Geometric Hyperparameter Optimisation of ALICE Monte Carlo Transport Simulations - Summer Student Report

Monte Carlo (MC) transport simulations often require a significant amount of computing resources due to the detailed physics simulations. There are various physics models and thousands of particles are transported through, in the particular case of this report, a complex geometry resembling the ALICE detector. This report outlines the development and use of automatic optimisation procedures using a geometric approach to reduce the number of relatively computationally heavy MC steps whilst removing a negligible number of hits in GEANT4. The approach was able to reduce the number of steps by roughly 8-9% whilst removing <1% of hits in any detector in the main barrel region of ALICE. Further study should be conducted, using the developed framework, to look at the absorber materials inside the detector. Alternative methods should be explored to optimise the region including the ZDC detectors.