摘要: |
Large ferry systems and ferry systems operating in major urban areas can often rely on regional travel models using the traditional four-step travel demand forecasting process to predict ridership levels for their routes. However, smaller agencies and agencies in rural areas often do not have either the data or the resources to develop and implement complex forecasting models for their systems. A variety of simplified forecasting methods are reviewed that can be implemented with inexpensive, commercially available software using ferry traffic counts and readily available socioeconomic data from published sources. Data from the Woods Hole, Martha's Vineyard and Nantucket Steamship Authority are used to illustrate and compare the applications of the following methods: simple annual compound growth rate (5- or 10-year historical trend, 12-month moving totals, weighted trends), simple linear trend extrapolation of time series data, and simple and multiple linear regression models using population, employment, and other independent variables for available outside sources. |