Abstract
This project investigates the design of a general-purpose programming framework for discrete-event system (DES) simulation algorithms, with a focus on the event-scheduling method. Based on a modular and extensible design, the framework defines core components such as the simulation clock, future event list (FEL), and queue structures, and provides reusable logic for time advancement, event processing, and performance statistics. The framework is applied to two representative case studies: a queue-based simulation of physical fitness test scheduling, and a track-guided hospital logistics robot scheduling system. In the first case, we simulate the queuing dynamics of student groups taking timed fitness tests, validate the model against real-world observations, and propose scheduling system improvements to reduce waiting time and improve fairness. In the second case, we build and simulate a logistics car scheduling system under both single and multi-controller scenarios across a multi-floor hospital, addressing routing, dispatching efficiency, and system load balance. Numerical results demonstrate the effectiveness of the simulation framework in modeling complex real-world systems. This project also explores further directions including the hybridization of event-scheduling and activity-scanning methods and data structure optimization.
Type