The creep behaviour of Ti-6Al-4V is an important factor in the design of safety-relevant components under thermo-mechanical load. The deformation-behaviour of the material is greatly affected by pre-hardening. However there is not yet enough knowledge available about the influence of pre-hardening on the creep behaviour for this material. Especially the beginning of the creep damage, in terms of primary creep, as well as its transition into the stable secondary creep stage, are of interest for this investigation. The execution of creep-experiments, with pre-hardened specimens should enhance the understanding of the creep behaviour of the material under static long-term block loadings. Most recently published studies in the field of creep damage are in the range of 0.4Tm (Melting Temperature) and higher. However, in Ti-6Al-4V extensive creep occurs even at lower temperatures. Even at room-temperature significant creep damage exists. In this thesis, static creep-experiments were performed at temperatures from 20°C to 370°C, constant weight load from 500MPa to 800MPa and pre-hardening from 0% to 5% plastic strain. With a time limit of about 100 hours per test, 23 static creep-tests were done, to evaluate the impact of pre-hardening. In addition, an amount of 14 cyclic block-experiments without pre-hardening were carried out, with a period time of 20 minutes per cycle. These tests were loaded at similar temperature and stress levels. Because of the cyclic load with dwell time at maximal and minimal stress, primary and secondary creep occurs in each cycle. The experimental results were evaluated as strain-time-plots as creep curves. Based on this findings, the creep stages are evaluated according to their feasibility for numerical implemtation within the finite elemente package. Firstly, a sensitivity study was performed in order to adjust the material parameters of the simulation to the time-dependent creep curves. The simulation is used to determine the behaviour of pre-hardened parts under thermo-mechanical, static and cyclic long-term load. Therefore, already in the FE-program Abaqus implemented creep models by Norton and Bailey as well as Kichenin were used. It turned out that these models are only able to display the cyclic creep experiments insufficiently and therefore a more holistically satisfying result could not be realised. For this reason a custom routine, to determine the creep parameters, needs to be developed in future work.