This paper investigates the sensitivity on choice of heat pump concept to uncertainties and variations in boundary conditions for large scale heat pumps with a Danish district heating system as case example. A performance mapbased heat pump model is used to evaluate performance of different heat pump concepts. The heat pump concepts are optimised in terms of concept design and prepared for evaluation of feedback control parameters and on/off strategies, which calls for a nonlinear optimiser such as genetic algorithms. The optimisation considers the operational price of the concept, regarding Coefficient Of Performance (COP), heating capacity and investment cost and is followed by investigations of the possibility for providing ancillary services. A sensitivity analysis is carried out to investigate whether, which and how variations in cost function weighting affects the optimisation and consequently the choice of heat pump concept. The results indicate a very robust optimised concept.