The Inconvenience of Waiting Time for Ferry– A Norwegian Case Study



The Inconvenience of Waiting Time for Ferry– A Norwegian Case Study

Authors

Stig Nyland Andersen, Norwegian Public Roads Administration, Trude Tørset, Norwegian University of Science and Technology

Description

Our research improves the modelling of the waiting time for ferries in the transport models. Five ferry connections with different characteristics are analysed. Arrivals registered continuously on both sides, with 96000 observations in total.

Abstract

Transport models are important tools in assessing new infrastructure projects. The effect of a new infrastructure projects relies heavily on the improvement in the level of service. A better representation of Level of Service data would give more precise traffic prognosis. Hence, the cost benefit analysis would improve from better estimations of both changes in generalised cost and traffic volumes.

The western coast of Norway consists of a wide range of fjords and islands connected by ferry services. The ferry service can be decomposed into three cost or inconvenience related elements in the transport model. The ferry ticket (direct cost), the crossing time on board the vessel and last, the waiting time. While the ferry ticket and the crossing time are set values from the timetable, the waiting time is composed of the service frequency and the arrival distribution of the passengers. The Norwegian transport model has the assumption that the waiting time can be represented by half the headway between two departures, maximum 120 minutes. However, this assumption relies on random arrivals to the ferry quay.

Today, the Norwegian transport model have issues when modelling ferry services. Generally, the model underestimates the traffic on most of the ferry services. One part of this problem might be explained by overestimation of waiting time for the ferry, resulting overestimations of generalised cost for trips including a ferry.

Our research aims to improve the modelling of the waiting time in the transport models. We analyse the arrival distribution of passengers to the ferry by an empirical study of five ferry services. All arrivals on both sides of the ferry connection was recorded continuously for one week or more on each ferry service. The registered data contains more than 96.000 observations in total. Furthermore, travel surveys were conducted on the ferries to give a better understanding of the type of passengers on the connection regarding trip purpose and mode of travel.

The ferry services represents a wide range of service types. Three of the ferries serve the E39 Coastal Highway Route with a larger proportion of long trips, while the other two connects islands to the mainland with a larger proportion of commuting trips. The headway between departures among the ferry services varies from 20 to 60 minutes during peak hours.

The arrival distribution will be analysed to give recommendations on how to improve the modelling of waiting time for ferries in the transport models. (1) How can the arrival distribution improve the modelling of waiting time for ferry services in the transport model? (2) Does the waiting time vary due to the type of connection and the ferry service?

We expected the study to show that passengers adjust their travels accordingly to the timetable to minimise the waiting time on the ferry quay. Furthermore, the average waiting time might vary according to the characteristics of the ferry service and the trip purpose of the passengers.

Publisher

Association for European Transport