To improve trip distribution, employment categories should be defined on travel behavior patterns rather than for convenience (using NAICS 2-digit groups as many MPO do).

NAICS were not designed for modeling travel behavior. We need to use it wisely. A set of establishment types were developed from the 6-digit NAICS based on travel behavior patterns.

The author combined the household travel survey trip data with attraction side establishment related information from InfoUSA2010 (including 6-digit NAICS and more).

The author estimated the resident workers own earning group based on the combined information of household income, number of workers in the household, education, years of experience, industry they worked for, if part/full-time job, age, and position classification. The groups were ‘validated’ with the Occupational Employment Survey data from the Bureau of Labor Statistics.

The following were analyzed:

1) Employee work trip patterns associated with their employer establishment type and employee earning level, in terms of trip production rate, trip distance/length, and mode share by time-of-day.

The analyses suggest that the household income or worker earning each alone would not be sufficient to distinguish different work trip behavior. It suggests using a combination of employer establishment type and the workers’ own earning level to ensure the right type of employees commute to the places with the right types of jobs (e.g., service-high earning).

2) Visitor/Customer trip attraction patterns associated with the attraction establishments (business and government) by establishment type.

The analyses suggest that the customer/visitor attraction patterns are related to carefully defined establishment types: industry/office combined (attract almost no individual customers), retail, service-low (or high) attraction rate per employee.

3) Commercial vehicle (goods and/or service, produce/attract/deliver) trip patterns by establishment type.

The analyses suggest using the establishment types in 2) is sufficient.

The application:

The base year relationship between the employment distribution by establishment type and the development type in the MPO land use data (CommunityViz) was established. The relationship is used to forecast the future employment by establishment type. CTPP workers distribution by industry (2-digit NAICS based) and earning is used to further forecast the employment by employee type.