First in MIL you can test the behavior of your desired function. So you con find

- understanding your function

- model errors regard to the abstracted function and implemented by your model

In Sil it depends. If you generate the code directly from your model by yusing embedded coder or targetlink than "compiler" errors or errors configuring the generation process will be found. If you implement the model by hand the you will find errors by coding the function

MILS: functional validation

SILS: functional validation of implemented code (auto-code or hand-code), code coverage analysis

PILS (Processor-in-the loop simulation): on-target functional validation, resource check (memory, execution time)

As of my knowledge SIL is used for Quantisation Errors(i.e., Arithmetic Problems )

And MIL is used for discovering the Overflow in Model Functionality.

Is this correct???