Development of the Monte Carlo event generator tuning software package Lagrange and its application to tune the PYTHIA model to the LHCb data

One of the general problems of modern high energy physics is a problem of comparing experimental data, measurements of observables in high energy collisions, to theory, which is represented by Monte Carlo simulations. This work is dedicated to further development of the tuning methodology and implementation of software tools for tuning of the PYTHIA Monte Carlo event generator for the LHCb experiment. The aim of this thesis is to create a fast analytical model of the Monte Carlo event generator and then fitting the model to the experimental data, recorded by the LHCb detector, considering statistical and computational uncertainties and estimating the best values for the tuned parameters, by simultaneous tuning of a group of phenomenological parameters in many-dimensional parameter-space. The fitting algorithm is interfaced to the LHCb software framework, which models the response of the LHCb detector. Typically, the tunings are done to the measurements which are corrected for detector effects. These corrections depend on the model which is input into the detector simulation, and the proposed approach allows to avoid this dependence. This tuning methodology and the developed software tools were tested on a small subset of kinematical parameters, which describe the gross fraction of the hadronic final states in high energy $pp$-collisions.