Suppose a simulation model has two random variables (eg. 'arrival time' and
'service time' in a queuing problem). What is wrong, if anything, with drawing a single random number on each trial and using this single number to determine the value for both random variables?
Analysing a service contract using simulation
The IDG Medical Fund is being considered as a group insurer for the employees of a large public company. If IDG wins the contract and insures the group, the daily frequency of claims is estimated as follows:
|
Number of Claims
0
|
Relative Frequency
0.05
|
|
1
|
0.06
|
|
2
|
0.08
|
|
3
|
0.10
|
|
4
|
0.33
|
|
5
|
0.14
|
|
6
|
0.11
|
|
7
|
0.07
|
|
8
|
0.04
|
|
9
|
0.02
|
|
|
1.00
|
The probability distribution of the cost of each claim has been estimated using historical data and is as follows:
|
Average Cost per Claim
$800
|
Probability
0.30
|
|
900
|
0.24
|
|
1000
|
0.22
|
|
1100
|
0.18
|
|
1200
|
0.06
|
|
|
1.00
|
(a) What is the probability that there will be 4 claims or fewer per day?
(b) Using the following random numbers from left to right simulate a 5-day working week and calculate the expected weekly cash outflow in settlement of claims.
Random numbers: Random numbers: Number of Claims Cost per Claim
40, 71, 18, 77, 42 14, 56, 20, 76, 29