Development of a Regional Forecasting Model Based on Google Transit Feed

Abstract

Travel forecasters struggle with the task of building and maintaining large transit network models. In large dense regions it is difficult to have a comprehensive data set for the entire transit network. This is especially true in the Greater Philadelphia region, which includes multiple rail and bus modes and more than ten different operating agencies. GTFS (General Transit Feed Specification) is a data standard defined by Google since 2006, which is used by many transit agencies to publish their schedule data for the use in Google Maps and other internet applications. The Delaware Valley Regional Planning Commission (DVRPC) has used GTFS as the main source for transit data in the development of its new regional forecasting model. Several benefits from the use of GTFS have become obvious during this work, such as data accuracy, format standardization, and regular updates. The model development process includes data translation, integration with other data sources, and calibration of model components that are not included in GTFS, such as access to transit, transfers, representation of fare and finally path-building and path-choice. The paper also explains challenges and benefits of the use of GTFS in a regional forecasting model and gives insight in the current state of model validation.

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 — 10 citations exist upstream.

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