Equation based and acausal modelling of behaviour, suitable for modeling complex physical systems which encompass disciplines that are very different in nature. Object Oriented Modeling approach adopts concepts of Object Oriented Programming, where features such as class, inheritance, encapsulation, abstraction, etc forms the architecture of the modeling language. Non-proprietary language, which encouraged the development of a large repository of open-source libraries across various disciplines. For the same reason, there is a large number of Modelica simulation environments that are available, both commercially and free of charge

you can design each component of your circuit as a single component, so it is not dependent from the others. It has a lot of libraries, even though most of them are not free, and in those libraries you can find and use most of the needed components of your circuit. So they are already modeled and you don't have to spend time doing that. Moreover, once the code is compiled, the simulations are very fast to run, compared to Matlab of other languages. also, you can work a lot graphically: you can connect graphically the components, design the circuit as you prefer, and use many different levels, since you can use components that are made of different components, and graphically you can see the whole system, and the inside of each component. Moreover, it is possible also to see the results on the system that you design, such as temperature, pressure and mass flow rate of different selected points, and its variations during the time.

The main functional gain over other simulation software with modelica is that is has cross platform support with Altair, ANSYS, Dassault Systems etc. This makes it easier to use across the industry.