News | CSULB’s Research on Tracking Truck Flows for Drayage Efficiency Analysis

Stop the Video




CSULB’s Research on Tracking Truck Flows for Drayage Efficiency Analysis

Thursday, December 15, 2016

by By Anusha Ramakrishnan, MSCS 2018


Two of the significant problems in the shipping industry in Southern California are congestion and pollution, caused in no small part by the growth of container volume at the Southern California Ports. Though the pollution problem has been mitigated through the Alternative Maritime Power for Vessels & Clean Truck Program, the congestion problem persists.

To further analyze and combat this prominent issue of tracking inefficient drayage, California State University, Long Beach Researchers Shui Lam, Byung Lee and Jay Patel embarked on a research project. As shared by them in their research report, the San Pedro Twin Ports are the 9th largest port in the world, and account for approximately 40% of the U.S. international container volume. Out of this volume, 50% of the unloaded cargo is bound for Southern California markets, and drayage is significant in Southern California. About the different cargo being carried across Southern California, and the associated drayage. Although cargo volume has been stagnant for a few years, the researchers reported that it will increase as per an earlier peak period, as indicated by the trends.

Their research involved building a (research) device, that makes use of the prevalent practice of tracking trucks using Global Positioning System (GPS). GPS tracking collects data on where a truck has been and at what time. The Cal State researchers fashioned their device using a server and tablets (for ease of use of truck drivers). Their software was built so as to:

  • Log GPS locations and events, and transmit the information to the server
  • Receive the data from the server, and perform preliminary analysis, responding to the Web client
  • Extract and organize data into well-formatted files


Mobile Application’s User Interface

Web Client’s User Interface

Five drivers from a drayage company participated in their data collection process for a period of 2 months, spanning different types of cargo processes – heavy-tag, delivery to rail, Target delivery, and store delivery.

Using this device, they tracked different kinds of transactions – picking up and dropping off loads, as well as empty boxes, and analyzed the turn time spent by trucks in terminals. Their analysis helped them understand that most of the truck transactions take around the same amount of turn time, but some take much more time than others, causing a long-tailed distribution graph as shown beside.

The researchers faced several obstacles such as errors in data logging by truck drivers, and insufficient data due to inadequate participation by customers. Circumventing these obstacles, they were able to make some concrete deductions about truck flows.

Source: Tracking Truck Flows with Programmable Mobile Devices for Drayage Efficiency Analysis, METRANS UTC

One of their deductions was that the average truck turn time is higher than previous studies on a single terminal and that some work types – such as Target delivery and delivery to rail, have a high percentage of non-productive travels. Their research report concluded on the note that there may be more statistics that can be deduced by breaking down data into categories if more data can be obtained without requiring driver participation for a longer term. The findings of the SoCal researchers could play a major role in assessing and resolving the current congestion problem in the drayage of the ports.

The full report of this project can be found here, on the METRANS website.


Anusha Ramakrishnan

Anusha is a student assistant at METRANS Transportation Center involved with the METRANS weekly newsletter. She is a first year Viterbi School of Engineering graduate student specializing in Computer Science. Her interests lie in the field of data science and machine learning to improve general accessibility and usability of technology. She is currently also pursuing a Natural Language Processing course at USC to help combine her two interests of linguistics and technology.