Abstract
This paper concerns the exploration of statistical models for the analysis of observational freeway flow data, and the development of empirical models to capture and predict shortโterm changes in traffic flow characteristics on sequences of links in a partially detectorized freeway network. A first set of analyses explores regression models for minuteโbyโminute traffic flows, taking into account time of day, day of the week, and recent upstream detectorโbased flows. Dayโ and linkโspecific random effects are used in a hierarchical statistical modelling framework. A second set of analyses captures dayโspecific idiosyncrasies in traffic patterns by including parameters that may vary throughout the day. Model fit and shortโterm predictions of flows are thus improved significantly. A third set of analyses includes recent downstream flows as additional predictors. These further improvements, though marginal in most cases, can be quite radically useful in cases of very marked breakdown of freeway flows on some links. These three modelling stages are described and developed in analyses of observational flow data from a set of links on Interstate Highway 5 (Iโ5) near Seattle. Copyright ยฉ 2002 John Wiley & Sons, Ltd.