Monte-Carlo event generation for the LHC

Abstract

This thesis discusses recent developments for the simulation of particle
physics in the light of the start-up of the Large Hadron Collider. Simulation
programs for fully exclusive events, dubbed Monte-Carlo event generators, are
improved in areas related to the perturbative as well as non-perturbative regions
of strong interactions. A short introduction to the main principles of event
generation is given to serve as a basis for the following discussion.
An existing algorithm for the correction of parton-shower emissions with the
help of exact tree-level matrix elements is revisited and significantly improved
as attested by first results.
In a next step, an automated implementation of the POWHEG method is
presented. It allows for the combination of parton showers with full
next-to-leading order QCD calculations and has been tested in several
processes.
These two methods are then combined into a more powerful framework which allows
to correct a parton shower with full next-to-leading order matrix elements
and higher-order tree-level matrix elements at the same time.
Turning to the non-perturbative aspects of event generation, a tuning of the
Pythia event generator within the Monte-Carlo working group of the ATLAS
experiment is presented. It is based on early ATLAS minimum
bias measurements obtained with minimal model dependence. The parts of the
detector relevant for these measurements are briefly explained.
Throughout the thesis, results obtained with the improvements are compared to
experimental measurements.