Intelligent Control of Vehicle-Based Internal Transport Systems

'Intelligent control of vehicle-based internal transport (VBIT) systems' copes with real-time dispatching and scheduling of internal-transport vehicles, such as forklifts and guided vehicles. VBIT systems can be found in warehouses, distribution centers, manufacturing plants, airport and transshipment terminals. Using simulation of two real-world environments, dispatching rules described in literature and several newly introduced rules are compared on performance. The performance evaluation suggests that in environments where queue space is not a restriction, distance-based dispatching rules such as shortest-travel-distance-first outperform time-based dispatching rules such as modified-first-come-first-served and using load pre-arrival information has a significant positive impact on reducing the average load waiting time. Experimental results also reveal that multi-attribute dispatching rules combining distance and time aspects of vehicles and loads are robust to variations in working conditions. In addition, multi-attribute rules which take vehicle empty travel distance and vehicle requirement at a station into account perform very well in heavy-traffic VBIT systems such as baggage handling systems.

Besides dispatching rules, the potential contribution of dynamic vehicle scheduling for VBIT systems is investigated. Experiments using simulation in combination with optimization show that when sufficient pre-arrival information is available a dynamic scheduling approach outperforms the dispatching approach. This thesis also evaluates the impact of guide-path layout, load arrival rate and variance, and the amount of load pre-arrival information on different vehicle control approaches (scheduling and dispatching). Based on experimental results, recommendations for selecting appropriate vehicle control approaches for specific situations are presented.