My question is for question 6 how to calculate the "Adjusted MapQuest times" for the sample of 500 deliveries?

5.A driver reads your answer to part (4) and makes the following comment:

MapQuest gives driving times for cars. Driving times for trucks will probably be longer because we can't go as fast, especially in city traffic. Is there some way to adjust the MapQuest prediction to make it fit trucks better?

Construct a fitted line plot (scatterplot with fitted linear regression line) using the MapQuest times as the predictor variable and the actual times as the response variable. Call the regression values 'Adjusted MapQuest' times. Provide the fitted line plot, regression equation and R² values associated with the 'Adjusted MapQuest' times. Interpret your results and indicate whether the driver's suggestion is valid or invalid.

The plot is given below,

From the above plot we can see that the regression equation is,

Actual times = 9.3392+0.8788*MapQuest

The R² value is 0.7937 implying very accurate model.

Now the regression model adjusts for both the smaller and large values and hence the about regression model would be useful. There are lots of value below the regression line implying the drivers claim is correct as well.

6.Use the regression equation given in part (5) to provide 'Adjusted MapQuest' times for the sample of 500 deliveries and calculate the Adjusted Deviations = Actual times - Adjusted MapQuest times. (You do NOT need to put those values here.) Construct an appropriate graph for comparing the deviations associated with the Old Computer Model, the MapQuest model and Adjusted MapQuest model and include it here. Interpret the graph and recommend the modeling tool (method of predicting delivery times) you feel is most accurate.

The histograms would be most appropriate. The obtained histograms are given below,

In part (4) we observed that the deviations for Mapquest was better than the old model. Here we can see that the deviation for Adjusted Map Quest is even better. Though the range may be almost same for adjusted and unadjusted Map Quest deviations but as the number of values close to 0 is higher for Adjusted model so the standard deviation and variation is smaller for Adjusted Map Quest model deviations.

Thus the most accurate model tool is the adjusted Map Quest model.