The interactive expert system for the winter maintenance service was developed by the University of Applied Sciences in the so called project "Interventionsmanagement Winterdienst (IMWD)" by order of the canton Lucerne. The aim of the expert system is to propose in an automatic computer based way spread- or a plough-actions for the roads maintained by the Canton Lucerne. For that, the actions are proposed by the expert system at least 30 minutes before a hypothetical critical state of a road. As a by-product theexpert systems allows to reduce stress for the people, which decide actions for the winter road maintenance. The algorithm of the IMWD expert system is based on 13 different decision cases, which are defined through various weather dependent criteria. The 13 decision cases are developed by the winter maintenance team of the Canton Lucerne in collaboration with meteorological experts (MeteoSwiss, Swiss Federal Institute of Technology Zurichand Meteoradar). The IMWD expert system collects the current sensor data of the road network and various meteorological data at regular intervals, the algorithm then periodically analyses the actual data based on the criteria of the 13 different cases and proposes a spread-plough-action for oneor several roads. This report describes the IMWD expert system and shows how the reliability of the interactive system is statistically evaluated. Results of the reliability are also presented. For the covering abstract see ITRD E143097.
Abstract