For my master's thesis I extended the Markov Reward Model Checker tool to use Interval Arithmetic. The new program provides rigorous error bounds, giving confidence that the results are not compromised by floating-point roundoff errors.

For this project I was supervised by dr. Pieter Collins.

Source code for IMRMC can be obtained from the git repository at git://git.ennoruijters.nl/imrmc.git.

The master's thesis can be found here.