How do you choose the best MCDM method for a problem?

1 Problem type

The first step is to identify the type of problem you are dealing with. Is it a choice problem, where you need to select one or a few alternatives from a set of options? Is it a ranking problem, where you need to order the alternatives from best to worst? Is it a sorting problem, where you need to assign the alternatives to predefined categories or classes? Or is it a description problem, where you need to describe the characteristics and preferences of the alternatives? Different MCDM methods are designed for different problem types, so you need to match your problem with the appropriate method.

2 Criteria structure

The next factor is the structure of the criteria that you use to evaluate the alternatives. Are the criteria independent, meaning that they do not affect each other, or are they dependent, meaning that they interact or influence each other? Are the criteria quantitative, meaning that they can be measured by numbers, or are they qualitative, meaning that they can be expressed by words or judgments? Are the criteria consistent, meaning that they reflect the same preferences across the alternatives, or are they inconsistent, meaning that they vary depending on the alternatives? Different MCDM methods have different assumptions and requirements for the criteria structure, so you need to choose a method that fits your criteria.

3 Decision maker involvement

The third factor is the level of involvement of the decision maker or decision makers in the MCDM process. Are you the sole decision maker, or are you working with a group of stakeholders or experts? How much information and feedback do you have or need from the decision makers? How much control do you want to have over the MCDM method and its parameters? Different MCDM methods have different degrees of flexibility and transparency for the decision maker involvement, so you need to select a method that suits your situation.

4 Data availability and quality

The fourth factor is the availability and quality of the data that you use to evaluate the alternatives based on the criteria. Do you have enough data to cover all the alternatives and criteria, or are there missing or incomplete data? How reliable and accurate are the data sources and methods? How do you handle uncertainty and ambiguity in the data? Different MCDM methods have different ways of dealing with data availability and quality issues, so you need to pick a method that can handle your data.

5 Method complexity and performance

The last factor is the complexity and performance of the MCDM method itself. How easy or difficult is it to understand, apply, and interpret the method? How much time and resources does it require to implement the method? How robust and sensitive is the method to changes in the data or the preferences? How well does the method match your expectations and goals? Different MCDM methods have different trade-offs between complexity and performance, so you need to balance the costs and benefits of the method.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?