In the Jet Copies Case Study, we are introduced to three college students: James, Ernie, and Terri. They have discovered that they have a possible solution to two separate problems that they face: the need to make money and the lack of adequate copy services near the university that they attend. The three students decide to open their own copy service but soon discovered that they may have acted too soon in their decision.
The first obstacle they had was the initial purchase of equipment. Without doing enough research, they purchased a copier for $18,000 that was obtained as a loan from Terri’s parents. After the purchase, they discovered that the particular brand they purchased had a history of frequently breakdowns. Although the copier manufacturer promised prompt repair service, the machine down time could range from one to four days.
In order to hopefully alleviate too much lost revenue for the beginning company, the trio wanted to purchase a backup machine to use during the primary machine breakdowns. The cost of the second machine would be $8,000. Before they approached their parents for another loan, they wanted to determine if it would actually be a profitable move or an unnecessary purchase on a whim.
To do this, they set up a simulation to determine how often the main copier could break down within a fifty-two week period. They decided that if the simulation showed that they lost more than $12,000 in revenue with just the one copier they would go ahead with the second purchase. They also estimated that they would produce between 2000 and 8000 copies per day at a fee of $0.10 per copy. They decided to estimate the possible daily copies by using a uniform probability distribution. They also determined that the time between breakdowns could be 0 to 6 weeks. Finally, they developed the following probability table for repair times: Repair Time (days) | Probability | 1 | 0.20 | 2 | 0.45 | 3 | 0.25 | 4 | 0.10 |

Upon completing the simulation, they were able to determine the following information regarding the days – to – repair ratio. In the beginning of the year, the ratio trended to the longer time of three days of downtime. In the middle of the year, the trend averaged to one or two days at a time. However at the end of the year, it trended back up to the three days of repair time. This was computed by using a range of pseudorandom numbers between 0 and 1. Once the range was completed, the simulation used the lookup function to identify which repair day correlated with the specific range. The average breakdown repair time for the entire year was 2.25 days.
The next ratio that was compared was the interval between breakdowns. In the beginning of the year, the breakdown intervals trended on the high side of four to five days between breakdowns. During the middle of the year, the intervals trended on the low side of one to two days. However, at the end of the year, it trended back up to the high side of four to five days. This was computed by using the formula of 6*sqrt of r, where r was the pseudorandom number computed.

The next ratio that was compared was the ratio of lost revenue. In the beginning of the year, the lost revenue associated with Day 1 and Day 2 trended to the high side while Day 3 and Day 4 trended to the lower side. In the middle of the year, the lost revenue seemed to trend to the high side on all for days of lost revenue. By the end of the year, the lost revenue seemed to be trending higher on Day 1 and Day 3 and lower on Day 2 and Day 4. This was computed by using the formula of 6r+2 to compute the number of copies lost in the thousands and then multiplying the answer by 100 to obtain the revenue that was lost.
The final step in the simulation was to determine the total revenue that was lost during the four day period of each week. In the beginning of the year, the revenue loss was trending fairly low ranging from 1300 to 2000 in lost revenue per week. In the middle of the year, the revenue loss was trending higher ranging from 1700 to 2200 in lost revenue per week. By the end of the year, the revenue loss was trending even higher ranging from 1900 to 2400 in lost revenue per week. The average of weekly lost revenue for the entire year was 1906.52.
Based on the confidence limits of Upper Confidence (1977.19 of lost revenue per week) and Lower Confidence (1835.85 of lost revenue per week), I can safely say that the simulation does show that there is a definite need for the purchase of a second copier because the potential lost revenue for the entire 52 weeks is definitely more than the 12,000 cap set in the beginning of the case study. The minimum revenue loss would be 95,464.20 (1835.85 X 52) while the maximum revenue loss would be 102,813.90 (1977.19 x 52).