Characterization of heavy fuel oil (HFO) is highly important to ensure technically, economically, and environmentally proper operation of the engines and power plants that use this source of energy. This applies in particular to the shipping industry. Here, we demonstrate that the combination of standard 1 H nuclear magnetic resonance (NMR) spectroscopy and multivariate data analysis can be employed for quick and accurate extraction of parameters pertaining to the physical and chemical properties of complex suspensions, such as HFO. For 82 HFO samples of known origin, good prediction models were obtained for a large number of characterization parameters, including the calculated aromaticity index, the density, gross and net calorific values, and water and sulfur contents, as well as micro-carbon residue.