Computing environment of the START model system

START stands for Simulation of Traffic for Application in Real-Time. The general objective of the START project is to develop a practical tool to predict traffic flows and travel times in real-time using the information provided by the On-line traffic monitoring system together with historic and network information. This report describes the computing environment of the START 2 models. It deals with the way the various models are mapped onto software modules, the way these software modules are supported by hardware. The communication procedures of the START system with its suppliers of input data, its databases, and its users are described. Attention is paid to the performance aspects of key parts of the system. Finally, some standard subroutine libraries are presented, which will be used in the implementation of the Statistical Model.