News / Is AI the right tool for logistics operators looking for 'milestone visibility'?

Photo: © Yota Nik | Dreamstime.com

In the second of our series on AI in logistics, we look at how it can help forwarders with their pain points

Embracing Artificial Intelligence (AI) can help forwarders address the “pain points” in their supply chains – but how looks different for each company.

Stephanie Herminjard, chief strategy officer at the world’s largest logistics AI platform, Raft, told The Loadstar: “Any process where lots of manual documents play a role is an area where you can really take advantage [of AI].”

An example, she highlighted is an operator that typically receives between 500 and 2,000 shipment update emails a day, all of which they need to read, contextualise and then use to update their transport management system (TMS).

This laborious manual procedure opens the door to human errors – “fat fingers”, joked Ms Herminjard.

She added: “Also, in many cases you might actually only update the absolute necessary data fields that are required, because you know you’re going to have another five or six emails related to the same shipment.

“Updating it numerous times is too time-consuming, whereas with AI you can extract all the information from an email and update the system a hundred times.”

This, she explained, means forwarders could offer their customers greater visibility in a shorter timeframe.

“Customers are looking for milestone visibility, especially on ocean phases, it’s always a pain point. If you have a lack of milestones, AI could help you think about which milestones you’re missing, for what shipments, how to update milestones and what information you need in your data fields to get them.”

Ms Herminjard added that AI was especially useful in logistics, due to “the number of parties the industry deals with”.

It could be particularly useful if a forwarder is asked for a quote outside the usual volumes and must then communicate with every member of the chain, truckers, ocean carriers, airlines, etc, to produce a final rate.

“There’s so many integrations that need to be done, and a lot of the processes are still very heavily determined by unstructured data – with emails, PDF documents, Excel files, commercial invoices, etc,” she continued.

“For freight forwarders, 60% or 70% of their most common volume is integrated in a structured way, but where it gets complicated is the remainder of the business where there are so many different files and structures that often go through inefficient ways.”

A spokesperson for forwarder Navia told The Loadstar the company used AI for accounts payable invoices (APIs), which they said was an “ideal choice”.

“Our invoice processing is part of our business that has crisp, defined rules associated with that data entry, which makes it a good proving ground for something like Raft,” they told The Loadstar.

“Before, our operations team manually entered data into our freight management systems to process APIs. This would inevitably result in discrepancies and errors. Some were made in reimbursements, which would mean unnecessary costs for Navia.

“Our finance team had to then reconcile all of the errors, meaning even more time spent on manual data entry.”

Indeed, a spokesperson for logistics company, and Raft customer, Scarbrough Global told The Loadstar “financial processes” were “the easiest” place to start with AI integration.

“A customer pays you a cheque for an amount of money; you have 10 open invoices and they paid three of them. Which three did they pay? A human could go through and add each one up individually, where a [AI] system should be able to do that in a matter of seconds,” they said.

Scarbrough Global CEO, Adam Hill added: “Our finance team has been the same size or smaller for the past five years, and our business is about two and a half times bigger. So, if you look at it from that perspective, it has been massively impactful.

“It has already helped us by saving time on manual data entry and preventing errors, which has allowed us to re-allocate our resources to higher value tasks.”

Other uses for AI could include statement reconciliation, shipment registration, route optimisation, demand forecasting or automated warehousing, but said Ms Hermenjard, the best place to introduce AI to a company is at the specific “pain point” it faces.

You can hear the whole podcast on how AI will revolutionise logistics here.