# Decision Trees and the Delphi Procedure

In the given case study, the IT department’s head is under duty to act on the company’s inventory control system (ICS). To address the issues connected with the ICS effectively, the head can opt for either of two alternatives. He or she should base the choosing of the alternatives on a decision tree along with a Delphi procedure. Notably, decision trees were originally used as graphical tools for illustrating the structural connections between different choices. Originally, they were presented as sequences of choices that were dichotomous (Hanke & Reitsch, 1992; Modis, 1992). Even then, they are now used in presenting more and more intricate choices as the existing appreciation of feedback loops increases. Indeed, the intricate natures of decision trees along with a Delphi procedures have seem them grow into the elementary basis of particular computer flowcharts.

The following is a tabulation of the factual data provided in the case study.

 Alternative Outcomes Productivity increase (\$) Probability (%) Total cost (\$) 1 Great benefit 15,000 40 10,000 Some benefit 10,000 50 10,000 No benefit 0 0 10,000 2 Great benefit 8,000 40 4,000 Some benefit 4,000 30 4,000 No benefit 0 0 4,000

To get the expected, or most likely, productivity increase, the stated productivity increase should be multiplied with the probability of its being released. That means under Alternative 1, the most likely “great benefit” productivity increase is:

15,000 * 40/100 = \$6,000

Under Alternative 1, the most likely “some benefit” productivity increase is:

10,000 * 50/100 = \$5,000

Under Alternative 1, the most likely “no benefit” productivity increase is:

0 * 0 = \$0

As well, it means that under Alternative 2, the most likely “great benefit” productivity increase is:

8,000 * 40/100 = \$3,200

Under Alternative 2, the most likely “some benefit” productivity increase is:

4,000 * 30/100 = \$1,200

Under Alternative 2, the most likely “no benefit” productivity increase is:

0 * 0 = \$0

 Alternative Outcomes Most likely productivity increase (\$) Total cost (\$) 1 Great benefit 6,000 10,000 Some benefit 5,000 10,000 No benefit 0 10,000 2 Great benefit 3,200 4,000 Some benefit 1,200 4,000 No benefit 0 4,000

To calculate the most likely net benefit owing to the adoption of any of the Alternatives, the corresponding difference between the projected total cost and the most likely productivity should be computed. That means under Alternative 1, the most likely “great benefit” net benefit increase is:

6,000 – 10,000 = -\$4,000

Under Alternative 1, the most likely “some benefit” net benefit is:

5,000 – 10,000 = -\$5,000

Under Alternative 1, the most likely “no benefit” net benefit is:

0 – 10,000 = -\$10,000

Under Alternative 2, the most likely “great benefit” net benefit is:

3,200 – 4,000 = -\$800

Under Alternative 2, the most likely “some benefit” net benefit is:

1,200 – 4,000 = -\$2,800

Under Alternative 2, the most likely “no benefit” net benefit is:

0 – 4,000 = -\$4000

The following decision tree should be used in choosing the most beneficial alternative

 Net Benefits Alternative 1 2 Great Benefit Some Benefit No Benefit Great Benefit Some Benefit No Benefit Expected net loss in benefits 4,000 5,000 10,000 800 2,800 4,000

From the tree, it is clear that the alternative that will result in the most minimum loss is the Alternative 2 “great benefit” choice. The company should purchase refresher training for enhancing employee productivity and design it to give a benefit of \$8,000.

Get Your Custom Paper From Professional Writers. 100% Plagiarism Free, No AI Generated Content and Good Grade Guarantee. We Have Experts In All Subjects.