Schedule coordination is the primary form of demand management used at busy airports. At its core, slot allocation involves a highly complex combinatorial problem. In response, optimization models have been developed to minimize the displacement of flight schedules from airline requests, subject to physical and administrative constraints. Existing approaches, however, may not result in the best itineraries for passengers. This paper pro- poses an original passenger-centric approach to airport slot allocation to maximize available itineraries and minimize connecting times. Because of the uncertainty regarding passenger demand, the proposed approach combines predictive analytics to forecast passenger flows in flight networks from historical data and prescriptive analytics to optimize airport slot assignments in view of flight-centric and passenger-centric considerations. The prob- lem is formulated as a mixed-integer nonconvex optimization model. To solve it, we propose an approximation scheme that alternates between flight-scheduling and passenger-accommodation modules and embed it into a large-scale neighborhood search algorithm. Using real-world data from the Singapore Changi and Lisbon Airports, we show that the proposed model and algorithm return solutions in acceptable computational times. Results suggest that slot-allocation outcomes can be made much more consistent with pas- senger flows at a relatively small cost in terms of flight displacement. Ultimately, this paper provides a new paradigm that can create more attractive flight schedules by bringing together airport-level considerations, airline-level considerations, and, for the first time, passenger-level considerations.