Professional Documents
Culture Documents
OP Case Analysis SWU
OP Case Analysis SWU
PROFESSIONAL STUDIES
Beverly S. Baruis
Pretty Edulan
JP Patatag
MSBA
1. SOUTHWESTERN UNIVERSITY CASE STUDY:
Brief Description:
Discussion Questions:
1. Develop a forecasting model, justifying its selection over other techniques, and project attendance
through 2011.
Group 2 selected three (3) models to develop a forecasting model in this case: (a)
naïve approach, (b) moving average, and (c) trend projections with seasonality.
A) Naïve Approach is to assume that demand in the next year/period will be equal to demand in the
most recent period.
Naïve Approach
Moving Average
C)Trend projections. According to the meaning of the trend projections that this technique fits a trend
line to a series of historical data points and then projects the line into the future for medium to long-
range forecasts.
Table C.1
Demand
Year Time period per year x² xy
21 1,191,900 91 4,373,400
= 238,998 attendees
Table C.2
= 250,530 attendees
Seasonal variations that in data are regular up-and-down movements in a time series
that relate to recurring events such as weather or holidays. Seasonality may be applied to hourly, daily,
weekly, monthly, or other recurring patterns. Similarly, understanding seasonal variations is important
for capacity planning in organizations that handle peak loads. The presence of seasonality makes
adjustments in trend-line forecasts necessary.
Table C.3
Seasonal Variations
198,649
We will be using the seasonal indices above to forecast at the five Saturday home
games in 2010.
Average
Average per Seasonal
Games 2004 2005 2006 2007 2008 2009 2010 Demand game Index
1 34,200 36,100 35,900 41,900 42,500 46,900 47,623 40,731 40,882 0.9963064
2a 39,800 40,200 46,500 46,100 48,200 50,100 54,321 46,460 40,882 1.1364415
3 38,200 39,100 43,100 43,900 44,200 45,900 51,012 43,630 40,882 1.0672178
4b 26,900 25,300 27,900 30,100 33,900 36,300 36,173 30,939 40,882 0.7567878
5 35,100 36,200 39,200 40,500 47,800 49,900 49,869 42,652 40,882 1.0432953
204,412
We will be using those seasonal indices above to forecast the attendance at the
five Saturday home games in 2011.
Based on the computation above only the trend projections showed that the
attendance from the past years games through 2011 has gone up. The naïve approach keeps the
projection the same as the previous year’s attendance. The moving average made the forecast far less
than the last years’ attendance.
On these factors Group 2 believes that the trend projection forecasting method
is the best fit for projecting in this case where attendance for Southwestern University will actually
increasing in 2010 and 2011. This method also gives leverage to the new coach to get him new stadium
that he requested.
An assumed average ticket price of $50 in 2010 and a 5% increase each year in
future prices.
Based upon the trend projection with seasonality forecasts, the revenue for both
2010 and 2011 were computed. By computing the sum of the attendees for all the five games each year
and multiply by the respective ticket price for that year we can forecast the revenue to be expected in
2010 and 2011.
Revenue:
1 47,623 49,920
2a 54,321 56,942
3 51,012 53,474
4b 36,173 37,919
5 49,869 52,275
The computation above shows the forecasted revenue in 2010 and 2011. The
revenue in 2010 is expected to $11,949,900.00 and $13,152,825.00 in 2011 respectively.
The option that this group will be taking is to build a new stadium as the
forecasts show that the stadium will be sold out somewhere in 2010 game days. It will continue to rise
and this should definitely be dealt with now to ensure there is no loss in revenue due to sold out games.
Questions:
1. What is the maximum number of applications per hour that can be handled by the present
configuration of the process?
Figure 2.1
Solution: The bottleneck in this system is in step 3 at 60-seconds per application, resulting in an hourly
system capacity of 1 hour x 60 minutes x 60 seconds/ 60-seonds per application = 60 applications/ hour.
Answer: The maximum numbers of applications per hour that can be handled by the present
configuration of the process are 60 applications/hour.
2. How many applications can be processed per hour if a second clerk is added to check for violation
(step 3)?
Figure 2.2
30”
Solution: In this case, second clerk added to step 3. The 60-seconds checking time of file for violations
and restrictions in step 3 represents the process time each station. Next, the process time of the
combined or additional clerk in step 3 is 60-seconds per two applications, or 30 seconds per application.
Therefore, step 4 becomes the bottleneck for the entire license-renewal process and the system process
time is 40-seconds. The capacity per hour equals 3,600-seconds per hour/40-seconds per application =
90 applications per hour.
Answer: The maximum number of application per hour if the second clerk is added to check for
violations and restrictions are 90 applications per hour.
3. If the second clerk could be added anywhere you choose (and not necessarily to check for violations,
as in question 2), what is the maximum number of applications the process can handle? What is the new
configuration?
Figure 2.3
15’’
Solution: If the second clerk will be added in step 2, process and record payments. The 30-seconds
processing and recording time of payments in step 2 represents the process time each station. Since we
added the second clerk in step 2 the process time with an additional clerk is 30-seconds per two
applications, or 15-seconds per application. Therefore, step 3 becomes the bottleneck for the entire
license-renewal process and the system process time is 60-seconds. So, the capacity per hour equals
3,600-seconds per hour/60-seconds per application = 60 applications/hour.
Answer: a. The maximum number of application per hour if the second clerk is added to step 2, process
and record payments are 60 applications per hour.
Table 2.2
4. How would you suggest modifying the process to accommodate 120 applications per hour? What is
the cost per application of this new configuration?
Considering a review of the jobs indicated that step 1, reviewing applications for
correctness, had to be performed before any other step could be taken. Similarly, step 6, issuing
temporary licenses, could not be performed until all the other steps were completed.
Table 2.3
Solution: We need to determine line balancing to minimize the imbalance between clerks while
meeting the required output/demand.
= 3,600/120 applications
= 30 seconds/application
The license-renewal process time in Table 2.3 shows that 6 operations are
necessary for a total operation time of 195 seconds.
= 6.5 or 7 workers
Based on the computation above that there are 30-seconds process time system
requirement and there is a total of 7 workers in this license-renewal process time. So, we need to
combined some activities and add workers on it.
The Table 2.4 below will be the suggested modifying process to accommodate
120 applications per hour.
Table 2.4
As per average time to perform on Table 2.4 the bottleneck time is 30-seconds. We combined/added
the three (3) clerks on step 1, review renewal application for correctness, process and record payments,
and conduct eye test, and 2 clerks on step 2, check file for violations and restrictions. On this the
capacity per hour equals 3,600 seconds per hour/ 30-seconds per application = 120 applications/ hour.
Table 2.5