The scientific literature lacks methodologies for assessing HRES investments because of complex interactions among HRES components, conflicting objectives, and uncertainty in the energy demand. This work fills this gap by presenting a methodology to assess long-term investments in HRES for off-grid buildings. To strengthen the reliability of the assessment, the methodology exploits an HRES optimal synthesis, sizing and operation simulating the system and embodying the interactions among HRES components for both electrical and heat generation, the building loads evaluation, and the verification of the effect of uncertainties in the energy demand. These aspects have already been investigated, but, to our best knowledge, have never been considered in a concurrent and integrated way. The resulting designs are optimised through an NSGA-II algorithm that minimises the system CO2 emissions and differential cost with respect to a solution that does not use renewable energy sources. The soundness of the solutions is evaluated through a scenario analysis and a sensitivity analysis using the Morris method. The applicability of the proposal was verified through a case study in an off-grid facility, showing possible savings up to 47 k€ and 320 tCO2. The methodology is thus strongly consistent with the building decarbonisation addresses set by the policymakers.