Machine learning is a form of artificial intelligence that is applicable in all fields of study. It incorporates many algorithms used in carrying out various tasks such as classification, predictions, estimations, comparisons, approximations, optimization and selections. In estimating original oil in place, which affords the explorationist the foresight on the total amount of crude oil that is potentially in reservoir. Machine learning is found to perform reserves estimation with speed and accuracy where insufficient data are available. These among other attributes of Machine Learning motivated a systematic literature review of studies undertaken between 2010 and 2021 and explore the strengths and limitations reported in the studies. In the oil industry, different types of data are gathered from subsurface and surface in order to know the reservoir hydrocarbon potential. Sensorsare known to be able to collect these data in large quantity, analyse and used to predict the output.3127 articles related to the study were collated from 4 databases and after a series of inclusion and exclusion criteria were conducted on the articles, 104 journal articles met the criteria and were used for the review. Results of the study reveal that between the years under review, 2019 had the greatest number of articles (20 of the 104) pertaining to the topic reviewed. 61% of authors reported inadequate data while 39 % reported under-performance of the algorithm. It was also revealed that machine learning was applied to perform predictions/forecasting in the industry than it was used to solve other problems, while Artificial Neural Network (ANN) was the most used artificial intelligence technique. The study opens another vista of knowledge for researchers to navigate machine learning in the estimation of original oil in place and petrophysics analysis. This emerging technology is smart and makes data evaluation easy and straightforward.