Sunday, October 16, 2016

Prioritization, prioritization matrix, and weighting

Assume that you want to buy a car and assume that there is a website that has search filters for your required specifications. Assume that you want a car with specific values for the following features:
[1] Horse power
[2] Center lock
[3] Fuel consumption
[4] Front speakers
[5] Maximum speed
[6] Driver airbag
The normal situation is that the search engine will treat all the parameters as they have the same importance (priority) to you and this may lead to results that are not logic. So, what if the search is not based on true or false, what if it is based on a total score, that meets your requirements.
How this will go?
First of all, the website will list all the features (attributes) the cars have, then you have to select the important features you are interested in to look for. The next step will be a survey to sort your priorities like the following image sequence below:

Assuming number of features is M, then the number of survey questions  N depends on the number of features M and it is the sum of the arithmetic series:

Or it can be simply calculated using this simple formula:
The survey may be a little bit long, but it really sets the right priorities in a scientific mistake-proof approach
At any time during the survey you can exit the survey and save your priorities which will be saved in what I call the "prioritization matrix" which is a square matrix that defines the relationship between each feature and looks like the following:

In the previous matrix, we count the number of "TRUE" in each row. The higher "TRUE" count, the higher priority.

Based on the previous matrix, these features are sorted in descending order based on my needs like that:
[1] Driver airbag
[2] Fuel consumption
[3] Maximum speed
[4] Horse power
[5] Center lock
[6] Front speakers

This way of prioritization can help setting weight factors for project management, risk management, time management, budget allocation, ... etc.
In this next post I will explain how set the weights and generate the score.