Semiconductor market sales have expanded massively to more than 200 billion dollars annually accompanied by increased pressure on the manufacturers to provide higher quality products at lower cost to remain competitive. Scheduling of semiconductor manufacturing is one of the keys to increasing productivity, however the complexity of manufacturing high capacity semiconductor devices and the cost considerations mean that it is impossible to experiment within the facility. There is an immense need for effective decision support models, characterizing and analyzing the manufacturing process, allowing the effect of changes in the production environment to be predicted in order to increase utilization and enhance system performance. Although many simulation models have been developed within semiconductor manufacturing very little research on the simulation of the photolithography process has been reported even though semiconductor manufacturers have recognized that the scheduling of photolithography is one of the most important and challenging tasks due to complex nature of the process. Traditional scheduling techniques and existing approaches show some benefits for solving small and medium sized, straightforward scheduling problems. However, they have had limited success in solving complex scheduling problems with stochastic elements in an economic timeframe. This thesis presents a new methodology combining advanced solution approaches such as simulation, artificial intelligence, system modeling and Taguchi methods, to schedule a photolithography toolset. A new structured approach was developed to effectively support building the simulation models. A single tool and complete toolset model were developed using this approach and shown to have less than 4% deviation from actual production values. The use of an intelligent scheduling agent for the toolset model shows an average of 15% improvement in simulated throughput time and is currently in use for scheduling the photolithography toolset in a manufacturing plant.