Analyses of drayage optimization applications shows that the required logistics fleets can be reduced by 21 percent.

FHWA researchers assessed drayage optimization algorithms to improve the productivity and efficiency of intermodal carriers in the United States

Date Posted
08/29/2014
Identifier
2014-B00919
TwitterLinkedInFacebook

Development of a Cross Town Improvement Project (C-TIP) Drayage Optimization Application

Summary Information

In cooperation with industry and on behalf of the U.S. DOT, the Federal Highway Administration, in 2013, finalized an assessment of a drayage optimization application sponsored under the Cross-Town Improvement Project (C-TIP). The purpose of this project was to test methods for leveraging technology to improve drayage operations. The evaluation of this application quantified benefits resulting from improved algorithms that optimize intermodal shipments from origin to destination.

The Cross-Town Improvement Project (C-TIP) has expanded under the Federal Highway Administration since it was introduced as a new project by the Intermodal Freight Technology Working Group (IFTWG). The IFTWG was seeking a solution to congestion, ineffectiveness, and increasing energy consumption that stemmed from the inefficiency of cross-town “rubber tire" interchanges (i.e., traffic from truck to rail, and rail to truck to rail) by developing and deploying an information sharing capability that enables the coordination of moves between parties to maximize loaded moves and minimize unproductive moves. In 2012, US DOT funded a project focusing on the Real-Time Traffic Monitoring and Dynamic Route Guidance components of C-TIP by incorporating optimization heuristics and a robust routing algorithm. The team selected by the U.S. DOT partnered with a drayage company in Memphis, TN to be the basis for developing and deploying the optimization algorithm and other related technology.

Approach

To effectively handle the complexity of the drayage problem inherent in the Memphis C-TIP and ensure that the algorithm was working properly:
  • The development and validation of the algorithm was divided into multiple iterations.
  • The algorithm performance was tested over a set of different problems ranging from well-known benchmark problems to specially customized ones.
  • The performance assessment of the algorithm relied on sets of real data that were collected from the daily operations of the drayage company in Memphis, TN over a six-month period.
To improve the capabilities of the optimization algorithm and to increase the robustness of its generated plans, data collection and testing was separated into two phases: one three-month Pre-Deployment Phase and one three-month Post-Deployment Phase. Each phase ensured that the algorithm design, development, and deployment were effective. Pre-Deployment and Post-Deployment phase issues were thoroughly examined, documenting all findings and response actions on a continual basis in order to minimize any impact to the project schedule, maintain user adherence to processes, and avoid similar issues in the future.

Results

In order to compare the Pre-Deployment and Post-Deployment phases, three levels of analysis – ranging from relaxed to restricted – were conducted on the data points collected during the Pre- and Post-Deployment periods. From each data set, 31 data points were selected to conduct a pairwise comparison of two data subsets with the same operational behavior. By comparing averages of the selected Pre- and Post-Deployment data points, the analysis found the following:
  • The required fleet reduced by 21 percent.
  • Total miles reduced by 9 percent.
  • Average miles per truck increased by 14 percent.
  • Total bobtail miles reduced by 13 percent.
This report, finalized in September 2013, is an assessment of research focusing on the use of technology to address drayage optimization. These findings along with the benefits provide a valuable resource to those considering the implementation of advanced technology for the optimization of intermodal freight logistics.
Deployment Locations