The field of predictive maintenance is moving towards more advanced and integrated approaches, combining machine learning, digital twins, and real-time data integration to improve equipment efficiency and reduce downtime. Researchers are focusing on developing more accurate and reliable models for remaining useful life estimation, as well as optimizing maintenance scheduling and decision-making under uncertainty. A key trend is the integration of prognostics with decision-making frameworks, allowing for more effective management of prediction errors and improved maintenance outcomes. Noteworthy papers include: A Quantile Regression Approach for Remaining Useful Life Estimation with State Space Models, which proposes a novel approach leveraging State Space Models for efficient long-term sequence modeling. Toward Decision-Oriented Prognostics: An Integrated Estimate-Optimize Framework for Predictive Maintenance, which introduces an integrated framework for jointly tuning predictive models and optimizing maintenance outcomes.