Originally Posted by RPZ
They slide off because there is nothing binding the top layer. It wouldn't take any engineering genius to design clamping braces that could be lowered on to the top layers holding them together, and remove before offloading the containers.
Otherwise containers could be redesigned to interlock horizontally as well as vertically, old ones phased out.
Other option would be to require container ships to have a containment structure integral with the hull and deck
structure the full height of a maximum load.
They don't simply slide off. Twist locks are at all four corners. For each container to container interaction, they lock in just as tight as the spreader bar used on the cranes to lift
them on and off the ship. You can find photos of containers almost 90 deg to horizontal and still being held on by the twistlocks. Granted, the weight of the entire box and cargo will have a big role in that. The lowest two tiers above deck (or cell guides, if fitted) are also lashed with crossbars and tightened with turnbuckles (to varying degree). The point of failure tends to be at twist lock connections right above the uppermost lashed container. The containers above that one can exert quite a lever arm and once they get moving side to side, or knocked into from an adjacent wayward stack, enormous forces are exerted on a relatively small connection that subsequently fails. It's rare for a single
box at the top of the stack to simply be washed over - unless it was never locked in properly in the first place.
Not all containers have equal heights and stack heights vary for many reasons so it's not as simple to just say to throw a truss over each stack. Early container ships did have lashings alongside the upper tiers that were intended to prevent sway as the ship rolled, but they fell out of favor at some point, probably as twist locks became more reliable. They would also be labor intensive to set and remove.
Intermodal containers are very much designed around their time spent at sea. No other mode, for example, utilizes the lashing points on the rear and sides of the corners - those are solely for lashing bars on ships. Rail and trucks just use twist locks. When packing a container that's going to sea, you pack it to protect the contents during that part of the voyage - it's the most dynamic and usually the longest.
Your last suggestion (higher cell guides) is the most realistic but won't come about without change in regulation (ie, everyone has to play by the same rules so no one loses a cost advantage since container shipping is extremely competitive, even among other kinds of shipping). These guides do allow the lashings to get to tiers higher up, which would exert less leverage on the entire stack as the ship rolls. However, higher cell guides is more steel
weight above G, which turns into less cargo weight that can be carried up there. You're also then imposing various container size restrictions within the cell guides (there are containers longer than 40's and they must go on deck at least one tier above the hatches or cell guides).
The industry (at least those of us who actually work on ships) is aware of this and it's been happening a lot more lately as it seems. It's frustrating for us as well. Cargo lost overboard is bad for other mariners, can be bad for the environment
, usually damages the ship and of course is bad for our customers. Until ship designs change, and in my view it will have to happen through regulation at the IMO level, all the crews can really do is check & tighten lashings and avoid the worst of the weather
as best as possible/practical.