You recommend using a design of experiments approach to hopefully realize a breakthrough improvement. The team brainstorms a long list of possible reasons why the wait times are so long. From that list, the team has reduced it down to 5 factors they want to include in an experiment. They do not suspect any interactions, so they conclude a resolution III will be adequate. They refer to the resolution matrix (in Analyzing Fractional Factorial Designs Lecture - see also your online textbook) and they find they can learn about the effects of those five factors in as little as 8 experiments. Construct an main effects plot for all five of the factors. We will do the first one for you as an example: Take an average of the wait time minutes (y) when "staff size" was at (-). Then, do the same thing for when the results (y) when "staff size" was at (1). When at (-1) When at (+1) 7 9 28 25 26 8 6 28 Average 16.75 17.50 The factors for the experiment were: A B C Staff size Order of treatment Method of treatment D Tracking software E Temp. in the waiting room - level 8 FIFO Iterative Product A 68 degrees + level 16 By priority All at once Product B 75 degrees Trial 1 Staff sz 1 Order 1 Method -1 Software -1 Rm temp -1 2 3 -1 1 -1 1 -1 1 1 1 1 -1 7 25 4 5 6 7 8 -1 -1 1 1 -1 -1 1 -1 -1 1 1 1 -1 1 -1 1 -1 1 -1 -1 -1 1 1 -1 1 28 26 8 28 6 Staff size Staff size + Treatment order Treatment order + Method Method + Software Software + Room temp Room temp + Actual value here Actual value here Actual value here Actual value here Actual value here Actual value here Actual value here Actual value here Actual value here Actual value here Results (Y) (patient wait time measured in minu 9
Purchase answer to see full attachment