AI: Shippers’ edge in mitigating higher rates


Electronic logging devices, blockchain software, cyber security, application programming interfaces, and cloud computing — all are flooding into a logistics and transportation business where email, if not the fax machine, often is the dominant technology. The amount of data being collected and generated by these systems has been compared to an “explosion.”

In an increasingly automated age, shippers that do not turn to technology for help eliminating inefficiencies will be saddled with increasingly higher costs, Waggoner warns. “Shippers have a right to be concerned” about the direction of pricing and capacity in the truckload and less-than-truckload (LTL) sectors in 2018, as a stronger US economy increases demand, so says Doug Waggoner, CEO of Echo Global Logistics.

“Shippers will continue to see higher prices in 2018,” Waggoner said in an interview. “As the economy heats up, you’ll find carriers will be more selective on freight,” he said. “They’ve got to optimize utilization and yield. I would expect not only truckload capacity to tighten, but LTL carriers to be more discerning about the freight they take and more surgical in pricing.”

That means the evolution in logistics and transportation technology that has been gaining speed over the last several years, enabled by rapid advances in cloud and mobile computing, is likely to accelerate. Shippers, brokers, and carriers will have to come to grips with big data, and increasingly will rely on automation and AI to manage and make sense of that data.

The opportunities for the transformation of transportation through technology are being explored from the high seas of ocean container shipping to the back roads of trucking and last-mile delivery. This year, Waggoner and other industry leaders believe, will be a time of testing for many technologies that up to this point have been discussed more than deployed.

Companies such as Maersk Line, Panalpina, and Flexport are trying to harness AI to tackle a variety of vexing industry problems, especially supply chain visibility issues that affect just-in-time transits and equipment availability. There are plenty of opportunities to use AI to better manage surface transportation networks as capacity tightens and costs rise.

“The business world has started to embrace analytics and behavioral economics,” Waggoner said. Many other business sectors are far more advanced in their use of automation to gather data and AI to identify important patterns in the data than transportation, he said. In addition to the example of Netflix, he cited Google and Amazon and, in particular, financial markets.

“Think of the world of high frequency traders,” he said. “For 10 years now, you’ve had mutual funds and trading shops hiring PhDs to create these trading algorithms and test them and tweak them using massive amounts of financial data. There are companies making real money trading futures and stocks. It’s all based on patterns they see and discover in the data.

”The general business community, including shipping, “is starting to catch up with what the financial trading industry has been doing now for five to 10 years,” Waggoner said. Among those trying to push the envelope on how AI is used in transportation is Echo Global Logistics, which has used technology to propel its own fast growth since it was founded in 2005.

Echo is one of a cadre of Chicago-based logistics companies founded in the 2000s that grew rapidly to become billion-dollar enterprises. Those firms, which include Coyote Logistics, now part of UPS, represented an early generation of disruptors now targeted by the more recent wave of digital brokers hoping to benefit from mobile apps and the “sharing” economy.

In brokerage, Waggoner believes, scale matters, and Echo’s ability to tap more than 40,000 trucking companies gives it an advantage over competitors trying to start marketplaces from scratch. But the ability to deploy technology, and to write the algorithms underlying it, is just as important, he said, and that ability increasingly is available to smaller businesses, too.

“Technology is more accessible, and math geeks are in vogue,” he said. “If you want to teach yourself Python [a programming language], you can start writing AI algorithms. A lot of AI solutions use open-source software from Google. A lot of this work that has been going on within certain disciplines is becoming more mainstream, and people can see the power of it.

”Within his own business, Waggoner is using AI to help identify underlying shipping patterns that would be impossible for humans to uncover by just sifting data. “We get thousands of emails every day from carriers,” he said. “We’ve got algorithms that literally read those e-mails and extract data and populate load boards, rather than having a person type in the data.

”One benefit he expects from broader deployment of AI technology in transportation is more efficient price discovery. “We can price better, once we know which carrier in our stable of 40,000 companies has the right capacity and the right price at the right time. If you can’t use algorithms to answer that question, you’ll have to make a hell of a lot of phone calls.

”The development of AI in supply chains is being driven by the demand for more rapid replenishment of inventory as e-commerce expands, the increasing complexity of transportation networks, and shipping costs, which if left unattended, will inexorably rise. Systems designed to address those issues will need AI not just to crunch operational data but also search for patterns.

“There’s so much data coming at us that a human being can’t understand what the data is telling them,” Waggoner said. “That’s why you see the rise of [AI]. AI can see patterns in data that humans would never discover, and from those patterns test the accuracy of predictions and get better over time. You have to figure out what the data is telling you.”

The opportunities, he said, are unlimited, but not easy to seize. “When we first started, I brought in a data scientist and said your assignment for the next three months is to find problems where we can apply AI. He came back with a list of 27 projects. They’re not always obvious. It can take deep thinking to identify what problems can be solved and what data exists to solve them.

“It’s a combination of understanding the business, human behavior, what the variables are, testing hypotheses, and finally when you uncover patterns and things that can be automated through algorithms then you have to incorporate that into your system so it’s just part of your system,” Waggoner said. “These are not necessarily things that are intuitive,” he said.

For example, “We found out that when we give a shipper customer a quote on an LTL shipment, the number of options we give them impacts our close rate,” meaning how often Echo closes the transaction with that customer, said Waggoner. “You can give the customer too many or too few pricing options. If you give them the right number, the close rate is significant.”

AI could play a critical role in finding truck capacity currently “hidden” from shippers, he said. “In the old days, a customer filled out a profile, saying we tend to have backhaul lanes between points A and B and C and D. It was almost like relying on folklore. But what if algorithms could catalogue that and then tell you where carriers need freight and shippers have backhauls?”

“If you can learn that over time, think of how that could help shippers determine where they need to go to efficiently find capacity,” Waggoner said. That will not happen overnight, but many shippers struggling with tight over-the-road and intermodal capacity and higher rates in many US markets as 2018 gets under way cannot wait to throw that switch.


Source: JOC, “AI: Shippers’ edge in mitigating higher rates“. ( William B. Cassidy, January 3, 2018. 

Image source:

Stay Up-To-Date With Our Newsletter!