APPLICATION OF NEURAL NETWORKS TO MODAL SPLIT MODELLING

J Adranka, J Jovic
1997 Yugoslav Journal of Operatio ns Resea rch   unpublished
A transport system user's cho ice of one among alternative travel models is based on acquired. experience and personal views concerning t he availelable alternatives. Hence ou r assumption that it is possible to create a neural netwo rk which can be ' fed" by real (collected) input data on user type, param eters relevan t for decision making and information about user behaviour under specific conditions. T his in formation is used to train the network (by e mploying some of the kno wn algorit
more » ... he kno wn algorit hms) to si mulate user behaviour. This paper briefly presents a special methodology we have used in performing research that permits filin g the gap in the knowledge of user behaviour to wards offered tran sport conditions. Based on this, a NEKOP (neural network and user behavior> model for user behaviour simulat ion has been developed. T he r esults obtained by modal split simu lat ion are compared with real data. The field of application and the const raints of the NEKOP model are described .
fatcat:agswnfnen5f6lpivc4rwxbv3sy