Access to Destinations: Refining Methods for Calculating Non-Auto Travel Times

Abstract

The functioning of the system of land use and travel networks in a region can be encapsulated into measures of the ease of reaching destinations from various locations, often referred to as accessibility measures. Regardless of the form used to specify accessibility, all measures require as inputs travel times between the zones of a region. For most transportation planning purposes, these travel time calculations are limited to motorized modes (auto and public transit), since these modes carry the bulk of all urban travel. In this research study, attention is focused on developing methods for calculating travel times by non-auto modes, including walking, bicycling and public transit. Unique networks for each mode are developed, accounting for the presence of special facilities such as pedestrian or bicycle trails and on-street bike lanes. A statistical model is estimated to identify the influence of special bicycle facilities on travel speeds, using GPS data collected from bicyclists in a real-world setting. These methods are demonstrated with an application to a section of the Twin Cities metropolitan region encompassing parts of the cities of Minneapolis, St. Paul and Bloomington. The output of the application of these methods are a set of maps depicting travel sheds from various locations within the study area. The data are displayed for three points in time: 1995, 2000 and 2005. Changes to these travel sheds over time are demonstrated with maps that show the difference in travel time between each set of origins and destinations for each pair of years. The research concludes with some suggestions about the uses of the travel time data, such as the calculation of multimodal, multipurpose measures of accessibility.

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 — 9 citations exist upstream.

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