The Big R-Book

♣27♣Multi Criteria Decision Analysis (MCDA)

27.1 What and Why

All models discussed previously use one common assumption: there is just one variable to model. While it might be complicated to find an optimal value for that variable or very complicated to predict that one variable, it does not prepare us well for the situation where there are more than one functions to optimize.

Think for example about the situation where you want to buy something – anything actually: then you certainly want to purchase it cheap but still a good quality. Unfortunately, goods of high quality will have high prices, so there is no such thing as an option that is optimal for both variables. Fortunately our brain is quite well suited to solve such problems: we are able to buy something, get married, decide what to eat, etc. The issue is that there is no clear bad choice but also no clear good choice. People tend hence to decide with gut‐feeling. This is great because it helps people to do something: fight or flight in a dangerous situation, start a business, explore a new path, get married, vote, etc.¹ The world would be very different if our brain was unable to solve such questions really efficient: such decisions come natural to us, but they are hard to explain to someone else.

However, in a corporate setting this becomes quite problematic. We will need to be able to explain why we recommend a certain solution. That is where Multi Criteria Decision Analysis (MCDA) comes in: it is merely a structured ...