A.I. Could ‘Remove All Human Touchpoints’ in Supply Chains. Here’s What That Means

Artificial intelligence is likely to shake up the transportation industry — transforming how supply chains are managed and reducing the number of jobs carried out by people, according to analysts and industry insiders.

Sidewalk robots, self-driving trucks and customer service bots are on their way, along with generative AI that can predict disruptions or explain why sales forecasts may have been missed, according to industry executives.

“AI may be able to totally (or nearly) remove all human touchpoints in the supply chain including ‘back office’ tasks,” Morgan Stanley’s analysts led by Ravi Shanker stated in a research note last month.

“The Freight Transportation space is on the cusp of a generational shift driven by disruptive technologies incl. Autonomous, EV, blockchain and drones. AI is the latest one of these potentially transformative technologies to emerge – and perhaps the most powerful to-date,” the analysts added.

For example, the bank said it expects several hundred autonomous trucks to begin operations in the U.S. in 2024, reducing the cost-per-mile by 25% to 30%, and eventually eliminating the need for drivers entirely (its timescale for this is “beyond three years”).

No more empty shelves?

Supply chains are often long and multifaceted: A company might source from manufacturers in different parts of the world, with components shipped to a central assembly plant before goods are distributed to customers globally.

Producing and transporting goods, already a complex process, was disrupted by the Covid-19 pandemic and the Russia-Ukraine war — which led to a shortage of components such as computer chips and the rerouting of shipments. That complexity means companies are often unaware of what happens to their products from one end of the process to the other.

“This is where AI (and machine learning) come in. By predicting what could go wrong with a fluid Transportation network … before it does, AI/ML systems could … potentially even avoid the disruption scenario entirely,” Morgan Stanley’s analysts added.

This is a theme picked up by analysts at investment firm Jefferies, who made multiple predictions about the effect that generative AI will have on transportation and logistics. That includes forecasting demand, predicting when trucks need maintenance, working out optimal shipping routes and tracking shipments in real time.

“A shortage of truck drivers, polar vortexes halting interstate commerce, and a dearth of baby formula on grocery store shelves will be a distant memory with the adoption of generative AI in the Trucking & Logistics space,” its analysts, led by Stephanie Moore, wrote in a research note published on June 6.

Going mainstream

Generative AI will be a big part of shipping giant Maersk’s operations, said its chief technology and information officer, Navneet Kapoor.

“AI and machine learning, they’ve existed for a very long time … Over the years, it has progressed from being interesting research projects to more ‘real’ projects within companies … And now, with the advent of generative AI … we have a real pivoting opportunity to take AI mainstream,” Kapoor told CNBC by phone.

Maersk has used AI for several years and is now “pursuing aggressively” ways to integrate it into its business processes and functions on a larger scale, Kapoor said. One way it is already being used is to help customers plan better.

“We are using AI to build what we call a predictive cargo arrival model to improve scheduled reliability for our customers … Reliability is a big deal, even post pandemic, so that they can plan their supply chain, their inventories better, and bring their costs down,” Kapoor said.

Maersk also wants to use AI to recommend solutions when shipping routes are congested, advising on whether goods should be flown or stored, for example. And, Kapoor said, the company wants to use a type of generative AI known as a large language model — which learns how to recognize, summarize and generate text and other types of content from vast amounts of data — to understand the sales process better.

“You can get a full view of all the transactions the customer has done with you in the last year, you can figure out the root causes of why [for example] you might lose deals in a certain business area,” Kapoor said.

Double-edged sword

And what of potential job losses?

“Generative AI, in my mind is, [a] once in a lifetime kind of disruption that’s going to happen … so there are going to be losses of jobs in the more traditional setting, but I also believe it’s going to create new jobs like every prior technology disruption has,” Kapoor said, adding that roles such as prompt engineers (people who train AI to give better responses) are likely to be more in demand.

One threat noted by Morgan Stanley is from “high tech digital entrants” to the industry, with analysts describing a double-edged sword for transportation companies: AI might help them become more efficient, but it could also reduce the need for services from the third-party logistics firms that organize packing, storage and shipping.

Maersk has invested in AI startups via its Maersk Growth venture arm, including Einride, a self-driving electric truck manufacturer; Pactum, a company that automates sales negotiations; and 7bridges, an AI platform that helps companies see where their stock is and anticipate delays.

“We look at [data startups] as definitely an enabler for our transformation, and an accelerator, but we are also watchful: we don’t want to be caught napping on this one … Data start-ups can be [an] intermediary between us and the customer and we need to make sure that we are staying ahead of the curve, but also learning from them,” Kapoor said.

Knowledge assistants

“Knowledge assistants” can help with another problem: the over- and under-ordering of goods, according to Igor Rikalo, president and chief operating officer of software company o9 Solutions, which helps firms centralize and analyze data. That’s often the result of a lack of communication between internal teams, with sales departments placing orders separately from those who work in supply chain management, he said.

“It’s a sub-optimal result, because sales [teams] might be investing into promoting the items that a supply chain is constrained on, so you’re wasting money,” Rikalo told CNBC by phone.

“We see a world where hopefully, every one of us will have what we call knowledge assistants that are powered by these AI, by these large language models,” he added, with such assistants being able to give insights into why a supplier has delivered less than what was ordered, for example.

Answering those questions usually requires input from sales, marketing, supply chain and procurement teams, but generative AI might be able to examine large data sets to provide answers.

It may also mean fewer people are needed in integrated business planning teams, which oversee long-term goals, revenue projections and forecast demand for particular products.

“A 1,000-person planning function today can probably be transformed to 100 people or less,” Rikalo said.

Written by:  Lucy Handley, Writer, and CNBC’s Cheyenne DeVon and Jonathan Vanian contributed to this report, for CNBC.