Analogies are important for the human aspect of decision-making. We have limited cognitive capability, so it is beneficial to describe a problem in a similar, but more-familiar context to help people build mental models to help them decide.

Dijkstra was a crazy smart guy, but I would have to disagree with this point. Analogies are useful tools for aiding understanding.