This paper analyses and evaluates the variability of seismic demand and capacity of a case study jacket offshore platform considering different sources of uncertainty. The aleatoric uncertainty due to variability of near-fault ground motions, as well as uncertain properties of the damping ratio, material elastic modulus, material yield strength, mass, and gravity load have been investigated. The main aim of this study is to pursue which sources of the uncertainty considerably affect the seismic response of the platform. To this end, the sensitivity analysis was conducted not only in the particular earthquake level, but also in all other ranges of intensity using incremental dynamic analysis (IDA) with a special focus on the seismic collapse fragility. In order to reduce the number of simulations, the Latin hypercube sampling (LHS) scheme has been utilized as an efficient sampling procedure to combine the effects of modeling random variables. The collapse fragility curves are derived for each of the model realizations created with LHS technique. Thereafter, the summarized random fragility curve is compared with the deterministic mean parameter model fragility curve. It is found that the uncertainty in the mass and gravity load on the platform are the most influential variables, which can notably alter the IDA and collapse fragility curves. Furthermore, the random combination of the considered sources of uncertainty shifts the median of collapse fragility away from the mean parameter model collapse fragility. Overall, the effects of the uncertainties do not lead to notable changes in the summarized collapse fragility curve.