How Is AI technology Impacting The Logistics Industry Today?

Last month, Descartes Systems Group (a Talking Logistics sponsor) held its “2023 Descartes Innovation Forum for Broker, Forwarder and Customs.” I served as a panelist on a session titled, “AI in Logistics Technology – Ally or Adversary?” I was joined on the virtual stage by Nelson Cabral, National Customs Manager at DSV Air & Sea Inc., and Glenn Palanacki, VP of Industry Strategy at Descartes.

The session is available on-demand for qualified registrants, so if you missed it live, I encourage you to visit the event website and watch it when you can. 

In today’s post, I’ll share some of my comments related to the following question that was asked during our conversation: How is AI technology impacting the logistics industry today? 

When it comes to Artificial Intelligence as a whole, many companies (including logistics service providers) are still in the learning phase — that is, they’re trying to understand the technology landscape and what’s possible, and it’s a bit challenging because everything is evolving very fast. 

In fact, in a survey we conducted this past February, we asked members of our Indago supply chain research community — who are all supply chain and logistics executives from manufacturing, retail, and distribution companies —  “Is your company using Artificial Intelligence in its supply chain or logistics operations?” More than two-thirds of the respondents (68%) said they weren’t using AI today.

As one supply chain executive said, “I look forward to utilizing AI in the future for more agile decision making with the understanding that it will be a ‘crawl, walk, run’ journey.”

Estimated/Predicted Time of Arrival (ETA) is probably the most prevalent AI use case in logistics today. And in the warehouse, AI is infused into the operating systems of autonomous mobile robots.

Generative AI is the cool new kid on the block, so we’re still in the very early stages of how this capability will be used.

The first use cases of Generative AI will likely be in customer service. We all know that the most common questions customers ask are, “Where’s my order? Where’s my shipment?” So, we’ll soon see chatbots powered by ChatGPT or Bard technology that can answer those questions quickly, that can help customers resolve problems and complaints, and do so in a more cost-effective and scalable way. 

Another use case is procurement. Walmart, for example, has used AI chatbots in procurement to negotiate with a relatively small set of equipment suppliers, and according to the Harvard Business Review article that discussed the case, the company has started using the technology to negotiate rates in transportation. We’re also seeing the emergence of autonomous procurement in transportation, which uses AI, machine learning, and applied behavioral science to develop carrier profiles and price predictions.

You’ll also see Generative AI technology embedded into the user interfaces of supply chain and logistics software applications. So, instead of having to look at various reports and dashboards, you engage in a conversation with the system: 

“Which inbound shipments are behind schedule today?”

The system responds in a few seconds with a list of all the delayed shipments.

Then you ask, “Do I have inventory in other locations that I can use to fulfill the affected orders?”

The system responds in seconds with inventory positions related to affected orders.

Then you ask, “What are the current rates to ship from those locations instead?”

And the conversation continues on until you have all the information you need to make a decision. Can a logistics manager do this analysis today with existing metrics and dashboards? Yes, but it might take them 20 minutes versus 5 minutes or less with an AI assistant.

Further into the future, we might be able to feed a generative AI engine with supply chain data to help companies generate visual maps of their supply chain. Today, supply chain mapping is very difficult, time-consuming, and costly. But if you feed a generative AI engine with data from purchase orders, advance ship notices, invoices, bills of lading, status updates, proof of deliveries, and other transactions flowing between trading partners on a business network, it could potentially generate a graphical supply chain map for you.

On the global trade front, trade compliance professionals will use AI to help with HS classification. The system can take the given attributes of an item and marry that with classification decisions applied in the past to assign an HS code automatically (although a trade expert, at least initially, would still review and approve the classification). 

They will also use the technology to help with Denied Party Screening. The system can use natural language processing, for example, to highlight similar-sounding names, nicknames, aliases, misspellings, and so on to more quickly and accurately perform these screenings (again, with oversight from trade experts).

These are just a few examples. I’m certain that if we have this conversation again a year from now, there will be many more AI use cases in logistics, including some we’re not even imaging today.