How AI can solve supply chain shortages and save Christmas


Covid-19 shines a spotlight on many networks around the world, from the internet in international air travel. But the supply chains that crisscross the world — the ships and trucks and trains that connect factories to ports and warehouses, carry almost everything we buy many thousands of miles from where. where it is made to the point where it is exhausted — facing more scrutiny than ever before.

“It’s fair to say that whatever you’re selling, there’s a problem right now,” said Jason Boyce, founder and CEO of Avenue7Media, a consulting firm that advises Amazon’s leading sellers. Boyce says he has clients who will turn over ten million dollars a year if they stay in stock. “We talk to clients every day that they just cry,” he says. “For months, they haven’t been fully in stock for a 30 day in a row.”

Digital twins seek to solve supply chain disruptions by anticipating them before they happen and then using AI to figure out a solution. The name takes the key idea of ​​simulating a complex system on a computer, creating a class twin that mirrors real-world objects — from ports to products — and the processes that about it. Simulations have been part of industry decision-making for a number of years, helping people explore different product designs or streamline the layout of a warehouse. But the availability of a lot of real-time data and computing power means that even more complex processes can be simulated for the first time, including the turmoil in the world’s supply chains that often rely on multiple sources. vendors and transportation networks.

This type of technology has given Amazon, which has the advantage of controlling its own trucks and warehouses, an added bonus for many years to come. Now some are accepting it again. Google is developing supply-chain digital twins that auto maker Renault announced it would start using in September. International shipping giants such as FedEx and DHL have built their own simulation software. And AI companies like Pathmind are creating special tools for anyone who can afford it. But not everyone will benefit. In fact, powerful new technology could widen the growing digital divide in the global economy.

Storm season

It is easy to blame pandemic on current supply-chain problems. Factory closure and lack of labor knocked out production and delivery hubs at the same time that a leap in online shopping and comfort buying sent the demand for home delivery rocketing.

But in reality, the pandemic only exacerbated the bad situation. “There are global forces driving it, all combined into one complete storm,” said D’Maris Coffman, an economist at University College London who studies the impact of pandemics on supply chains.

Stopping this storm will require sinking trillions of dollars into global infrastructure, expanding ports and delivery vessels, and investing in better governance, better working conditions, and better trade deals. “Technology can’t solve these problems. It won’t allow ships to carry a lot of containers,” said David Simchi-Levi, who heads the data science lab at the Massachusetts Institute of Technology and helped create digital twins. for a lot of big companies. But AI can help companies deal with the worst of it. “The digital twin allows us to figure out issues before they happen,” he said.

According to Hans Thalbauer, managing director of Google’s supply chains and logistics team, the biggest problem businesses face is the inability to predict chain events. “It doesn’t matter which company you talk to,” he said. “Everyone in the supply-chain world will tell you they don’t see what they need to make decisions”

It’s supply-chain visibility that allows Amazon, for example, to predict when something will show up at your door. For every item Amazon provides on its own — and that includes the millions of items it provides for third-party vendors like Boyce and his clients — it gives an accurate estimate of when -a it will come. It may not be so much, says Boyce, but if Amazon is wrong in these predictions, it will start to lose customers — especially during the holiday season, when people buy last-minute gifts and rely on Amazon to deliver it. “It takes a lot of computing power just to show the simple little delivery day,” he said. “But people are afraid of hell if they don’t get their things on time.”

According to Deliverr, a U.S. company that handles shipping logistics for many e-commerce companies including Amazon, Walmart, eBay, and Shopify, an estimated shipping time is two days compared to seven. up to 10 -day sales increase of 40%; the estimated shipping time of one day increases sales by 70%.

No wonder some want a crystal ball of their own. Supply chains at the right time are about to die. The disruptions of the past two years have plunged many businesses into the pursuit of over -efficiency. Warehouse space is expensive, and paying to store inventory you don’t need for a week can be just as expensive at most times. But if next week’s stock doesn’t show, you’re not selling.

“Before the pandemic, most companies focused on cutting costs,” Simchi-Levi said. Now they are willing to pay for strength, but focusing only on strength is also a mistake: you have to find the right balance between the two. This is the real power of simulations. “We’re seeing more and more companies starting to stress-test their supply chains using digital twins,” he said.

What if?

By exploring a variety of possible scenarios, companies can determine the balance between efficiency and robustness that is best for them. Add in-depth learning to reinforcement, allowing an AI to learn by trial and error what actions to take in different situations, and digital twins become machines for exploration. on what-if questions. What if there was a drought in Taiwan and water shortages stopped microchip production? A digital twin can predict the risk of this happening, track its impact on your supply chain, and — using reinforcement learning — suggest what actions to take to minimize damage.

If you are a U.S. Midwest auto manufacturer, a digital twin may suggest that you purchase additional components from a West Coast distributor with a surplus. But combining multiple scenarios and things can soon get even more complicated. For example, according to Simchi-Levi, Ford maintains more than 50 plants worldwide, using 35 billion parts to produce 6 million cars and trucks each year. It has nearly 1,400 suppliers spread across 4,400 manufacturing sites directly associated with it, and a stack of suppliers and suppliers of up to 10 layers in depth between Ford and the raw materials contained in its vehicles. Any links can be broken, and a good stress test should check each of them.



Source link

admin

Leave a Reply

Your email address will not be published. Required fields are marked *