01625613

Component

The problem of supplier selection has received a great deal of attention by researchers. It is also one of the main components of Freight Activity Microsimulation Estimator (FAME), which is a disaggregate and comprehensive framework that simulates the freight movements for all industries and all commodities in the U.S. However, the supplier selection and supplier evaluation models in the FAME model face computational issues. Using the result of a nationwide establishment survey which was performed by the FAME developers, this study analyzes the supplier selection problem by evaluating the potential suppliers and then reducing the computational issues in FAME. The buyer’s behavior on selecting the distance range in which the trade forms is analyzed using a decision tree model and an ordered probit model. The results indicate that the decision tree model is more reliable in forecasting the consumer`s behaviors. In contrast to the initial theory that local businesses would perform their trade locally and in a limited range of distance, the modeling results show that the firm’s primary business type, as well as factors such as the commodity type, number of orders, and average number of weekly shipment could affect the distance to the main supplier.

This paper was sponsored by TRB committee AT015 Standing Committee on Freight Transportation Planning and Logistics.

01618707

17-01210

English

17p

2017

Transportation Research Board 96th Annual Meeting

Digital/other

Figures; Maps; References; Tables

Business practices; Consumer behavior; Decision trees; Freight transportation; Microsimulation; Probits; Suppliers

Freight Transportation; Operations and Traffic Management; Planning and Forecasting

Transportation Research Board Annual Meeting 2017 Paper #17-01210

TRIS, TRB, ATRI

Dec 8 2016 10:22AM