We’ve been working with Visy’s New Zealand operation on something that sounds simple but gets complicated fast.
They have trucks moving all sorts of stuff. Some of it weighs a couple of hundred kilos. Some of it weighs several tonnes. And they have a rule: heavy items cannot be placed on top of light items. Makes sense. You don’t want a two-tonne pallet crushing a lighter one.
The catch is that this rule has to hold throughout the entire route. Not just at the start. At every point during the journey, the load ordering has to be right. Heavy items get loaded first and unloaded last. Light items go on top and come off first.
When the Solver Doesn’t Respect the Real World
Our original schedule optimiser wasn’t respecting this properly. It would produce efficient routes, but the load ordering sometimes didn’t work. And when you’re dealing with restricted access at delivery sites, that’s a problem. The driver shows up and can’t unload in the right order.
This isn’t just an inconvenience. When a truck arrives at a delivery point and the heavy items are sitting on top, the driver has to either unload everything to get to what’s underneath, or call back to the depot and ask what to do. Neither option is good. Both waste time and create frustration.
Visy’s drivers shouldn’t have to be amateur load planners. They should be able to show up, unload, and move on.
So we went back and added doubled load ordering options to the DTO. The solver now handles the constraint properly, even with the exponential growth in possibilities that comes with multiple drops.
Why This Matters
Load ordering constraints are common in transport. Different industries have different rules. Sometimes it’s about weight. Sometimes it’s about fragility. Sometimes it’s about delivery sequence. The principle is the same: the order in which things are loaded determines what’s possible at each stop.
When your solver ignores these constraints, you end up with schedules that look good on a screen but fall apart on the ground. Drivers get frustrated. Turnarounds take longer. Customers wait.
When your solver handles them properly, the opposite happens. Routes get executed as planned. Drivers stay happy. Productive time goes up.
The Result
Visy NZ gets routes that work in the real world, not just on a screen. Their drivers can unload without having to rearrange the whole truck at each stop, which means faster turnarounds and less frustration on the ground.
That’s the kind of fix that makes a real difference to the people doing the work every day. It’s not flashy. But it’s the difference between a plan that works and a plan that doesn’t.
