COMPARISON OF COST MODELS FOR URBAN TRANSIT

Abstract

This article reviews and evaluates three distinct approaches to modeling transit operating costs: unit cost models, four-variable regression models, and slowness regression models. Six individual operating cost models were reviewed: a four-variable unit cost model incorporating measures of annual vehicle hours, annual vehicle miles, number of peak-hour vehicles and annual revenue passengers; a four-variable regression model; a daily cost four-variable model incorporating measures of average daily vehicle miles of service, average daily vehicle hours of service, average daily passenger revenue on route and peak vehicles needed on route; two slowness function models; and an urban environment cost model that includes variables that help explain differences in operating cost such as population density, age of fleet, age of city, scheduled speed and labor costs. These models were tested on data from a number of U.S. transit agencies to determine the predictive ability of the models in terms of (1) the degree to which the model explained variation in cost and (2) the data requirements for estimating parameters using the model. Results of this analysis indicated that the simplest model in terms of data requirements is the slowness model, however it is the least desirable for general application. The four-variable unit-cost model is much better suited to generalization. Although it, too, does not encompass all factors influencing operating costs, it is well suited for short-range planning purposes and was preferable to the four-variable regression model.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 3 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.