Work zones represent typical bottlenecks on motorway sections. Due to aging of Slovenian motorway network, roadworks and the corresponding delays are becoming a major issue. While the majority of roadworks is planned, it is not straightforward to predict the delays they will cause even with a good prediction of the traffic volumes. The problem is to determine the work zone capacity, which is affected by several work zone characteristics, road elements and traffic structure. Besides, the stochastic nature of capacity cannot be ignored, since the same traffic values can result in different traffic conditions. This paper presents work zone capacity estimation based on the data of the recent work zones on Slovenian motorways. We collected data of various short- and long-term work zones that were recently present on Slovenian motorway network. We have studied the effect of heavy vehicles, longitudinal grade, crossovers and lane narrowing on the work zone capacity. Two statistical methods were applied to study those factors, namely regression analysis and survival analysis combined with maximum likelihood method. For each work zone type we have estimated the Weibull cumulative distribution function of capacity and queue discharge rate. The estimated capacity can provide an estimate of breakdown probabilities, delays and levels of service of motorway sections in work zones and as such it provides a useful tool in planning maintenance of the motorway network.

capacity, cumulative distribution function, maximum likelihood method, regression analysis, work zone