# Delta Song Case Analysis

Â Possible cost drivers that will allow us to estimate a salary cost function for Delta are: available seat miles, number of departures, available ton miles, revenue passenger miles, and revenue ton miles. The two cost drivers we chose were revenue passenger miles and available ton miles. The salaries consist of payments to pilots, flight attendants and ticket agents. Their salaries are determined by the number of passengers and cargoes and the miles or hours flown. This is why we chose revenue passenger miles and available ton miles.

After calculation we found that the R2 of revenue passenger miles is . 1764, and the R2 of available ton miles is . 5577. We used scatter plots to show this: The available ton miles scatter plot shows a more linear relationship between the two variables. Low point (3132, 1145), high point (4029, 1514) Salary=0. 4114xavailable ton miles-143. 50 The greatest advantage about this technique is that it only uses two data so it is convenient.

The disadvantages are that the data is inefficient. This is because the data is based on cost function for only two periods, meaning it is less accurate.

Simple Regression Using simpler regression to estimate the salary cost with available ton miles as the cost driver. These are the results: Coefficients Intercept X Variable 1 -682. 643 0. 551693 Standard deviation 282. 6033 0. 79698 Salary= 0. 5517x available ton miles- 682. 63 R2=0. 5577, and the coefficients are larger than the deviations so it is valid. Regression analysis is more reliable at measuring cost behavior than other measurement methods. This is because this technique uses statistics to fit a cost function in all historical data.

The regression analysis technique is an improvement compared to the high low method. It also allows analysts to pick out the best cost driver. A disadvantage with the regression analysis method is that only one cost driver is considered, so it canâ€™t completely explain the variation of salaries. Multiple Regression We chose revenue passenger miles and available ton miles. The results we got are as follows: Coefficients Intercept X Variable 1 X Variable 2 -1144. 55 1. 051937 -72. 2955 Standard Deviation 243. 2101 0. 120829 14. 88974 Salary=-1144. 55+1. 05xavailable ton miles-72. 0xrevenue passenger miles R2=0. 5577, and strandard deviations are smaller than coefficients, so it is statistically valid. This technique is an improvement over simple regression. This is because this technique uses more cost drivers and provides results are closer to the data given. Questions 1-3 If the conditions are not met, the cost functions will be less useful. The const functions in Question 1-3 are based on the assumption that the wages per hour remain the same and there is no addition labor needed, so itâ€™s useful only under certain conditions.

Using the background of the industry and companyâ€™s circumstance into consideration, we think these are important: ? The present equilibrium between Delta and the labor unions is not interfered. However, issues concerning lowering pilotsâ€™ salaries to industry level will not be obstructed by union forces. Also, employees other than pilots will not join labor unions to require higher payments. If not, adjustments on cost structure will be weak and Delta Song will turn out to be another Delta Express. ? Reducing staffs or cutting salaries are to be formulated.

Restrictions about layoffs will directly lead to weak control over budgets, and in turn create similar problems as high salaries do. ? Lastly, the new fixed cost caused by new security directives after September 11 terrorist attacks. However, it shouldnâ€™t become a major concern for song because security costs can be expected amid the whole industry. Salary Cost for Jet Blue According to Question 1, available ton miles should be used to estimate the salary cost. However, available ton miles of 2002 Q3 are low. We drew a scatter plot to show this: In this situation, available ton miles and salaries are not linear.

The scatter plot of revenue passenger miles and salaries is as follows: Revenue passenger miles and salaries are quite linear. We use the high low method technique to estimate the salary cost with revenue passenger miles as cost driver. Low point (599. 4, 16000), high point (2016. 2,49000) Salary=23. 29xrevenue passenger miles+2038. 83 Estimating the salaries cost for Delta Song in its first year To estimate the salaries cost of Song, we used the historical salaries cost of JetBlue to predict Songâ€™s salary because Song does not have any historical data for reference.

We assumed that Song can achieve the same revenue passenger miles as JetBlue in every quarter, and Jetblueâ€™s salaries are linear with time series. To determine Songâ€™s salary, three deviations must be taken into consideration: ? Song pilotsâ€™ per hour wage are $100 more than those of JetBlue on average. We numbered each quarter in 2001 and 2002 from 1 to 8, and made simple regression between the time series and revenue air hours Quarter