The efficient and prompt delivery of commodities, which is one of a port's main functions, is of the utmost importance. An essential component for the success of both businesses and economies is the optimization of these processes. The power of machine learning algorithms in improving port operations has been demonstrated in recent years. They are skilled in foretelling and improving various supply chain components, which makes this apparent. In this paper, we present a case example demonstrating the use of machine learning techniques to improve hydrocarbon operations at the Tangier Med Port. A quartet of commonly used machine learning algorithms was chosen, and they were put through a battery of tests to see how well they predicted cargo volumes and optimized the storage of hydrocarbon liquids. Our research shows that the random forest method outperforms its competitors, predicting cargo volume with an accuracy of more than 90% while also significantly increasing the efficiency of storing hydrocarbon liquids. By implementing this method in a real-world setting, turnaround times were significantly shortened, productivity increased, and customer satisfaction increased. Our research intends to highlight the benefits that machine learning algorithms provide in improving port operations and, at the same time, provide essential insights into their smooth integration within real-time settings. As we look to the future, potential projects might explore other machine learning techniques. They also have the potential to be combined with auxiliary technologies, such as the Internet of Things (IoT), to advance the development of port operation optimization. General Terms: Smart transport, Maritime transport, Hydrocarbons, Artificial intelligence, Prediction, Tangier Med port, Use case.

smart management, maritime transport, hydrocarbons, artificial intelligence, prediction, Tangier Med Port.