Why density changes the dispatch math

Not all snow removal is the same problem. A contractor servicing farm driveways spread across a county and a contractor servicing eighty driveways packed into a few blocks are both "doing snow removal," but the math underneath their operations could not be more different. Understanding that difference is the key to understanding when dynamic dispatch is a marginal improvement and when it is a decisive one.

Two regimes: travel-bound and decision-bound

Picture the sparse operation first. Properties are far apart; the dominant cost is the drive between them. In this travel-bound regime, the sequence you visit properties in barely matters once it is roughly sensible, because the long drives swamp everything else. A route planned the night before is nearly optimal, and re-optimizing on the fly buys you very little — there is not much to gain because the drives are fixed and long.

Now picture the dense operation. Properties are close together; drives are short. Here the dominant variable is not travel — it is the decision about what to do next. This is the decision-bound regime, and it behaves completely differently. Because re-deciding the next stop is nearly free (the next property is right there), the fleet can constantly rebalance toward whatever is most urgent or best-positioned. Small, frequent reallocations that would be pointless in the sparse case compound into real coverage gains.

On sparse work the drives dominate and the plan changes slowly. On dense work the decisions dominate and the plan should change constantly. Dynamic dispatch is built for the second world.

Why density amplifies the swarm advantage

There are a few reinforcing reasons dense clusters favor continuous dispatch.

The cost of re-deciding is low

When any vehicle is a few hundred metres from several open properties, changing your mind about which to service next costs almost nothing. The fleet can chase urgency — the property closest to breaching its window — without paying a travel penalty for the flexibility.

Idle and overlap time shrink

The classic waste in a dense operation on fixed routes is two machines working near each other in a way that made sense on paper but wastes coverage in practice, while a third area waits. Continuous claiming, with each machine always taking its highest-value next property, squeezes out that idle-and-overlap time — which is precisely the slack that density creates and fixed routes cannot reclaim.

Density cohesion beats ping-ponging

The counter-risk of "always take the best property" is that vehicles scatter and ping-pong across town. A well-designed swarm guards against this: it sweeps a dense block before leaving it and discourages a vehicle from grabbing a driveway right next to a peer already working there. The result is tight, coherent coverage of a cluster rather than a scramble.

The practical implication

If your work is genuinely high-density, the question is not whether dynamic dispatch helps — it is how much coverage you are leaving on the table with fixed routes. The honest way to find out is to simulate your real cluster at different vehicle counts and compare. Density is the condition under which fewer machines can cover more properties on time, and it is exactly the condition a fixed list cannot exploit.

Key takeaways

• Sparse work is travel-bound — sequence barely matters, so fixed routes are nearly fine.
• Dense work is decision-bound — the "what's next" decision dominates, and re-deciding is nearly free.
• Density amplifies the swarm because it lowers the cost of re-deciding and squeezes out idle-and-overlap time.
• Good swarm logic keeps dense coverage coherent (sweep the block, avoid stealing a neighbor's property) instead of ping-ponging.