Tools and Techniques for Business Improvement

 

Changing
Minds

 

Creating
Minds

 

Quality
Toolbook

 

Tools of
the Trade

 

Improvement
Encyclopedia

 

 

Home
Page

 

C style
(book)

 

Business
Articles

 

Thinking
Stories

 

My
Photos

   

 

Here's my
latest book!

Add/share/save
this page:

 

 

 

 

Decision Tree: Examples

The Quality Toolbook > Decision Tree > Examples

When to use it | How to understand it | Example | How to use it | Practical variations

 

<-- Previous | Next -->

Example

The management team of a manufacturer of space heaters were trying to decide whether to release a revolutionary new heater, or whether to continue testing it. Releasing it immediately would ensure good sales, as no other manufacturers had anything like it. However, this would be at an estimated 10% risk of serious problems which would result in halving the revenue. Another six months testing would reduce the chance of problems to 5%, but at an estimated 8% risk of a comparable heater being introduced by the competition. They decided to use a Decision Tree to calculate the Expected Monetary Value (EMV) of the options in order to help decide what to do.

The figure below shows the figures and the Decision Tree. The final figures indicated that in terms of EMV, the options had very similar value, as the threat of competition canceled out the gains in reliability. They followed this up by repeating the calculations with a small range of probability figures, to investigate the sensitivity of the model. This showed that a change in problem risk had the greatest effect, so they decided to continue testing.

 

 

Fig. 1. Calculating a Decision Tree

 

Other examples

  • A machine shop team use a Decision Tree to balance the possible future demand for increased productivity with methods of achieving this. They find that a 10% increase can be best met with training and improved work processes, but a further 10% increase will require new machinery.
  • The safety team in a power plant use a Decision Tree to understand the overload situation and to develop contingency measures for unacceptable hazards.
  • A doctor investigating a disease uses a Decision Tree to understand the possible effects of using imperfect tests and possible developments.

 

<-- Previous | Next -->

 

Contact —  — My page

 

 

  Syque 2002-2013

  Massive Content -- Maximum Speed

TOP