by Tom Cloyd - 4 min. read - (reviewed 2024-05-02)
[under development]
Choosing the best option from a set of possible problem solutions is often poorly done, yet there is a fairly easy way to do much better. Then, when the best option is acted upon, many people think they are finished. However, there is one additional step which can seriously improve your situation: assessment of outcomes. With outcomes assessment you have what you need to start a feedback loop in which you can reassess all that you done, from problem description to solution locations and then to taking action.
Special note: This part of the problem-solving article is very much still incomplete.
Most problems involve more than one issue. Let’s take the common problem of needing transportation to work. And let’s say that you’ve already decided you need a motorbike. Which one is your best choice?
Here’s an example of how this could work out. Let’s say and office-worker, Ann, is looking to solve her work transportation problem in this way. Ann is poor, and she needs to make a good decision. If her chosen motorbike breaks down or costs too much to repair, she will be in real trouble.
In trying to find her best motorbike purchase option, she has several concerns:
With expert assistance (consulting some mechanics), and some Internet research, information on each of these 4 factors can be reduced to a number that reflects actual cost. But if one is considering, say, 6 different motorbikes, there are now 6 times 4 (=24) data points to keep in mind. This is far beyond the ability of ordinary humans!
No one can do this without error, but our problem-solving process uses a decision tool than can prevent this. This matters, because what usually happens in this situation is that people just ignore information so that they can move on to making a decision. This does NOT lead to a good decision!
In the motorbike example above, each of Ann’s concerns about her purchase - purchase cost, reliability, operating cost, operating cost, security issues, and insurance cost - is a factor she wants to consider. Many people call these factors “attributes” - another word for the same thing. Keeping all these attributes involved in making a decision, when there are multiple problem solutions, is handled neatly by what is call “Multi-attribute Decision Making” - or, simply, the MAD model.
How does it do it? There is no magic. The crucial information is simply arranged in a data table, and ordinary arithmetic is used to reveal the best choice for a problem or decision.
The numbers that go in the table are measures of the various factors. These numbers may seem almost impossible to get, but there is a simple solution which usually works well.
Consider the following real problems in the motorbike example:
Purchase cost can be measured in units of currency - dollars or shillings or whatever, but how about reliability? There is no obvious unit of measure for this.
Purchase cost could easily be stated in thousands of currency units, but operating costs are more likely to be in hundreds or even tens of units. A purchase cost difference of 500 units could completely wipe out an operating cost difference of 50 units.
Security issues, like reliability, have no clear unit of measure. How are we to get a number comparing the security risk of various motor bikes?
Practical experience using multi-attribute decision making shows us these problems are much more easily solved than one might think. Working through our imaginary motorbike example will illustrate how it’s done.
Let’s reduce the number of motorbikes to consider to 3, to keep our example as brief as possible. Here is a data table for these 3 motorbikes, with each of them measured on the 5 key attributes Ann is concerned about. Following the table is an explanation of how the numbers were arrived at. As you will see, it is not difficult.
(to be finished)
(to be finished)
(to be finished) ### Considering alternate views ^
(to be finished)
(to be finished)
(to be finished)
☀ ☀ ☀