Despite the potential shown by previous investigations on the use of ultrasound for the remediation of oil-contaminated sand, the influence and interactions among ultrasonic parameters and oily sand are unclear, leading to possible ineffective treatment and high-power consumption. In order to improve the process efficiency, this work analyzes the effects of ultrasonic power, frequency, and load toward the cleaning of crude oil�contaminated sand, using two different sample positions and sand types. Crude oil�contaminated beach sand and produced sand from offshore oil well were used as samples. They were cleaned in custom-made ultrasonic bath reactor for 10 min with power from 30 to 120 W, frequency covering 25�60 kHz, and sand load of 10�100 g. With experimental design consisting multiple factors and levels, the interactions between factors in all possible combinations were determined using ANOVA (n = 210). From p-value based at 95 confidence interval and extensive F test, the three most significant factors were the sand type, the ultrasonic frequency, and the interaction between sand type and frequency. The best setting for suspended samples involved high frequency of 60 kHz, whereas bottom samples preferred low frequency at 28 kHz. This finding was justified when the acoustic pressure attenuation, standing wave pattern, and surface pitting/cracking were found in correlation with the cleaning results. Overall, the maximum treatment under ultrasonic bath solely gained around 60, improvable by hybrid cleaning with other techniques such as chemical, biological, mechanical, and thermal. © 2021, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.