The growing sophistication, size, and interconnectedness of the current transportation systems have led to the paradigm shift of the time-honored methods of deterministic models to more data-driven artificial intelligence (AI)-driven models. A combination of road, rail, metro, bus, freight, and new mobility services create multi-modal transport systems which produce huge amounts of heterogeneous data and are characterized by tight spatialtemporal coupling, nonlinear behavior, and uncertainty. AI and sophisticated optimization methods can offer potent means of deriving actionable data insights, predicting system behaviours, optimising processes and helping in the real-time decision-making process at various levels of decision-making. In this chapter, a detailed and substantive account is made of data-driven modeling, artificial intelligence algorithms, and optimization of multi-modal mode of transportation. It addresses the source of data, preprocessing, predictive and explanatory AI models, multi-objective optimization and control, reinforcement learning and digital twin assisted decision support. The chapter further doubly focuses critically on the issues associated with generalization, fairness, transparency, privacy, and governance, and possibilities of future research directions of trustworthy, scalable, and real-time AI-enabled multi-modal mobility systems.