The bus industry is plagued by high accident costs and risks of passenger injuries. A bus simulator may offer a method of reducing accident rates by delivering targeted training to bus drivers who are most at risk. The first part of this thesis describes the design of the UK's first bus simulator, the fidelity of which was based on a thorough analysis of bus crashes. The second part describes the first studies in a multi-staged method to evaluate the training effectiveness of the simulator: face validity, effects of bus driver experience and stress on simulated performance and simulator sickness. This approach ensured that the ABS has a reasonable level of fidelity, is capable of eliciting behaviourally valid responses from bus drivers and is the first step is achieving training transfer effectiveness. The final study investigated the occurrence of self-bias in bus drivers. The conclusions drove the design of simulated scenarios to be used for bus driver training. Keywords: Bus, Simulator, Fidelity, Validity, Accidents, Driving, Stress, Training

No description available.