Taz Lake, MBA, PMP, Atlanta-based Technical Project Manager specializing in Web and Internet – IA, BA, KMS, DMS, CMS, CRM

You probably use multi-criteria project selection every day, but you probably don’t call it that on the consumer side of your life.  Whenever you make a purchasing or investment decision you are using some variant of the tool.  Whether you are purchasing a new TV or deciding which route to take to work, you are thinking about many variables which inform your decision.  You break the decision into categories and then might have specific criteria under those categories.  Then you compare and score and make a decision.  By the way, making a decision not to purchase or proceed is also a decision.  All of these actions tend to be intuitive for many and everyone approaches it differently and may emphasize certain criteria more than others.

This happens in business too, but the hope is that the selection process is rigorous and justifiable.  One mechanism to help with this is a multi-criteria model.  The multi-criteria model I use below is adapted from Information Systems Project Management: A Process and Team Approach by Mark Fuller, Joe Valacich, and Joey George.  This multi-criteria analysis can be used to weigh project alternatives and provide a mathematical model for selection of projects or products.  The number of requirements and constraints is not important, but it is important to keep the decision factors narrowed to those the organization views as most important. 

Rules:

• The total weight for requirements and constraints should equal 100. 
• You may bias the model toward one or another of the items, but for in general start by balancing the requirements and constraints with an equal weight at 50 points total. 
• The authors of the textbook use a 1,2,3,4,5 scoring model.  I prefer a 1,3,9 scoring model because it forces tough decisions and creates separation in the model. 
•  The individual weights assigned to individual requirements and constraints are a matter of negotiation among the project selection team.  In this model, larger numbers indicate the project is better along that line item.
• The scoring and summary columns are calculated fields and should not be modified.

Example:

The organization has determined that X system is needed.  We have been asked to evaluate three platforms using the organization’s multi-criteria analysis.  Step 1, requirements and constraints have been set based on negotiation and review of the organization’s needs.  Step 2, weights for the criteria are applied based on negotiated values.  Step 3, alternatives are selected from a “consideration set” and ratings are applied, giving us a final total.  The larger number wins.

Step 1: Multi-Criteria Analysis

Step 2: Multi-Criteria Analysis

Step 3: Multi-Criteria Analysis

Note: This example is not prescriptive… criteria, weights and ratings are unique to each selection process, as well as each organization.