How AI Could Help Streamline Logistics Operations
[Stay on top of transportation news: .]
As artificial intelligence continues to advance, technology developers are beginning to introduce AI-powered virtual agents and assistants that could change how freight brokers, dispatchers and fleet managers perform their jobs.
This latest wave of intelligent software and services includes AI agents that are designed to automate specific work functions and even make some decisions autonomously, essentially acting like virtual employees.
Lately, a growing number of startup firms and tech companies have been bringing these types of capabilities to the freight transportation industry with the goal of enhancing the efficiency and productivity of logistics operations.
“Tools are popping up literally every single day that help with deploying virtual agents in voice, text, email, chat and others,” said , a provider of AI-enabled business communications technology.
CloneOps.ai's Work previously held leadership postions at Transportation Insight Holding Co., Coyote Logistics and UPS Freight. (LinkedIn)
The company’s virtual agents automate phone calls, emails and text messages to handle various routine tasks, such as making outbound tracking calls.
More Q2 iTECH
►How AI Could Help Streamline Logistics Operations
►TMS Suppliers Prepare for AI Implementation
►Tech Vendors Roll Out AI Features at TIA
►Dysart:How Trucking Companies Can Harness YouTube Tools
Other opportunities for virtual agents to keep freight moving with fewer bottlenecks and less friction include automating communications on load availability, load confirmation and scheduling, he added.
“The sweet spot for us right now is supervised automation, where AI does the heavy lifting, but people still steer,” said Work, who recently joined CloneOps after previously holding technology leadership roles at Transportation Insight Holding Co., Coyote Logistics and UPS Freight.
Other tech vendors shared similar views on how AI and human workers will interact in the near term.
In general, tech companies are building AI tools designed to help employees at freight brokerages and motor carriers work more efficiently, while some also are working to develop increasingly autonomous AI agents over time.
Tech startups, such as Arnata, Envoy AI and HappyRobot, are among those actively developing AI workers for logistics and supply chain businesses.
Others are pursuing a more incremental rollout of AI-enabled workflow automation.
“Freight is still a service industry. Relationships matter," said Sandhu, co-founder of Chain. (BETA Podcast Network via YouTube)
“Instead of jumping to full agentic AI where you have agents pretending to be people, we should be focused on AI that augments the people who already understand the business,” said Annalise Sandhu, a former trucking company owner and a , an AI-powered supply chain visibility platform. “Freight is still a service industry. Relationships matter. So the winning platforms won’t be the ones that remove humans, they’ll be the ones that make humans better.”
Rather than drawing a hard line between AI assistants and AI agents, Sandhu suggested that it’s more useful to consider degrees of augmentation for the user of the technology, starting with tasks like making check calls and fielding emails.
From there, AI models could advance further by understanding users’ standard operating procedures, integrating with their technology applications and learning what to do next based on context and history, she said.
“You can’t get to that level unless your AI is built on a solid foundation that connects data across systems, brings brokers and carriers into the same workflows, and removes the silos,” Sandhu said.
The “final frontier” of full AI autonomy is still distant, she added.
“A lot of people are buying into the hype today and think we are already there, or are close, but this is years away,” Sandhu said. “In theory, once every piece of software in logistics has an AI layer and can speak to each other using a common protocol, AI can coordinate freight across businesses with minimal to no humans in the loop.”
Other technology exports noted that agentic AI and AI assistants or co-pilots can complement each other.
“I see a lot of value in AI models, whether it’s the agent model or co-pilot,” said .
As a cloud-based software platform that automates freight appointment scheduling for 3PLs, carriers and shippers, Qued uses AI to find prime appointment slots to optimize profitability.
“I see a lot of value in AI models, whether it’s the agent model or co-pilot,” said Qued CEO Gollapalli. (The Freight Pod via YouTube)
The platform is designed to perform a task from start to finish “without having to depend on a human unless there is an extreme situation — that is, an exception,” Gollapalli said.
For example, an exception can arise when a shipper requests that a load be picked up tomorrow, but the carrier or broker has no appointment availability that day.
“Before you pick the day after tomorrow as the next available slot, you definitely want to talk to your customer and make sure that they’re OK with you going beyond the estimated pickup,” Gollapalli said.
The Future of Logistics Operations
The introduction of increasingly advanced AI capabilities also raises questions about the long-term ramifications for logistics operations, said Chris Torrence, chief strategy officer at Optym, a provider of AI-enabled truckload dispatch optimization.
Some startups “have effectively automated the entire carrier sales experience,” he said. “What does that mean [for] a freight broker that has a thousand brokers on the floor? What does the future office look like? Do you have a fraction of that because [of] these tools and agentic AI bots [and] virtual assistants?”
Looking ahead to wider implementation of AI models in more aspects of the freight business, Torrence predicted that companies will pre-program defined parameters and tolerances for activities like rate negotiation.
TT's Seth Clevenger, Michael Freeze and Keiron Greenhalgh break down what ACT Expo revealed about trucking's road to sustainability.Tune in above or by going to .
However, he wondered whether AI agents will possess the level of expertise to interject when necessary and to reconcile exceptions during sales processes.
CloneOps’ Work said successful deployment of AI models has little to do with the availability of the technology or even where and how it will be used.
“It has much more to do with the readiness of the organization for change,” he said.
The advent of AI models also has implications for business valuations, according to The Tenney Group, which consults on mergers and acquisitions of transportation and logistics companies.
“The place where it comes up initially is for folks who are trying to combat increased operating expenses,” said the firm’s president and CEO.
“They’re considering some of these [AI] options as a way to both protect and build business value. I’ve had the opportunity to shadow some of these technologies, and they’re quite remarkable in the way that they can remove a tremendous amount of non-revenue-producing activity within an operation.”
Lean Solutions Group, which offers staffing alternatives in Colombia, Guatemala, Mexico and the Philippines for third-party logistics providers in the United States, also sees opportunities for agentic AI to improve efficiency and productivity.
For example, if a broker is outsourcing track and trace work to Lean Solutions, implementing agentic AI could dramatically increase the volume of check calls per person per day, said , the staffing firm’s chief technology officer.
“All the ops we do in logistics are either voice-based or email-based,” Quijano said. “AI voice mimics a humanlike conversation. It can answer rapidly.”
AI agents would conduct voice calls automatically, talking to drivers for status updates. Further, Quijano said he is looking forward to deploying AI voice agents for appointment scheduling and carrier sales, “where there’s a lot of negotiation happening.”
People in those roles typically must complete 100 or more calls or other types of information exchanges daily, he said.
“It can be exhausting for a person. There’s a lot of turnover involved [with] a people-only team,” Quijano said. “[They] can just do so much more with AI voice.”
AI Development Accelerates
Within the past several months, OpenAI, Google and Anthropic have introduced new products or projects conceived to ease and accelerate implementation of artificial intelligence.
Anthropic, developer of the AI chatbot Claude, in November released its Model Context Protocol, an open standard for connecting AI assistants with data sources.
This protocol, which contains rules for formatting and processing data, can replace fragmented integrations and can lead to a simpler, more reliable way to give AI systems access to the data they need, Anthropic said.
Quijano said he viewed MCP as a way for virtual agents to communicate with “all the different platforms there are, automatically, without having any integrations done.”
For example, the AI agents could communicate with a transportation management system or a financial application like QuickBooks, he said.
Want more news? Listen to today's daily briefing above or go here for more info
In January, OpenAI announced that it was developing an agent called Operator, which can browse webpages and interact with them in various ways, such as filling out forms, placing online orders and scheduling appointments.
“Operator is one of our first agents, which are AIs capable of doing work for you independently — you give it a task, and it will execute it,” OpenAI said in the announcement.
In April, Google announced a new protocol, Agent2Agent, designed to help AI agents collaborate. “AI agents offer a unique opportunity to help people be more productive by autonomously handling many daily recurring or complex tasks,” including aspects of customer service and supply chain planning, Google said in announcing A2A.
The protocol was launched with support from more than 50 technology partners and paves the way for AI agents, regardless of their underlying technologies, to collaboratively automate complex enterprise workflows, Google said.
To make informed business decisions, logistics companies will need to keep up to date on this rapidly changing technology landscape, Quijano said.
“I would encourage every single operator, every single person in logistics, to know what’s coming,” he said.
Complex supply chains can be managed with relatively few people and software tools, Pando's Jayakrishnan says. (LinkedIn)
Some large companies that operate internationally already are using AI agents in their supply chain operations, said , CEO of Pando, a logistics AI company.
Pando in February announced the launch of Pi, a suite of AI agents designed to automate freight procurement, dispatch planning, and freight auditing and payment processes for manufacturers, distributors and retailers.
Pi interacts with carriers by negotiating rates, verifying invoices and tracking shipment delays, among other activities.
Pando’s view is that complex supply chains often “can actually be managed with very few people and very few software tools,” Jayakrishnan said.
Global shippers deal with many different logistics service providers, he noted, adding that these logistics providers typically are not sharing their technology with each other.
At the same time, each carrier has its own nomenclature and ways of invoicing and carrying out other frequent tasks, Jayakrishnan said.
Pando processes millions of invoices daily — managing them as they come in, approving and paying them or rejecting them — without any human intervention, Jayakrishnan said.
