In this work, we present a sim-learnheuristic approach to solving the Multi-Source Team Orienteering Problem (MS-TOP) under stochastic and dynamic conditions. By combining deterministic, stochastic, and dynamic components, our methodology is able to efficiently address this complex variant of the MS-TOP. Specifically, we conduct a case study using real-world data from electric bicycle stations in Barcelona, gathered from Open Data Barcelona. The study involves efficiently distributing bicycles to various stations starting from different hubs, and finishing at a central depot. This task becomes significantly more complex with the inclusion of stochastic travel times and dynamically changing factors such as weather or traffic congestion. Our approach is able to find high-quality solutions in short computational times by combining heuristic algorithms, simulation, and machine learning components. The case study demonstrates the practical applicability and effectiveness of our method in a real-world context.