Owing to its excellent qualities as a natural sorbent, rice husk (RH), a significant agricultural waste product obtained from the milling process, is employed as a biosorbent for engine oil. Engine oil spillages in rivers will flow to the ocean, exposing marine life to deadly contaminants. To date, there are very few natural sorbent studies specifically targeting engine oil removal. The purpose of this study was to optimise the significant factors in the efficiency of engine oil sorption by RH. Spectroscopic analyses using Fourier transform infrared spectroscopy (FTIR) and scanning electron microscopy (SEM) were performed to characterise the chemical composition and surface morphology of RH sorbent after pre- and post-treatment. A conventional optimisation approach, one-factor-at-a-time (OFAT), was used to assess the range of factors affecting the efficiency of engine oil sorption through heat treatment, heating time, packing density, and concentration of engine oil. The efficiency of engine oil removal obtained from this method was 74.5%. All the factors were assessed using a Plackett–Burman design (PBD) to eliminate non-significant factors. Furthermore, a central composite design (CCD) was employed to explore significant interactions among the factors. The quadratic model generated (R2 = 0.9723) fitted the data well. The optimised conditions from the CCD were 160 °C, 20 min, 0.16 g/cm3, and 12.5% (v/v), with improved oil sorption from 74.5% (OFAT) to 78.89% (RSM).