Ship performance monitoring is integral to maritime operations, encompassing propulsion, safety, and environmental considerations. Accurate prediction of ship fuel consumption is vital for optimizing efficiency, reducing costs, and achieving sustainability objectives. In this context, machine learning (ML) techniques have emerged as powerful tools for analyzing vast amounts of data collected from ships onboard sensors. However, the black-box nature of conventional ML models presents challenges in understanding the rationale behind their decisions and predictions, hindering their adoption in critical maritime applications. This paper explores the integration of explainable machine learning (XAI) techniques into ship performance monitoring systems to enhance transparency, interpretability, and trustworthiness. We conduct a comprehensive review of XAI methods suitable for maritime applications, considering their ability to provide insights into ship operational data while maintaining computational efficiency. Furthermore, we present a case study demonstrating the application of XAI techniques to analyze historical ship performance data and identify critical factors influencing fuel efficiency. Our findings illustrate the effectiveness of XAI in uncovering actionable insights, enabling ship operators to make informed decisions to optimize performance, reduce operational costs, and mitigate risks.