Economics and thermodynamics

As I read about economics, I keep hearing the word "equilibrium", which makes me immediately think about thermodynamics. The parallels between economics and thermodynamics seem interesting and strong. Both are fundamentally statistical sciences: thermodynamics is statistical mechanics applied to kinetics, and macroeconomics is thermodynamics as applied to microeconomics. At the micro level, there seems to be some parallels: individual molecules and individual economic actors are doing their individual things based on local information, and their individual behavior is in a sense "random".

Thermodynamics and economics are random in different senses: the randomness of individual molecules is statistical: we know precisely what two individual molecules will do when they meet; the randomness is in that we don't know the actual properties of individual molecules. In economics, we're dealing only with a few million or billion human beings (rather than a few trillion trillion molecules) so it's actually possible to get direct information about individual properties in economics; we don't, however, know how to predict detailed individual economic choices from those properties. Still, randomness is randomness, and there should be some similarities.

One interesting aspect of the parallels is that thermodynamics does not — apparently unlike some economic theories — simply stop at the conclusion that in the long run everything will be in thermodynamic equilibrium. Thermodynamics gives us important tools to manipulate thermodynamic systems to create nifty things like steam engines and jet planes. To what extent have economic theorists drawn on the physics of thermodynamics to understand economic behavior? If not (yet) to any great extent, then I'm extremely curious about what serious, detailed study of the parallels could achieve in economics.