ABSTRACT
One of the difficulties in the development of artificial neural network (ANN) models is that, unlike statistical modelling where estimates of sample size can be initially computed, the number of samples or observations needed for training ANN models cannot be determined in advance. This chapter discusses an issue in the context of a neural network freeway incident detection model that was developed using ‘real world’ incident and traffic data. From a practical perspective, the impact of sample size on model performance will provide an insight into the sample size of 'real world' data required to train ANN incident detection models. The 'real world' data that was collected for model development and validation are then described and the procedures implemented for pre-processing this data before using it in the study are discussed. The issue of why the model’s performance ceases to improve as a result of increasing the sample size beyond 25 incidents are also discussed.
