Two weeks ago, I wrote about how complex the world was, and I quoted an article about “an almost unfathomably complex, algorithmically calibrated, partly automated … global supply chain.” Well, let me tell you, I’ve been inundated with an email asking about this.

However, that presents a problem: I’ve spent much of my career working as a data scientist, and it’s hard to talk about algorithms and make it fun. They say that economists know the price of everything and the value of nothing. But that’s not true, as I know the value of explaining the algorithmic Supply Chain to you and the price of making it dull. Walking the tightrope of writing in detail and making it funny is mission impossible. But when faced with mission impossible, Tom Cruise didn’t quit, and neither shall I*.

No discussion of this topic can begin without looking at Moore’s Law, which you’re likely familiar with. It states that the number of transistors on an integrated circuit doubles about every two years. I’ve previously argued that the semiconductor is one of the greatest technological achievements of all time.

At the very same time, the cost of data storage has dramatically decreased.

That these two phenomenons have happened together is crazy. Of course, it hasn’t just happened. It’s the immense effort of clever people in a system which enables this progress. We should never take this for granted. 

For more than a long time, back to Babbage’s difference engine, through to the Turing machine and Konrad Zuse’s inventions, above all else, computers have been used for calculating. The automated and Algorithmic Supply Chain is nothing more than millions upon millions of calculations, whose output is passed on to the next stage of the supply chain. This has been made possible by the growing power and the reduced cost of computing.

To see how this looks, we need to go to the east coast of China and visit Qingdao port. Billed as the world's first fully automated port, only Rotterdam is a rival in automation. China is at the heart of the world’s global supply chain: in 2021, as a percentage of global GDP, China accounted for 27.4% of manufacturing value added. At Qingdao, goods arrive on the back of a truck, the container is lifted off and … that’s where humans stop for a little while.

At Qingdao port, we have an impressive combination of networking (5G), software (optimisation algorithms) and hardware (cranes, robotics and containers). Simultaneously, hundreds of containers are loaded and offloaded, coexisting harmoniously thanks to path planning algorithms. A few months ago, I was outside Singapore port, and let me tell you, the scale is vast! You have to see it to appreciate it. Then it’s onto the container ship, stored using container stack arrangement algorithms, like one big game of Tetris.

Apart from the Mary Celeste, ships that can sail themselves have yet to arrive. Or have they? GPS autopilot navigation is the modern enterprising captain of the seas. The ship may speed up or slow down to reduce costs, reduce waiting time, or observe the beauty of Mauritius as it passes by (okay, I made that one up); all determined by a hodgepodge of satellites, weather reports, fuel prices, forecasting algorithms and crew availability. Every aspect of the journey is continuously optimised and adjusted.

Once the container arrives at its end destination, it’s put on a train, and the goods make their way onto a van for delivery. Thirty years ago, Bob, the van driver, chose the route. Vehicle routing algorithms now determine the order of parcels to be delivered.

All of what I’ve just described is operations research, statistics and technology, rebranded as artificial intelligence because AI gets VC money. Whatever it’s called, here’s why you should care: if you are a buyer, you can access more goods, more quickly, than ever before. If you are a seller, you can access a wider market than ever before - your market is the entire world (except North Korea). If you are an investor, the complexity of the world creates new markets. There are opportunities for new businesses and investment. For the industrialist, it’s more vertical integration and conglomerate building opportunities. If you’re a geek, this stuff is cool. And for all of us, the world is closer together.

The reality of the modern industrial economy is a million miles away from common perception. While many have a quaint view, the reality is a technological, algorithmic network operating incredibly efficiently, connecting 800 ports and 5,500 container ships transporting 800 million twenty-foot equivalent units (TEUs) a year. If you doubted the importance of global supply chains, consider this: disruptions in 2021 knocked between 0.5 and 1.0% off global economic growth. Ouch.

The BBC summed it up well: “The fact that it's hidden from most people's sight, and that it has become so utterly reliable and efficient to the point of transparency, doesn't make it any less of an achievement of human technical endeavour.”

*If you want to know how the sequels fit into this, you’ll have to work that out yourself.