Little Field simulation Strategy and report

B6016 Managing Business

Operations

Report on Littlefield Technologies Simulation Exercise

By Group 4:

Anise Tan Qing Ye AneelGautam

Chu KarHwa, Leonard Tan Kok Wei

Ranking Reflecting on the simulation exercise, we have made both correct and incorrect decisions. Nevertheless, although we ranked 4th (Exhibit 1: OVERALL TEAM STANDING), we believe we gained a deeper understanding of queuing theory and have obtained invaluable experience from this exercise.

Rank Team 1 2 3 4 5 6 bigmoney1 techwizard Cash Balance ($) 1,346,320 1,312,368 makebigmoney 1,141,686 beaters123 donothing mas001 895,405 588,054 472,296 Exhibit 1 : OVERALL TEAM STANDING

Decisions Made A summary of the rationale behind the key decisions madewould perhaps best explain the results we achieved. .

Decision 1 Day Parameter Value 54 station 1 machine count 2

When the exercise started, we decided thatwhen the lead time hit 1 day, we would buy one station 1 machine based on our analysis that station 1 takes the longest time which is 0.221 hrs simulation time per batch. (Exhibit 2: Average time per batch of each station).

As day 7 and day 8 have 0 jobarrivals, we used day 1-6 figures to calculate the average time for each station to process 1 batch of job arrivals. Base on the average time taken to process 1 batch of job arrivals, we were able to figure out how many batches each machine can handle. number of day jobsarriving each day utilization of station 1, averaged over each day utilization of station 2, averaged over each day utilization of station 3, averaged over each day 1 2 3 4 2 2 1 0 0.433 0.272 0.398 0 0.079 0.158 0.106 0.052 0.019 0.108 0.165 0 5 2 0.33 0.079 0.06 6 1 0.332 0.158 0.013 Average process 0.221 0.079 0.046 time per batch Maximum batch 4.52 12.66 21.74 processed a day Exhibit 2: Average time per batch of each station

We did not pay much attention to the station 3 machine at this stage even though there appears to be significant variation in processing time, as the usage appears to be very low.

Decision 2 Day Parameter Value 56 station 2 machine count 2 This turned out to be an incorrect decision as we have yet to figure out “how the system works”. We bought an additional machine at station 2, even though the queue accumulation was not that significant, as we wanted to eliminate the queue to 0. (See Exhibit 4). Later on, we figured that we should look at the utilization of the machine as opposed to the queue not being 0.

Exhibit 4

We did a further projection by using regression analysis consolidating the job arrivals from day 1- 60 to figure out whether we should have bought machine 2 at station 2, and realized our mistake.(Exhibit 5: Regression Analysis of day 1 -60)

numberof jobs arriving each day108642y = 0.055x + 1.175R2 = 0.200numberof jobs arriving each day线性(numberof jobs arriving each day)600-202040 Exhibit 5: Regression Analysis of day 1 -60

Based on the formula y=0.0559x+1.1753 (y=12.66-the maximum machine 2 can handle, we can figure out x=day 205. This means we should have bought machine 2 only after job arrivals exceed 12 batches, and that will be at day 205 based on the regression analysis. We bought it too early and thus lost the opportunity to earn more interest on cash.

Decision 3: Day Parameter Value 132 station 1 machine count 3 134 station 1 machine count 4 135 station 2 scheduling rule pri4 142 station 2 scheduling rule pri2

On Day 132, we noticed that jobs arrivals had increased significantly and our lead time had already hit 5 days. Machine utilization had also reached100% and our ranking had dropped. Please see Exhibit 6. We immediately bought one more machine at station 1. Later on, we figured out we need another machine at station 1 as our lead time was already 5 days, and the queue will not be reduced without the additional machine. So on the following day, we bought another machine at Station 1. As regards Station 3, based on our former calculation, there seems to be no need to buy another machine, because a machine can process 21.74 batches a day. If that is the case, why is there a queue at station 3? We then recalculated how many batches

one machine at Station 3 can handle using the figures from day 1 to day 134, and the average time to process 1 batch for one machine at Station 3 is 0.09 hrs simulation time and the most it can handle is 11.11 batches a day. On day 135, we decided to buy another machine at Station 3.Also, we set the priority to stage 4 as now our priority is to meet the lead time.

Exhibit 6:

Decision 4 : Day Parameter Value 152 station 2 machine count 1 152 station 3 machine count 1 152 station 1 machine count 3 This is a wrong decision. The simulation briefing materials indicated that from day 150-180, the job arrivals will be fixed. We did closely monitor this and thus did not take action on day 150. The average arrivals for day 150 – 152 are around 10 batches, and concluded that day 150- 180 will be around 10 batches. In that case, two machines at Station 1, one machine at Station 2, and one machine at Station 3 should be able to handle 10 batches a day. We made a decision to sell the “extra machines” in hope of earning extra interest. Also, we set priority to stage 2 as we needed to fully utilize stage 3 when the job arrivals fluctuate.

Decision 5:

Day Parameter Value 157 station 2 scheduling rule pri4 157 station 2 scheduling rule fifo 158 station 2 scheduling rule pri4 166 station 3 machine count 2 168 station 2 scheduling rule fifo 169 station 2 scheduling rule pri2 169 station 2 scheduling rule pri4 169 station 2 machine count 2 170 station 1 machine count 4 On day 157, we realised that the lead time hit 1.26 days and also noticed that the job arrivals did not stay the same. Instead, they seemed to continually increase. As we hoped that job arrivals will start to drop “later on”, we scheduled priority to stage 4 to shorten the lead time. It seemed to workas the lead time dropped to 0.6 on day 163. However, on Day 166, the lead time shot up to 2.5 days (Exhibit 7). We quickly bought back all the machines which we sold as we still have some time till day 180 and if we do not do so, we will not be able to make a penny for around 30 days which equals 30x1000per batch x 15 batches a day =450,000, which is more than the purchase money for 3 machines.

Exhibit 7

Decision 6 : We decided not to sell any machines at the end of the exerciseas we treat the manufacturing plant as a going concern. Another reason is that the sale of the machines will not make a difference in the rankings.

Conclusion: In order to earn the interest which is small amount (65202/268=243 a day)(Exhibit 8), as compared to the daily revenue which will amount to 15000 a day with the machines in place,we mistakenly sold the machines. If we did not sell and buy back the machines, we should be able to rank NO.2 at least.

Exhibit 8: Sources and Uses of Cash

Description Starting Cash Cash Sources Revenue Interest Amount($) 1,000,000 . 1,688,202 65,202 machine retirement 30,000 Cash Uses debt interest . 0 machine purchases 720,000 Inventory Cash Balance

1,168,000 895,405

本文来源：https://www.bwwdw.com/article/hzev.html