an edge in niche markets during the wait for the data to be prepared. Secondly, having smaller and more constant production runs will correct our timing of market entry and exit. Regarding our operation problems of depleted inventory and insufficient shipping quantities to regional warehouses, our recommendations are to monitor prices better and proactively change prices every 5 days to remain competitive. In addition, we want to enhance our communication between sales, production, and shipping and explore factors that affect shipping through increased analysis.

4

Introduction

Our group, Team O, ranked No. 2 in the ERPsim Advanced Game over a series of nine rounds. We divided our roles into pricing, production, shipping, and marketing. Together, our goal was to work cohesively in order to earn the maximum potential net income and company valuation. Our strategy before starting the game was to sell only a few product lines and place our focus on distribution channels 12 and 14 in order to utilize price leverage via marketing to gain higher margins. Despite our slow start in round one, we were able to quickly solidify ourselves as one of the top performing teams. However, after five rounds of play, we not able to keep our first place position. Ultimately, due to a combination of external and internal issues, we ended the nine rounds in second place.

After analyzing our performance, we identified a strategic and an operational issue that we were unable to correct during the nine rounds of play. We were also able to identify several potential causes of these two main issues. Regarding the strategic problem we faced, we failed to identify opportune markets to produce and push our products into to obtain significant market share. Operationally, our inventory sold out quickly in several rounds, leaving little inventory in our regional warehouses to sell. In this report, we will go in-depth about the strategic and operational issues we faced, explore the various potential causes of these problems, and discuss recommendations we would make going forward into the final six rounds of the Advanced Game.

Strategic Problem

The primary strategic problem we faced was that we failed to identify the most favorable markets to enter, as well as exit unfavorable markets ahead of our competitors. Our strategy was to produce just a few products in order to minimize production time and focus on the markets for those specific products. The products we chose to initially produce and sell were 500g Nut Muesli, 500g Raisin Muesli, 1kg Nut Muesli, and 1k Original Muesli. Later, beginning in round four, we made the decision to produce 500g Blueberry Muesli and 1kg Strawberry Muesli to add to our product mix and maximize our overall profit. Due to our narrow product mix strategy, we needed to be alert about where those few products were selling well and gain an edge before the competition entered the same market. While we were able to achieve around 95% productivity each round, our team struggled to find the sweet spot for our few products to maximize profit and market demand.

Graphs to Illustrate Strategic Problem

Figure 1.1 illustrates our initial struggle to maximize market demand in round one. Our company, represented by the green bars, sold nearly an even quantity of 500g Nut Muesli and 500g Raisin Muesli. However, the graph displays that 500g Nut Muesli had a much higher market quantity demand than 500g Raisin Muesli (59.93% to 40.06%). Proportionally, as shown in Figure 1.2, we should have sold much more 500g Nut

5

Muesli than 500g Raisin Muesli than we did. The issue shown in these two visualizations was that we were unable to sell more 500g Nut Muesli with regard to the market quantity demanded, thus supporting our main strategic issue that we were unable to properly identify and take advantage of the market.

Figure 1.1: Round 1 Quantity Sold per Product by Sales Organization

Figure 1.2: Net Value per Material Description for Round 1

Figure 1.3 shows the top five product markets in terms of net value for round five. Even in round five, our team was unable to identify the top performing markets and adjust our product mix accordingly. By round five, we moved into production for 1kg Strawberry and were able to utilize the market well for that product. However, 1kg Blueberry, 500g Mixed Fruit, and 500g Raisin were the other top three performing products for round five and we were unable to produce and sell those specific products. Out of the top five products of that round, we only produced and sold two of them: 500g Nut Muesli and

6

1kg Strawberry Muesli. Identifying which markets to enter was a problem we consistently encountered throughout the nine rounds.

Figure 1.3: Net Value per Product by Sales Organization for Top 5 Markets in Round 5

Below, Figure 1.4 shows the sale of 500g Blueberry Muesli over the nine rounds and compares our company performance with our competitors and the market. Figure 1.4 corresponds with Figure 1.5, which shows the difference in overall market sales revenue over the nine rounds. In Figure 1.4, we can see that we entered the 500g Blueberry Muesli market in rounds four, five, and six. However, in all of those rounds, the market net value dropped considerably compared to previous rounds. We sold the most 500g Blueberry Muesli in rounds four and five, but that was when market demand for 500g Blueberry was relatively low.

After noticing this decline, we made the decision to halt our production of 500g Blueberry. We then committed to another product in our sales forecast. Unfortunately, we sold the last of our 500g Blueberry Muesli in round six, exiting the blueberry market right as 500g Blueberry Muesli sales began to spike again. At this time, it was too late to switch back into this product without first producing the other products we had committed to.

7

Figure 1.4: 500g Blueberry Sales Over Nine Rounds

Figure 1.5: 500g Blueberry Muesli Net Value Changes Over Nine Rounds

8

Reasoning for Our Strategic Problem

1. Timing of Market Entrance and Exit

One of the possible reasons for our inability to take advantage of booming markets was that we were too slow to enter into markets that we saw as up-and-coming. This issue would lead to our inability to take advantage of market booms and caused us to be stuck with large amounts of relatively unpopular stock.

Figure 2.1 shows the percentage change in market demand of 1kg Strawberry from round to round. There is no bar for round one because there is no round zero sales to compare it to, but the total amount sold in round one was $1,234,539 [$5.09*240,000] (this calculation was derived from our pivot table of sales data from SAP database). The market for 1kg Strawberry increased steadily from round one to round four. It then began to steadily decrease (with a couple of spikes in between) for the rest of the game. The large increase at the end in round nine can be attributed to our dumping of inventory at low prices instead of an actual increase in market demand.

Figure 2.1: Percentage Change in Market Demand for 1kg Strawberry from Previous Round

Figure 2.2 illustrates how our production of 1kg Strawberry begins in round 3 and steadily increases until the end of the game. Rather than lower our production of 1kg Strawberry as demand decreased, we actually increased our production.

9

Figure 2.2: Production of 1kg Strawberry by Round

It is clear based on the above charts that our team correctly read the market in order to produce 1kg Strawberry in time for the market spike in round four. However, we failed to continue watching market trends in order to know when to stop producing 1kg Strawberry. We increased our production of 1kg Strawberry from round four to round nine, even though the market report shows that the product line started declining after round four.

2. Placing Large Material Requisition Orders

Another aspect of the strategic problem we identified was that we placed large material requisition orders. Our goal of placing large material requisition orders was to maximize production immediately after the goods were received. We did not want low productivity or idle time in between orders, so we wanted to load up on materials whenever we ordered. However, looking back, this might not have been the best strategy.

Figure 2.3 illustrates the material requisition orders received on certain days from rounds one to five. Looking at day 11 of round three in particular, we received a quantity of 861,000 units of material for production. Figure 2.4 corresponds to the spike of 861,000 units shown in Figure 2.3 and shows the breakdown of the 861,000 units in detail. We intended to make three main products of Strawberry, Raisin, and Nut Muesli.

Purchasing such large quantities of materials for only a few select products meant that we did not have to worry about low productivity, but also meant that we did not have the flexibility to change between products efficiently. This was a huge problem since we couldn’t take advantage of the markets we perceived as successful. One such market that we identified was 1kg Blueberry. Although we saw the demand increasing, we were

10

not able to switch into this product with such a large requisition order. By having material for only a few products, we were limited in producing only those specific products and thus limited what products we could sell and the markets we could enter into.

Figure 2.3: Quantity of Goods Receipt Rounds 1-5

Figure 2.4: Detailed View of Day 11 and Breakdown of 861,000 Units Shown in Figure 4.4

11

We included Table 2.1 to again illustrate how production was locked in for Strawberry, Raisin and Nut Muesli late in round three. By the time we finished producing these items, it was already late in round four and we felt like we missed the opportunity to take advantage of the 1kg Blueberry Muesli market.

Table 2.1: Product Delivery

Recommendations

In addressing the two potential causes of our strategic problem, we have determined two main recommendations that could resolve our strategic issue. As we described above, part of our strategic issue was entering markets right as the demand spiked and exiting markets after demand already fell unfavorably. Though markets are volatile and sometimes unpredictable, there are times when we should have known to enter or exit a market based on extensive market analysis.

Our recommendation going forward is to not only follow the successful markets after the ZMARKET data is released every five days but to consider evaluating where we might be able to gain an edge in niche markets as well. Since we wait a full five days for the ZMARKET report data, there could be a lot of market changes in the meantime that we are not accounting for. At that point, when the ZMARKET report comes out it might be too late to switch our production to meet the market current market needs.

We will utilize Lumira and SAP Cloud during the game itself to predict the patterns in the market. Initially, each of our team members was responsible for a relatively simple task within one given department. Now that we know how to properly carry out our assigned tasks, we can designate more team members to do actual analysis throughout the duration of the game. Our team will be able to make important executive decisions based on predictive analysis and real data rather than guessing which markets might be outperforming others and whether we should switch production lines during the five-day downtime.

We will use a mixture of several charts and graphs that we believe will highlight the market trends. The first chart will show the products sold by round and by sales organization. This will show us where the large markets are, where our competitors are selling the most, and where we have a good market share. In addition, we will be analyzing individual products to determine whether the demand is increasing, beginning

12

to level out, or decreasing. Partnering these two charts together should allow us to avoid the issues we had by waiting for the ZMARKET report, while also giving us additional insight into our own market power.

In addition, we mentioned that another cause to our strategic issue of improper timing of entering and leaving certain markets was due to our large production runs. We would often look to purchase way too many products at one time and would have production scheduled for over 15 days at a time. We were so blinded by our productivity percentage that we failed to efficiently switch quickly between products.

In order to counteract this, our second recommendation is to have smaller and more frequent production runs. With our investment in setup-time reduction, we are able to maintain high productivity while also having the flexibility to switch into emerging products. After creating a budget production schedule with smaller batch sizes and strategically timed purchase order conversions, we can still operate around 90% productivity. We just need to communicate more efficiently to ensure we are running MRP and creating purchase orders early enough to ensure we don’t run into any slack time. We believe our lower production runs will allow us to take advantage of market trends in order to offer a wider and more coveted product mix.

In short, our two recommendations are 1) using Lumira and SAP CloudAnalytics to create graphs in game and 2) having shorter production runs. These two potential solutions should help us enter and exit markets smarter and quicker.

13

Operational Problem

The main operational problem we faced during the Advanced Game was that we were often running out of products too quickly which left us with little finished goods inventory to sustain us through the next production run. As mentioned in the section addressing our strategic issues, our initial strategy was to produce only a few products in order to maximize production and reduce switching time between products. For the first round, we produced 500g Nut Muesli, 500g Raisin Muesli, 1kg Nut Muesli, and 1kg Original Muesli. Out of the twelve available products across two different sizes for distribution, we only produced four. This meant that we had to carefully control the level of finished goods inventory on hand to ensure we would have enough to sell while other products were still in production. However, we struggled to keep a sufficient level of inventory, with stock-outs leading to gaps in sales negatively affecting our bottom line.

Graphs to Illustrate the Operational Problem:

Figure 3.1 shows the production and sales for 500g Nut Muesli in round one. The blue bars represent production quantity and the green bars represent sales quantity. The chart is designed to show our operational problem; our products sell out too quickly. Ideal pricing and production result in sales spread over the period of time between one production run and the next. According to the graph, we lost out of five days worth of sales (day 13-18) because we sold out too quickly.

Figure 3.1: Production and Sales for 500g Nut Muesli for Round 1

Figure 3.2 shows the sales and production runs for 500g Nut Muesli in rounds one through nine. The blue data points correspond to our production runs and the green data points correspond to sales taking place. Any breaks in the green line represent times when we sold out of the product too early and lost out on potential profits. According to the chart, there was only one production run that actually lasted into the next run.

14

Figure 3.2: Production and Sales for 500g Nut Muesli Rounds 1-9

Reasoning for Our Operational Problem

1. Low Prices Leading to High Demand Resulted in Depleted Inventory

One of the potential causes for our inventory being depleted was because our team priced our products too low with the goal of being competitive in prices, especially in distribution channels 10 and 12 where the prices are very elastic. However, our prices being set too low may have resulted in a substantial increase in demand. This resulted in our inventory being sold way too quickly, leaving little in the warehouses before we finished another production run for that specific product.

We created a heat map (Figure 4.1) that visualizes our prices in each distribution channel comparatively with our competition over the nine rounds. This graph is a simple representation that shows how our prices were set much lower than the average competitor's prices for distribution channel 10. For distribution channel 12, our prices were set lower than the average competition’s prices for the first four rounds before we started pricing higher than the average. For distribution channel 14, we priced our products lower than our competition’s average prices due to a system malfunction that locked our price in place for three rounds. Because distribution channel 14 is comprised of independent grocers who are highly susceptible to marketing and are less elastic to changes in prices, we did not fully utilize our price leverage with this particular product distribution channel. It makes sense that we priced a little lower for distribution channel 12 and 10 because they are more price-sensitive merchandisers.

15

Figure 4.1: Average Price per Distribution Channel; Company vs Competitors

Figures 4.2-4.3 illustrate the various price changes for 500g Nut Muesli in distribution channels 12 and 14 in comparison with the market’s prices for all nine rounds, while Figure 4.4-4.5 show the same dimensions for 1k Strawberry Muesli. Looking at the 500g Nut Muesli in DC 14, the prices our team set for this product fell below the market price between rounds three and six due to the system malfunction. Since DC14 is the least price sensitive, we were not taking advantage of the market prices to maximize profits. In contrast, the prices set for 500g Nut Muesli in DC12 were relatively even.

1kg Strawberry, on the other hand, was below market price for the majority of all nine rounds for both DC10 and DC12. Our goal from this was to price competitively in these markets since DC10 and DC12 are both relatively price elastic and not susceptible to the influence of marketing. We thought that by pricing low, we would be able to sell more and thus gain more revenue. However, these lower-than-market prices may explain the high demand for our product and thus the reason for why our finished goods inventory was depleted very quickly, supporting our overall problem that we did not have enough inventory to last us through another round of production.

An additional issue we noticed was that since the market report comes out every five days, our prices tended to lag behind the market. This is displayed best in Figure 4.2 from rounds three to six. Our company’s price is in green and our price lags behind the market average by one round. Given this approach, we are not optimally pricing to what the market demands but merely just following what we know the market bore in the past. Our pricing strategy was to price around 5 percent over the market price in DC12, but this graph shows that we were unsuccessful in doing so.

16

Figure 4.2 Figure 4.3

Figure 4.4 Figure 4.5

After looking at our prices for 500g Nut Muesli for DC 14, we wanted to dive deeper into the effects of pricing that low. We noticed that our prices for rounds three to five were well below the market rate and as a result, we felt that this would negatively affect our net income. As shown in Figure 4.6, we first made a graph depicting the net value of the market and our company by distribution channel in rounds three through five. However, this graph was not obvious enough to the naked eye to discern anything. We then took the raw data from the graph and manipulated it in Excel to see our proportion of sales to each distribution channel. This analysis can be found in Table 4.1. Our analysis shows our company did not earn as much revenue from DC10 compared to the market (-6%) and earned too much of our revenue from DC14 (+9%). If our company priced correctly, this distribution would be favorable; but since our prices in DC12 were higher, this was not a favorable outcome.

17

Figures 4.6: Net Value per Distribution Channel for the Market and Company

Table 4.1: Table of Net Value per Distribution Channel

From this graph and table, we were able to find that because of our low prices in DC14, we were selling a disproportionate amount to that channel. Our company was selling 9% more in DC14 than the overall market which means that we were receiving lower margins on those products than if our sales were consistent with the market proportions.

In order to tie back into our hypothesis that we had depleted inventory, we decided to look into whether or not our 500g products were being sold too quickly as a result of our

18

low prices. As seen in Figures 4.7-4.9, we looked at the 500g products that we produced in rounds three to five. Like on the graph seen above, the blue lines represent the production runs for each product and the green lines represent sales. Anytime there is a break in the green line, it represents a stockout and thus a loss of sales for that period of time.

Figure 4.7: Production and Sales for 500g Nut Muesli Rounds 3-5

Figure 4.8: Production and Sales for 500g Blueberry Muesli Rounds 3- 5

19

Figure 4.9: Production and Sales for 500g Raisin Muesli Rounds 3-5

As the charts depict, we consistently underpriced our products which lead to constant sellouts and lost revenue that could have come from increased profit margin if we had priced higher. There are large variations in stockout time between production runs from three days in Figure 4.8 to 14 days in Figure 4.9. With this high variation, there is no easy change we can make to prices in order to fix all of our stockouts. Instead, we need to be more attentive to prices and adjust them in real time according to historical sales data.

2. Not shipping enough products to our regional warehouses to keep up with production

Another cause of our operational problem was that we were not shipping enough products to the regional warehouses to keep up with production. By not shipping enough of our products, we were not maximizing the amount that could be sold, thus leading to the operational problem of running out of products too quickly in the regional warehouses. If we managed to ship out all of our products to the regional warehouses, then the warehouses would be more adequately stocked to handle the outflow of product.

Figure 4.10 emphasizes our company’s total production in comparison to the total quantity of goods shipped out to the regional warehouses. Both this graph and Figure 4.11 were created by linking the yield and goods movement databases in order to make the comparison. The only difference between the two graphs is that Figure 4.11 shows production vs quantity shipped over all nine rounds. For round one in particular, we had produced a lot more than was shipped out. This also happened in round eight, where we under-shipped our total production by 28,854 units [468,000-439,146]. This was especially detrimental to our inventory levels because we were selling out especially fast in round eight, where we needed a sufficient level of inventory to support the demand.

20

Figure 4.10: Total Quantity Produced and Total Quantity Shipped to Regional Warehouses in Round 1

Figure 4.11: Quantity Produced and Quantity Shipped to Regional Warehouses Over Nine Rounds

21

Table 4.2 calculates the quantity that was over-shipped or under-shipped for each of the nine rounds. The column furthest to the right shows the cumulative shipments not made over the nine rounds. Round one had the highest percentage of goods not shipped because the 88,487 units left in the warehouse made up 21.22% of the total quantity produced [88,487÷417,000=0.2122]. We had four rounds where we under-shipped the goods we produced. Rounds four, seven, and eight were the other rounds that we failed to ship all of our goods with 3% [64,048÷464,000], 0.02% [48,000÷461,000], 6.17% [76,854÷468,000] of units not being shipped out.

As a result of the first round having such a high quantity of produced goods not shipped out, we constantly had to keep up throughout the next nine rounds to make up for the 21.22% under-shipped goods. As you can see in the column furthest to the right, we were unable to make up for it until the last round. Though these numbers are low, they are very inconsistent and fluctuate a lot. The inconsistency of remaining units in the main warehouse causes uncertainty regarding the units we keep in each of the regional warehouses. For this reason, we believe that it could be a potential cause of our inventory levels being so low.

Table 4.2: Over/Under Shipments to Regional Warehouses

22

Recommendations

As illustrated above, the reasons for our depletion of inventory was due to a mixture of low prices and not shipping enough products to our regional warehouses. After running our analysis on low prices, we have determined the best course of action is to actively monitor pricing trends and be more proactive on switching prices every five days. As a company, we need to get better at monitoring pricing trends and making educated guesses on where the prices will be. In the past, we were pricing to what the average prices were and as a result, we could not follow our pricing strategy. Noticing that prices are on an upward or downward trend will help our company take advantage of high demand environments and hedge our losses when demand is soft.

Secondly, we will switch prices every five days in order to stay competitive within the markets and make sure we are executing the pricing strategy we have set for ourselves. Since the market reports come out every five days, we will get immediate feedback if our prediction on the market is holding true or not. Our company would rather take a bolder and riskier stance on the market than always be five days behind our competitors.

We will rectify our second issue, of not shipping enough to regional warehouses, by having better communication between our product, sales, and shipping segments of our team and better analysis of factors that affect shipping. In order to have better communication, our team plans on talking more with each other on product sales and products planned for production in order to get our shipping right. With better communication, we are able to quickly adapt to certain areas buying different products, and not lose out on sales because products are still in our main warehouse.

Lastly, we plan on having a better analysis of the factors that affect shipping. These factors include inventory reports, sales reports, and our production schedule. We plan on shifting some positions on our team so that one person is dedicated to analyzing shipping. While communication between roles can lead to the facilitation of better analysis, having one person whose specialty is analyzing the inventory and sales of regions will greatly benefit our team.

In conclusion, we determined that we had low inventory because of our imperfect pricing and shipping schedule. We believe the pricing hindered our ability to make the margins we expected to achieve and the shipping did not allow us to sell enough. Going forward we plan on having more analysis of the data by shifting around the roles of our team and better communication of our strategy during game time. These two solutions executed properly should fix our inventory problem and help the team earn the number one company valuation.

23

Conclusion

While our team performed quite well overall in the extended game, we definitely had some areas in which we can improve in going forward. Our initial strategy was to only produce around two product lines in order to maximize production capacity and minimize production costs. Then around round three, we saw how different product lines were experiencing breakout sales and so we decided to change our product mix and move into those markets. We also had trouble finding optimal prices that moved our products quickly enough to conserve warehousing space but slowly enough to prevent stockouts.

While taking advantage of high performing markets is a good strategy, our execution was poor as we identified and moved into some markets too late, while failing to exit other markets when we should have identified a downturn. While we achieved our goal of having high productivity, we failed to adapt quickly enough to change our product mix to take advantage of booming markets.

Going into the next game, we have a revised strategy that combines the high productivity from the last game with more efficient market analysis allowing us to move into new markets more concisely. Additionally, we plan to run production in smaller batches in order to both save on warehousing costs and to be able to switch between product lines more efficiently when the market demand shifts. This revised strategy will allow us to maintain high productivity while being able to better react to market changes; which will in turn translate to higher profitability across the board.

After determining our problems our recommendations are as follows: • Strategic

1. Utilize Lumira and SAP CloudAnalytics to create graphs of market trends 2. Have shorter production runs

• Operational 1. Monitor prices better 2. Proactively switch prices every 5 days 3. Better communication between sales, production, and shipping 4. More analysis of factors that affect shipping

Our recommendations are a combination of better communication, analysis, and proactive action. With our grasp of the game, we will be able to consolidate roles on our team so we can free up one more person for full-time analysis of the market. The successful implementation of these strategies will lead to better performance in the final game and a number one overall class rank.