A Gravity Model Using Spatial Correlation of Time Series Data for Freight Distribution Estimation: A Case Study of the County-Level Coal Distribution Estimation in United States
With an interdisciplinary nature of transportation studies, many transportation theories were adopted from physics, computer science, statistics, or other engineering researches. Similarly, the gravity model, originated from Newton’s law of universal gravitation, has been used in freight distribution estimation, and passenger demand analysis, for decades. Researchers have utilized transformation, optimizing parameter estimates, or adding adjustment factors to improve the gravity models. However, a zone-to-zone distance matrix, the source for impedance factor of gravity models in most practices, has many fundamental limitations. One assumption of using the distance matrix is that the friction factor is a decreasing function over a distance; i.e., the estimated shipment volume (or trips) decreases as the distance increases with a given production and attraction volume. This has been proven wrong in many cases by research and in practice. Nevertheless, the zone-to-zone distance matrix is still most-widely used in gravity models for both passenger and freight distribution estimation. The objective of this paper is to provide an alternative impedance factor of gravity model for freight distribution estimation, using a spatial correlation matrix constructed upon the correlation between the production of origin and the attraction of destination in time series. The results are compared to the gravity model using the county-to-county great circle distance, by transportation mode, with four transformation types. The U.S. domestic coal distribution data from the Energy Information Administration was used to validate the proposed approach. The case study shows that the proposed approach outperforms the conventional gravity model using a distance matrix.
-
Supplemental Notes:
- This paper was sponsored by TRB committee AT015 Standing Committee on Freight Transportation Planning and Logistics.
-
Corporate Authors:
500 Fifth Street, NW
Washington, DC United States 20001 -
Authors:
- Lim, Hyeonsup
- Chin, Shih Miao
- Hwang, Ho-Ling
- Han, Lee D
-
Conference:
- Transportation Research Board 96th Annual Meeting
- Location: Washington DC, United States
- Date: 2017-1-8 to 2017-1-12
- Date: 2017
Language
- English
Media Info
- Media Type: Digital/other
- Features: Figures; References; Tables;
- Pagination: 17p
- Monograph Title: TRB 96th Annual Meeting Compendium of Papers
Subject/Index Terms
- TRT Terms: Case studies; Coal; Estimation theory; Freight traffic; Gravity models; Spatial analysis; Time series analysis
- Uncontrolled Terms: Spatial correlation
- Geographic Terms: United States
- Subject Areas: Freight Transportation; Operations and Traffic Management; Planning and Forecasting;
Filing Info
- Accession Number: 01630295
- Record Type: Publication
- Report/Paper Numbers: 17-06189
- Files: TRIS, TRB, ATRI
- Created Date: Mar 27 2017 9:34AM