Econometric Count Data Modeling Applied to Stochastic Demand in Truckload Routing

Document Type

Conference Proceeding

Publication/Presentation Date



Decision Sciences Institute Annual Meeting at San Antonio. Nov. 2006


Truckload routing has always been a challenge. The routing problem itself is hard, but the complexity of solving the problem increases due to the stochastic nature of truckload demand. This paper examines truckload demand through the use of an econometric discrete choice model known as a Count Data Model. It ultimately uses Negative Binomial regression as its base. Two outcomes are produced, a count and the likelihood that the count will occur. The outcomes provide the ability to produce a more robust forecasting of truckload demand.

This document is currently not available here.