Weighing the pros and cons of artificial intelligence in international shipping and forwarding
 From Melbourne's iT Wire comes a report that first appears to be expository before becoming promotional about Artificial Intelligence (AI) in container tracking.
Even before its manipulative nature is revealed, one wonders what real advantage AI offers in container tracking that isn't already available - RFID (radio frequency identification) comes to mind. And that's been around for 20 years. Despite, the universal ballyhoo about AI these days, in this examination, it continues to be much ado about not very much.
But you be the judge. Let's start off with the iT Wire's conclusion: "The use of artificial intelligence in tracking shipping containers represents a transformative shift in the logistics industry, offering unparalleled levels of accuracy, efficiency and predictive capabilities.
"From predictive maintenance and route optimisation to cargo monitoring and risk management, AI-powered tracking systems are revolutionising the way containers are monitored and managed throughout their journey. As technology continues to advance and new innovations emerge, the future of container tracking holds immense promise for enhancing supply chain visibility, efficiency and sustainability on a global scale," it concluded.
The theme of "holding promise for the future is often repeated in many contexts. One is also struck by the fact that that so much of the AI tracking data is useless to the shipper because it is beyond his control as a shipper, such as route selection that is in the hands of ever-fewer market segmenting shipping lines, with one differing little from another in this regard.
The article begins cheerily enough: "Central to this process is the tracking of shipping containers, ensuring timely delivery, minimising losses and optimising supply chain operations. With the advent of artificial intelligence (AI), tracking systems have undergone a transformation, offering unprecedented levels of accuracy, efficiency and predictive capabilities.
"Traditionally, tracking shipping containers relied on manual methods such as barcodes, RFID tags and GPS devices. While effective to some extent, these methods had limitations in terms of accuracy, coverage and real-time visibility. With the rise of AI and machine learning technologies, tracking systems have evolved to incorporate advanced data analytics, predictive modelling and automation capabilities, revolutionising the way containers are monitored and managed throughout their journey."
One doubts the seemingly sound promise of AI-powered route optimisation. These provide algorithmic leverage and real-time data on weather conditions, traffic patterns, port congestion and other variables to identify the most efficient and cost-effective shipping routes. But how meaningful is this? Is a shipper to accept as gospel the weather conditions affecting his consignment on every day it passes under that patches of sky of it transoceanic multi-sea voyage? Will this, as claimed, minimise transit times and fuel consumption, or will it, as AI route optimisation promises, streamline supply chains and produce cost savings.
It might be said that any forwarder worth his salt will already know his options and make a better choice than an AI programme working on old and incomplete data. At times, a shipper's sense of the particular conditions at a particular time in a particular peak season might decide to move Savannah-bound Vietnamese cargo west through Suez - or, alternately, east through Panama.
Much as one respects Elon Musk and Dr Jordan Peterson, it is difficult to share their dread of the risks that AI posed to the human race. Mr Musk's vision of hundreds of square miles of desert covered with solar panels seems costly and unrealistic. Dr Peterson's favourable assessment of an AI-generated graduate paper proves nothing, as it based on previously published information presented in a new and original way.
The Shipping Gazette has deployed AI in a number of articles and found it wanting. While polished, the writing was lifeless, some of it out of date and in one or two cases simply wrong. AI, however, did produce two or three recyclable sentences, and presented new lines of enquiry. Which one suspects will be the case when applied to logistics. Not worthless, but hardly game changing or revolutionary.
In the area of cargo monitoring, much is promised, but much that has already been promised and delivered 20 years ago by RFID. "This data allows logistics companies to ensure the integrity and quality of perishable goods, pharmaceuticals and other sensitive cargo throughout the shipping process," it is claimed. Is that true? Does it really "ensure the integrity and quality of perishable goods", or just allow the shipper to worry helplessly about it?
Once such consignments get to the marine terminal, aboard a ship, and are shifted about in a container yard, it's pretty much a crap shoot of whether it arrives in good shape. Of course, supply chains have much improved over 20 years, and little goes wrong these days. But the point is that knowing what goes wrong when it goes wrong provides little value unless one can make good losses in time before one displeases a customer.
The pitch goes on: "AI analytics platforms analyze historical data, market trends, geopolitical events and other factors to assess and mitigate risks associated with shipping operations. By identifying potential risks such as port disruptions, container theft or trade disputes, risk management systems enable proactive decision-making and contingency planning to safeguard cargo and minimise losses."
And on: "AI algorithms analyse transactional data, shipping documents and other sources of information to detect fraudulent activities such as invoice fraud, cargo theft or customs violations. By flagging suspicious patterns and anomalies, fraud detection systems help prevent financial losses and protect the integrity of the supply chain."
Sixty years ago, a popular TV show, "Here come the Seventies" predicted family helicopters, medical marvels and other exciting futuristic advances that did not come to pass, or have only done so in recent years.
Today's freight forwarder and beneficial cargo owner (BCO) will likely depend on their own savvy for what's best for consignments in their charge than placing any reliance of an artificial intelligence technology.
It would be far better to keep and eye out and be on guard against its present applications in dystopian realms of social credit as practiced in China to enforce the edicts of a police state. AI's threat is mostly regulatory - not technological. |
