Question

The data in the table below is from a study conducted by an insurance company to determine the effect of changing the process by which insurance claims are approved. The goal was to improve policyholder satisfaction by speeding up the process and eliminating some non-value-added approval steps in the process. The response measured was the average time required to approve and mail all claims initiated in a week. The new procedure was tested for 12 weeks, and the results were compared to the process performance for the 12 weeks prior to instituting the change.

Table: Insurance Claim Approval Times (days)

Old Process	Elapsed Time	New Process	Elapsed Time
Week		Week
1	31.7	13	24
2	27	14	25.8
3	33.8	15	31
4	30	16	23.5
5	32.5	17	28.5
6	33.5	18	25.6
7	38.2	19	28.7
8	37.5	20	27.4
9	29	21	28.5
10	31.5	22	25.2
11	38.6	23	24.5
12	39.3	24	23.5

Use the date in the table above and answer the following questions in the space provided below:

1. What was the average effect of the process change? Did the process average increase or decrease and by how much?
2. Analyze the data using the regression model y = b0 + b1 x, where y = time to approve and mail a claim (weekly average), x = 0 for the old process, and x = 1 for the new process.
3. How does this model measure the effect of the process change?
4. How much did the process performance change on the average? (Hint: Compare the values of b1 and the average of new process performance minus the average of the performance of the old process.)

Answer 1

(a) The mean elapsed time is found for old process is 33.55 and for new process is 26.35 and the difference in mean are found statistically significant as pvalue is less than 0.05 i.e. 0.000 in table1. The elapsed time is decreasing by the new process and the difference is 7.15

(b) The regreaaion equation is found to predict the elapsed time for old and new process is

Elapsed time = 33.55 - 7.2 * type of process

If the process is new then type of process = 1 otherwise 0

The effect of the process is significantly impact on the elapsed time as the pvalue is less than 0.05 i.e. 0.000

The process performance change on the average is - 7.2 which is the difference in mean of old and new process elapsed time.