decision analysis
TRANSCRIPT
CRYSTAL BALL ASSIGNMENT
ITM 501 – Simulation Report
CRYSTAL BALL ASSIGNMENT
Stephen Uthayakumaran
Kiana Arashi:
Syed Huda
Mariam El Badan
Summary
ABC is a company that offers affordable prices with
outstanding service, and that’s what unique about it. They have
always had effective customer service, especially with their help
desk. As a result the variety of clients started increasing and
caused the volume of business to increase. ABC’s clients were all
from North America and to help serve them better there shifts use
to take place for 6 hours according to an Eastern Time zone
(11pm-5 am). But now as the business volume increased they
started offering 24/7 customer service, by hiring more employee’s
to work 8-hour shift. ABC’s concerned with losing their customers
due to the expansion of the business. Therefore they ask their
employee to concentrate on one customer at a time, and give them
their full attention and ignore multi-tasking. ABC now needs to
figure out the number of employee to hire in order to minimize
the company’s cost.
Objective
The Objective of this report is to find out the number
of staff that should be hired for the help desk position, in
order to minimize cost and serve customers in the best possible
way.
Assumptions
To better understand the purpose of the report and to
give better decision-making ideas, we must examine the case and
look for all the assumptions that we have. Here are some of the
assumptions we have gathered:
ABC objective is to minimize cost.
ABC offers 24/7 services.
ABC has three shifts, and each shift will have at least onestaff member at a help desk.
ABC does not allow their help desk staff to multi-task,
therefore they will use the first come, first serve base.
Their help desk staffs are usually at the same skill andknowledge level, and their shifts are on a rotating basis.
Staff’s shifts are for the 8-hours straight.
The maximum number of staff that are working at the helpdesk at a certain shift is 2.
ABC accepts requests that are either through the phone,email or real-time (text) chat over the web.
Simulation
There were a couple of ways to simulate the conditions
required for this project. We chose to use crystal ball to factor
in all the different variations of the same randomized data by
running 500 trials (to represent the 500 weeks). Observing our
results, we conclude that the best approach towards this problem
would be to have 1 staff per shift to run the ABC Company. This
will approximately maintain the least amount of delay costs (on
average) compared to any other options available which will
improve customer service greatly. We have chosen staff to be
hired at 600 per week based on our budget. We had to test
different types of staffing formations to determine an optimal
staff count. In order to utilize all the different options, we
simulated the trials in all 8 different possible settings. The
average increase on our budget, factoring in delay costs, will be
about 331.72 dollars giving us a total expense of 2131.72 per
week. In contrast with all the other options, (1,1,1) is the
least costly and as such the best logical choice.
Results for 500 trials in all
formations
(1,1,1)
Forecast: Delay Cost (Total)Statistic Forecast valuesTrials 500Mean $331.72Median $332.62Mode '---Standard Deviation $110.31
Variance $12,167.54Skewness -0.1486Kurtosis $3.93Coeff. of Variability $0.33Minimum $9.73Maximum $710.27Mean Std. Error $4.93
(1,1,1)
Cost of StaffShift 1 Shift 2 Shift 3
# ofstaff 1 1 1Price /week $600.00 $600.00 $600.00Cost /shift(Total) $600.00 $600.00 $600.00
Staff Cost (Total) $1,800.00
Overall Costs(Total) $2,131.72
(1,1,2)
Forecast: Total Delay CostStatistic Forecast valuesTrials 500Mean $331.65Median $313.14Mode '---Standard Deviation $103.73Variance $10,760.13Skewness 1.27Kurtosis $5.92Coeff. of Variability $0.31Minimum $151.35Maximum $905.96Mean Std. Error $4.64
(1,1,2)
Staff CostsShift 1 Shift 2 Shift 3
# ofstaff 1 1 2Price /week $600.00 $600.00 $600.00Cost /shift(Total) $600.00 $600.00 $1,200.00
Staff Cost (Total) $2,400.00
Overall Costs(Total)
$2,731.65
(1,2,1)
Forecast: Total Delay CostStatistic Forecast
valuesTrials 500Mean $116.97Median $111.55Mode '---Standard Deviation $35.81Variance $1,282.71Skewness $1.04Kurtosis 4.59Coeff. of Variability 0.3062Minimum $53.48Maximum $287.97Mean Std. Error $1.60
(1,2,1)
Staff CostsShift 1
Shift 2 Shift 3
# ofstaff 1 2 1Price /week $600.0
0$600.00 $600.00
Cost /shift(Total) $600.00
$1,200.00
$600.00
Staff Cost (Total)
$2,400.00
Overall Costs(Total)
$2,516.97
(2,1,1)
Forecast: Total Delay CostStatistic Forecast
valuesTrials 500Mean $267.86Median $247.64Mode '---Standard Deviation $94.14Variance $8,862.74Skewness $1.12Kurtosis 4.6Coeff. of Variability 0.3515Minimum $91.79Maximum $678.61Mean Std. Error $4.21
(2,1,1)
Staff CostsShift 1 Shift 2 Shift 3
# ofstaff 2 1 1Price /week $600.00 $600.00 $600.00Cost /shift(Total) $1,200.00 $600.00 $600.00
Staff Cost (Total) $2,400.00
Overall Costs(Total) $2,667.86
(1,2,2)
Forecast: Total Delay CostStatistic Forecast
valuesTrials 500Mean $100.68Median $95.47Mode '---Standard Deviation $34.37Variance $1,181.41Skewness $1.46Kurtosis 8.76Coeff. of Variability 0.3414Minimum $34.23Maximum $340.96Mean Std. Error $1.54
(1,2,2)
Staff CostsShift 1
Shift 2 Shift 3
# of staff 1 2 2Price /week $600.0
0$600.00 $600.00
Cost /shift(Total)
$600.00
$1,200.00
$1,200.00
Staff Cost (Total)
$3,000.00
Overall Costs(Total)
$3,100.68
(2,1,2)
Forecast: Total Delay CostStatistic Forecast
valuesTrials 500Mean $244.51Median $230.65Mode '---Standard Deviation $86.58Variance $7,495.56Skewness $1.18Kurtosis 5.42Coeff. of Variability 0.3541Minimum $65.54Maximum $692.34Mean Std. Error $3.87
(2,1,2)
Staff CostsShift 1 Shift 2 Shift 3
# ofstaff 2 1 2Price /week $600.00 $600.00 $600.00Cost /shift(Total) $1,200.00 $600.00 $1,200.00
Staff Cost (Total)
$3,000.00
Overall Costs(Total)
$3,244.51
(2,2,1)
Forecast: Total Delay CostStatistic Forecast valuesTrials 500Mean $41.24Median $39.34Mode '---Standard Deviation $13.50
Variance $182.17Skewness $0.89Kurtosis 4.15Coeff. of Variability 0.3273Minimum $12.55Maximum $100.18Mean Std. Error $0.60
(2,2,1)
Staff CostsShift 1 Shift 2 Shift 3
# ofstaff 2 2 1Price /week $600.00 $600.00 $600.00Cost /shift(Total)
$1,200.00 $1,200.00 $600.00
Staff Cost (Total)
$3,000.00
Overall Costs(Total)
$3,041.24
(2,2,2)
Forecast: Total Delay CostStatistic Forecast valuesTrials 500Mean $12.04Median $11.17Mode '---Standard Deviation $4.74Variance $22.49Skewness $0.75Kurtosis 3.46Coeff. of Variability 0.3938Minimum $1.67Maximum $28.78Mean Std. Error $0.21
(2,2,2)
Staff CostsShift 1 Shift 2 Shift 3
# ofstaff 2 2 2Price /week $600.00 $600.00 $600.00Cost /shift(Total) $1,200.00 $1,200.00 $1,200.00
Staff Cost (Total)
$3,600.00
Overall Costs(Total)
$3,612.04
Conclusion
Our simulation report objective was to determine
the number of staff that should be hired to maintain ABC’s
customer service reputation and cover for the new shifts at the
help desk. Our simulation concluded that ABC must hire at least
one person in each of the three shifts, for the business to keep
up their customer service reputation and minimize expenses. ABC
will have to maintain the same 1-1-1 structure to be most
effective in keeping the lowest costs because the additional cost
for an extra staff member is 600 dollars which does not compensate
enough from the delay costs. Despite this however, the highest
frequency of customers are going to be during the second shift
which our analysis would suggest that maybe having a better
compensation for the staff working on the second shift would give
them more motivation. Having said that, it is important that ABC
also concentrate on existing customers and offer them the same or
higher level of customer service as they use to receive, in
additional to the new customers. By using the simulation method,
and the crystal ball tool, we have learned how to deal with issues
that require large data, and how simulation makes decision-making
much easier.