What capacity actually means when the physical building does not change There was a fulfillment building designed around a maximum capacity of roughly 16,000 units per day. On our best high-volume event, we broke 60,000. Normal days eventually became greater than 25,000 and usually greater than 30,000.

The building did not become four times larger. We did not install four times as many dock doors. We did not create four times as much floor space. People did not simply work four times harder.

That is what makes the number useful. It forces a better question: What is capacity?

THE BUILDING IS NOT THE NUMBER

Capacity is often treated as if it were a permanent physical property. This machine can process X. This building can hold Y. This department can produce Z.

There are certainly hard physical limits. A conveyor has a maximum speed. A trailer has a fixed cube. A dock has a finite number of doors.

A staging area occupies real square footage. Only so many people can safely work in the same physical space. But the practical capacity of an operation is not determined by one limit. It is determined by the relationship between many limits.

Flow. Timing. Space. Equipment.

Labor. Information. Routing. Sequencing.

Container availability. Door availability. Downstream capacity. The system reaches its practical limit when one of those relationships prevents additional useful work.

Change the relationship and the practical capacity can change dramatically even though the building does not.

THE FIRST SYSTEM FIT ON NOTEBOOK PAPER

I still have a handwritten CPT chase list from ALB1. Four departures were written on it: 9:00 p.m. 10:15 p.m.

2:30 a.m. 5:00 a.m. The notes were simple. Start grabbing CPT freight an hour in advance.

Re-induct before cutoff. Close out unused cells containing CPT freight. Load dwelling CPT cages. Help where the operation is backed up.

That page is useful because it captures an early version of the operating system. CPT means Critical Pull Time. A CPT is a deadline, but managing the deadline at the deadline is already too late. A 9:00 departure is not a 9:00 problem.

It is an 8:00 problem. And a 7:30 problem. And potentially a problem created much earlier when freight was routed, staged, inducted or allowed to dwell somewhere it should not have. Once we started organizing around what had to become true before the deadline arrived, rather than reacting when the deadline was already visible, the operation changed.

The paper was not sophisticated. The logic was.

FROM FOUR DEPARTURES TO TWENTY-ONE

The operation eventually went from that handful of CPTs to 21 CPT departures in a day, plus sweeper loads. That is not merely a scheduling change. Every additional departure interacts with the physical dock. Doors.

Trailers. Freight. Staging. People.

Equipment. Container state. Yard movement. A departure consumes resources before the trailer ever moves.

If the freight arrives at the door too early, it can occupy space needed for something else. If it arrives too late, the departure is at risk. If a door is assigned poorly, the wrong freight may have to travel farther. If a container is allowed to dwell, the space remains occupied and the future workload grows.

If a trailer is not available, the plan changes. If one process gets ahead of another, the faster process may simply create a larger queue. The system therefore cannot be improved by maximizing each component independently. The components have to operate as a sequence.

CAPACITY HIDES IN DWELL

One of the easiest ways to lose capacity is to let work stop without noticing that stopped work is still consuming the system. A dwelling cage is not neutral. It occupies space. It contains freight.

It may contain freight with a deadline. Someone will have to touch it later. The system has already spent effort creating that state, and it will have to spend more effort resolving it. Enough dwell turns into congestion.

Congestion changes travel paths. Longer travel paths consume labor. Crowded staging makes it harder to see what is important. Important freight becomes harder to identify.

Now the system spends more labor managing the consequence of work it already performed. That is negative capacity. Removing dwell can therefore create capacity without adding a single square foot. The building did not get bigger.

More of the building became available for useful work.

CAPACITY HIDES IN TIMING

The same thing happens with time. If the operation treats every unit as equally urgent, the operator has to continuously rediscover priority. That consumes attention. If the system knows what is due next and begins preparing early, priority becomes part of the operating cadence.

Now the deadline does not arrive as a surprise. This matters because a dock is not a continuous-flow abstraction. It has clocks. A trailer leaves.

That creates a hard state transition. Freight that makes the departure becomes somebody else's next problem. Freight that misses it remains yours and becomes an exception. The difference may be minutes.

Capacity therefore depends partly on whether the operation understands time as a resource. A process can have plenty of theoretical hourly throughput and still fail if the work reaches the right place after the useful window closes.

CAPACITY HIDES IN THE ROUTE

I physically and virtually worked on conveyance in that environment. Install. Troubleshooting. Sensors.

Speed. Diversion logic. Stacking areas. Container rules.

Dock-door scheduling. Where a box goes matters. Every unnecessary movement consumes something. Conveyor time.

Walking. Equipment. Space. Attention.

If routing creates congestion at one merge, the effect can appear far away from the original decision. If the wrong container enters an area where it cannot be handled correctly, the system creates an exception. If freight is routed to a location that later requires another move, the first move may have been technically successful and operationally wasteful. Good flow reduces the number of times the system has to reconsider the same unit.

That creates capacity. Not because the conveyor magically became larger. Because more of its movement became useful.

CAPACITY HIDES IN THE HANDOFF

A fulfillment center is full of handoffs. Inbound to inventory. Inventory to pick. Pick to pack.

Pack to sort. Sort to dock. Dock to transportation. Transportation to the next node.

Every handoff can create a queue. Every queue can become a buffer. Every buffer can become congestion. The system is not limited only by how fast each department can work.

It is limited by whether the output of one process can become valid input to the next process at the rate and state the next process requires. This is why making one department faster can make the building worse. If process A produces 1,000 units an hour and process B can absorb only 700, process A's extra 300 units are not necessarily additional capacity. They may be inventory accumulating between processes.

The real capacity is constrained by the relationship.

THE BUILDING WAS A MACHINE MADE OF PEOPLE AND MACHINES

It is tempting to describe high-volume operations as a mechanical problem. They are not. People are part of the architecture. A person has limited attention.

A manager can only hold so many active exceptions in their head. An associate needs to understand the next correct action. A new employee cannot instantly inherit years of operating context. A plan that depends on everybody remembering everything is not a scalable plan.

That means information design affects physical capacity. Can the operator see the next CPT? Can the associate identify the correct lane? Can the manager see where freight is dwelling?

Can the team distinguish normal work from an exception? Can the next shift understand the state they inherited? If not, the building spends labor converting information into action manually. That labor is part of the capacity equation too.

YOU CANNOT YELL YOUR WAY TO 60,000

This is the point I care about most. You do not sustainably take a building designed around roughly 16,000 units and push it beyond 60,000 by demanding four times the effort. The geometry does not allow it. There are only so many bodies that fit safely in the space.

Only so many doors. Only so many lanes. Only so many containers. Only so much equipment.

Only so many minutes before a departure. At some point, exhortation stops producing output. The system has to change. Work has to move differently.

Priority has to become clearer. Dwell has to fall. Routing has to improve. The cadence has to fit the deadlines.

The physical layout has to be used more intelligently. People have to spend less attention rediscovering the state of the operation. Constraints have to be found and removed one at a time. Then the next constraint appears.

That is engineering.

THE NUMBER IS NOT THE STORY

Sixty thousand is a great number. It is also dangerous as a standalone résumé bullet. Without the mechanism, it can sound like bragging. The useful story is what the number demonstrates.

The building's original practical assumptions did not describe its eventual practical capability. Capacity was distributed across the architecture. Some of it was trapped in dwell. Some in timing.

Some in routing. Some in handoffs. Some in information. Some in door utilization.

Some in the relationship between people and equipment. We kept finding it. Then we kept discovering the next limit. Eventually the same building behaved like a very different system.

THE NEXT CONSTRAINT LEFT THE BUILDING

There is an important ending to this story. Once a building becomes capable of producing much more, the problem does not disappear. The constraint moves. Now the dock has more freight.

Now transportation needs more capacity. Now downstream buildings receive more volume. Now equipment balance changes. Now the regional network becomes part of the engineering problem.

That is why my work eventually moved beyond one department and beyond one building. Success changes the input to the next system. The 60,000-unit day was not the end of the engineering. It was evidence that the boundary of the engineering had moved.

WHAT CAPACITY REALLY MEANS

Capacity is not simply the number printed in the design document. It is not simply the maximum speed of the equipment. It is not simply headcount. It is not simply square footage.

Capacity is what the complete system can move through valid states, at the required time, without creating a larger failure somewhere else. That definition is harder. It is also much more useful. A 16,000-unit building doing more than 60,000 did not violate physics.

We learned where the real constraints were. Then we changed the system around them.