The Tunnel under the River Warnow in Rostock, north east Germany was let as a tolled concession in the 1990s. It opened to traffic in 2003. Openingyear forecast flows had been in excess of 20,000 vehicles per day. However, the outturn flows were considerably lower, of the order of 7000 vehicles per day. This meant that the revenue generated from the tunnel was insufficient to cover the financing costs. Faber Maunsell were appointed to investigate why the previous forecasts had been so inaccurate and to produce new, more reliable forecasts. This paper provides an investigation into the reasons for the original forecasting errors separating them into errors that could not have reasonably been foreseen and those that should have been foreseen. The issues covered include matrix building, network assumptions (including traffic calming) and assignment parameters (including willingness to pay). An assessment was made as to how much errors in each of these items had contributed to the overall error. There is then a descriptionof the methods used to produce updated forecasts taking into account observed traffic, new surveys and updated socio-economic forecasts. The surveys include traffic counts; origin destination studies by post card questionnaire and registration plate matching; and stated intention and revealed preference studies to determine willingness to pay tolls. An investigation was made into other means of increasing concession revenue including complementary measures (traffic calming on the alternative route) which could increase scheme capture. The results of this traffic study are being used in the refinancing of the concession. For the covering abstract see ITRD E135582
Abstract