Your Manual Station Count Is a Packaging Decision

The list that names the stations

In an interview at a food plant, the person who owns improvement work was asked where his priorities sat for the rest of the year. He answered in three buckets: labor, downtime, scrap. Then he named the places. Product fillers. Hoppers. Label makers. Box makers. Then he named the real question. Some lines run one filler. Some run two. Some run three. Where there are three, can it be two. Where there are two, can it be one.

That last question arrives dressed as a staffing question. It is not one. He had just described four station types, every one of them created by the pack format, and he was asking whether the count was a requirement or an inheritance.

Station count is designed, not staffed

Work the backward chain on any manual station and it lands on a specification. Fill weight and tolerance set how much manual correction a filler needs. Container geometry sets whether a hopper feeds cleanly or needs a person standing by it. Label count and placement set how many label stations exist. Case pattern and count set the work at the end of the line. None of those were chosen by the operations team, and none of them were chosen with a crew number attached.

The tell is variation. When two lines running comparable product carry different station counts, the higher count is not a requirement of the work. It is a requirement of a format, a layout, or a decision nobody wrote down. That difference is measurable today, with no study, by walking both lines and counting.

A time study cannot reach this. A study measures the task in front of it, and the task in front of it is real: the operator is busy, the motion is necessary, the rate is what it is. The study can make three fillers faster. It cannot ask whether the third filler exists because the fill tolerance is tighter than the customer specification requires, or because a hopper geometry forces intervention at a rate the original design never priced. Those questions sit outside the station, in documents the plant does not own.

This is why labor opportunity that looks exhausted often is not. The plant has optimized every station it was given. The count it was given was never examined.

Where the decision belongs

Build the station map first. For every manual station on the line, write the format or layout requirement that creates it, in one sentence, naming the document it comes from. Most entries will resolve cleanly. The ones where nobody can name the requirement are the candidates, and they are usually the oldest.

Price each candidate as crew minutes per thousand units, not as a headcount. A station is paid for in minutes every time the format runs, so minutes per thousand units is the number that survives a mix change and the number a specification review can act on. Then take the map where format changes are actually approved. That is a specification and layout decision, and it belongs with the people who sign those, with the plant's minutes attached to it.

The order matters. Stabilize before you automate. Automating a station whose existence nobody can justify buys equipment to perform work the format should not require, and it makes the station permanent by putting capital behind it. Name the requirement, remove the stations whose requirement nobody can name, then automate what is left.

The rule I apply is to ask for the station map before the time studies. The map tells you which stations should exist. The studies tell you how fast the ones that do exist run. Running them in that order costs nothing and changes what the project is about.

Keep the value states separate. Removing a station is an implemented change. It becomes margin only when the hours are converted: reduced paid overtime, a hire avoided as volume grows, or a crew redeployed to work currently not getting done. A station removed with no conversion named is a cleaner line and the same cost.

What a well-designed line reads like

Every manual station on the line traces to a named requirement in a current specification or layout document. Two lines running comparable product carry the same station count, and where they do not, the reason is written down. Crew minutes per thousand units exists per station, by product family. Any specification or format change states the manual stations and crew minutes it adds or removes before it is approved, and that statement is part of the approval, not a follow-up. The count of stations whose requirement nobody can name is zero.

Count the stations on two lines

Pick two lines that run comparable product and count the manual stations on each. If the counts differ, you have found a design choice being paid for as labor, and you can size it this week with a clipboard. Then ask one question per station on the higher-count line: which document requires this. The stations with no answer are the work.