Here’s an innovation that’s going to be popular at container terminals. It’s a fixed set of frames allowing containers to be stacked in individual pigeonholes. They’re placed and removed by stacker cranes running through the aisles. The system has been in development and testing since 2018.
Hardware innovations take a long time to develop, and they require a place for testing. The partnership of DP World and SMS Group, a German firm, combined the expertise and the need and test bed to create the product.
Stacking density may be improved up to 4 times using the Boxbay system.
This is not a new idea. Auto manufacturers and shippers have been using such arrays to store cars since the 90’s. I was shown a picture of one in Japan by Ernest Konigsberg, a Berkeley operations research professor who was familiar with the design and the optimizing software written to decide which locations to place vehicles. We would call it AI today, but then it was simply optimization software. It’s been around a long time.
An interesting question is why this technology is only emerging now. One answer is the Great Congestion about the time of the COVID epidemic. Yard storage was a significant problem during the supply chain crisis. This kind of system can improve the utilization of scarce container yard land. It’s a natural type of tech to invest in.
I’m sure we will see more such systems if container shipping demand comes back and exceeds the congestion period levels. But if ports aren’t handling so many containers, they may not be so eager to invest in this technology. I don’t see the large US ports jumping on this so soon, with traffic falling.
The article has a nice picture of Pusan Newport terminal, and judging by the stacking disarray in the right part of the picture, they can use this system1
Here we see another bet on ammonia power. One partner is the Swiss firm WinGD, which has a long history of providing power for merchant shipping. Their website gives no indication that they have a hydrogen or ammonia-powered engine yet. But they do offer dual-fuel engines that will utilize gas for fuel. Perhaps they are not so far away.
Another development partner, CMB.TECH, a clean technology private firm based in Antwerp, already has some design expertise for ammonia engines. A look at their site shows some hydrogen-powered vessels operating, and some ammonia-powered vessels on order. They also have some hydrogen-powered fixed engines and other machines such as an excavator that they say can be ordered today.
Handling of hydrogen or ammonia on board and at ports is a major concern with this technology. Watch for the designs to see how difficult this will be.
Lora Cecere, the Supply Chain Shaman, has given us once again something to think about– big time. She goes off on some of the current fads in supply chain that she thinks are not worth pursuing.
It’s always wise to listen to Lora. She has a wealth of experience and many years of consulting with top companies to inform her thinking.
Here are some of her dead-end streets that you should be avoiding.
Dashboards
Lights-out planning
Real-time planning
DDMRP (demand-driven material requirements planning)
Forecast Sharing
Sales Forecasting
She believes future applications are adaptive and distributed. Read the article to see what she means, and how those two keywords play out.
The SCOR methodology she claims is useful in devising a new approach to planning. Figure 3 shows an outside-in model of planning for a company.
I’m particularly intrigued by her comments on dashboards. I’ve felt they were more about visibility than improving processes. While some of that is necessary, the real value will be if you can make changes quickly. And most dashboards, I’m afraid, don’t let the operators do that. They look cool but don’t help making critical changes.
Her remarks about real-time planning are also on point. If you rerun too frequently, the plans thrash— they oscillate between one action and another completely different one. There is value in following a consistent pattern and making slow changes rather than swinging back and forth. We saw that in spades during the Covid epidemic. Look at the auto manufacturers who canceled their chip orders and then six months later could not place them again because the capacity had been diverted to other chips that were more profitable as it happened. Acting too quickly caused them anguish that has persisted for three years now. Reruns generate noise that affects partners and destroys relationships.
I also share her skepticism about DDMRP which bases an entire manufacturing progression on orders. It’s better than not looking at orders, but it fails to capture the richness of what might happen in the future. We can point to the same auto manufacturers and see that they have hundreds of cars in an ‘almost-complete’ state while they wait for certain electronics to be delivered.
And she provides some good references on checking how forecasting is improved by getting the customer’s forecast in advance. Statistical tests indicate that for the most part they are not helpful. What characterizes the few that are? That’s worth research.
And finally, salesmen are the worst forecasters. First of all, they’re liars. And the best ones are the worst liars. Why? Especially if they are paid on commission, or via a plan that pays for meeting targets or provides spiffs for beating them, they will try to get the target set low. And for commission sales often they are not guaranteed, as to size especially. So the salesman might predict low-ball, because he’s sure he will get that as a target. Then the big order comes as an extra. But for manufacturing planning, that is not reality. Often big orders do not come exactly when anticipated, so they might not fall at quarter end when the financials are due. Timing is harder to predict than size.
In the past at a mid-size manufacturer, we had better luck eyeballing a steady rate of production from history (moderated by some anticipation of the future market) and then trying to get salespeople to tell us about big orders and when they anticipated them. We laid the big order forecasts on top of the ‘run-rate’ forecast to determine the timing of production. A few days’ change in big order timing didn’t then blow our whole manufacturing plan.
Salespeople are a good source of info about products and customers, but as manufacturing forecasters they are awful. Yet you need to cultivate them to get that key information about their expectations which is valuable.
Supply Chain and Logistics 