This research investigates the advanced use of artificial intelligence algorithms in transforming port management and operations. It focuses on delivering intelligent solutions to the complex challenges faced by modern ports through innovative computational and analytical models. To achieve this, the study implements two advanced models: a genetic algorithm for optimizing logistical operations and a deep LSTM neural network for intelligent forecasting. The genetic algorithm, applied in an AnyLogic simulation with 25 key nodes, reduced operation time by up to 35%, fuel consumption by 28%, and operational costs by 32%. The LSTM model, with two hidden layers and 60 neurons per layer, was trained on 120,000 data records. It reached 90% accuracy in traffic forecasting, surpassing traditional models like ARIMA, which achieved 84%. Beyond operational improvements, these technologies also contributed to lowering carbon emissions and boosting overall port productivity. This dual benefit is crucial amid growing environmental and sustainability demands in the maritime sector. The study also addresses key challenges, including the need for investment in cyber-physical infrastructure, workforce training, cybersecurity, and organizational resistance to digital transformation. Scientifically, the research presents significant contributions. It proposes an integrated framework to evaluate AI's impact on port performance, develops hybrid models with high adaptability, and analyzes the economic, operational, and environmental implications in a multi-dimensional context. Furthermore, it offers actionable strategies to overcome technical and organizational hurdles. Ultimately, the findings serve as a strategic guide for port managers, maritime policymakers, and researchers exploring smart technologies in logistics, paving the way for more efficient, sustainable, and intelligent port systems.