Model Workers Trip Appropriately Based On Travel Behavior Differences
Corresponding Author: Mei Ingram, Institute for Transportation Research and Education
Presented By: Mei Ingram, Institute for Transportation Research and Education
Abstract
This will be a follow-up of the author’s presentation at the 2015 Planning Application conference on employer establishment category “Better trip distribution starts from developing travel behavior based employment categories”, in which the author argued that the NAICS 2-digit codes do not provide a good employment categories for modeling travel behaviors and appropriate employment categories need to be developed carefully based on trip attraction behaviors (if not also other factors).
Workers commute and at work travel behaviors are major parts of the regional travel, and naturally major components in a forecast model. The author would argue that the workers travel behaviors need to be modeled appropriately based on their travel behavior differences.
The author will show the workers travel behavior analysis of the 2016 Greater Triangle Household Travel Survey (2016HTS), in terms of trip frequency, trip rate, trip time length and distance, mode share, by time of day and by purposes of home-based work (chained or unchained), at-work-work related and at-work personal.
The analysis should focus on the workers’ travel behavior differences by employer establishment type (industry, office, service – low / high individual customer attraction rate per employee, or retail); and further, by employee type which is a combination of employer establishment type and workers’ own earning level (industry/office – low/high earning, service – low/high earning; and retail – low/high earning) as a surrogate to the within-organization ‘occupation’.
The author will provide suggestions on how to model the workers’ travel behaviors appropriately. The author will argue that the current time-of-day practice of simply slicing daily (or peak and off-peak) trip table (after motorized trip mode share) into time-of-day pieces desired is not appropriate, as the time-of-day behavior varies in terms of trip purpose composition to trip length/distance and mode and so on.
Finally, an approach on forecasting workers’ employee type for future year is provided: a relationship can be developed using the base year CTPP tables of employment by census Industry and Earning level at both residence and work location, the InfoUSA data (for employer establishment type), and CommunityViza; and applied to future year.