Klau - Roll-Off Dispatch Planning

When I explain container chaining to dispatchers, I don't just show them job sequences like "delivery → removal → delivery." That looks the same as "removal → delivery → delivery"—what's the difference? The magic becomes visible when you show every step of every job.

Making the Work Visible

Instead of thinking in jobs, think in steps:

Load empty container at yard
Drive to customer site
Unload empty container
Drive to next customer
Load full container
Drive to dump site
Dump container
Drive to next job
Unload empty container

When you see all the work—not just the job names—it becomes obvious where the efficiency is. Every step that can be eliminated is time saved. Every container swap avoided is 20-30 minutes back in your day.

Perfect Chains: The Goal, Not the Guarantee

Will you achieve perfect chains every day? No. Perfect chains happen occasionally, but they won't be the standard—at least not yet. The real win is "better than before."

What we celebrate is the most efficient and most safe groupings of jobs possible, along with excellent visibility into the real work happening. When dispatchers can see every step, they make better decisions. When software handles the heavy lifting, dispatchers have time to make those decisions.

Down the road, I see an opportunity to use pricing leverage on customer self-service flows—incentivizing customers to schedule in ways that fill in perfect chains. Once customers are actively helping optimize their own schedules, perfect chains could become the standard. But that's the future. For now, we focus on making every day better than it would have been.

Beyond Container Size: Dump Site Timing

Container size isn't the only compatibility factor. Dump site availability is huge—and often overlooked.

Some materials have to go to specific sites. Those sites may be close or far, and they have operating hours that require careful planning. When I was driving roll-off, I had a couple super frustrating instances where I got caught in traffic and arrived at the dump site just minutes after they'd stopped accepting loads for the day.

Now I had a full container on my truck that I needed to set off full to finish my day. Not the end of the world, but it impacts the schedule and means work carried over to the next day unexpectedly. Good optimization accounts for these constraints before they become problems.

The Human + Algorithm Balance

There's always going to be special circumstances that aren't modeled well in data or algorithms. The owner's buddy wants a container removed first thing tomorrow morning. Instead of leaving the yard with an empty to deliver before doing the removal, you schedule the "imperfect" chain.

The key is having this flexibility without extra frustration. Mark the job as "pinned"—locked in place—and optimize the rest of the day around it. The algorithm handles the complexity; the dispatcher handles the relationships.

This is what "augment, not replace" looks like in practice. Software does what software does well (evaluating thousands of sequences in seconds). Humans do what humans do well (knowing that this particular customer needs special handling today).

Building Flexibility Intelligence

Smart carry-over and smart chaining both require knowing which customers are flexible. In our early stages, this intelligence comes from what the dispatcher already knows—marking jobs as "flexible scheduling" based on their relationship with the customer.

In later stages, when we have customer self-service, we can offer flexibility in multiple ways: generic flexibility on the account ("yes, I'm flexible with scheduling") and per-job flexibility potentially driven by discounts or pricing incentives. Customers who help us optimize get better rates. Everyone wins.

The Bottom Line

Container optimization isn't about achieving perfection. It's about making the invisible visible, eliminating unnecessary steps, and giving dispatchers the tools to make smart decisions without spending hours on manual planning. The industry has accepted inefficiency as inevitable for too long. It's not inevitable—it's just unseen.