The thing about pick to light battery replacement maintenance is that the expensive part is never the battery. It is the light dying quietly. A picker stands in front of a slot waiting for the light, the light never comes on, he freezes for a second, turns to find a supervisor, the supervisor goes back to the paper list, and the rhythm of the whole order breaks. The principle is as simple as a flashlight: each slot carries a small module with a lamp and a numeric display, the system lights up the slot, the picker takes the quantity shown and presses the confirm button to kill the light. Think of a whole pick line as hundreds or thousands of flashlights taking turns lighting up. Flashlights need new batteries. So does this system, except it needs several hundred at once.

Before last year's peak season I was doing an inspection at a third-party warehouse that handles household goods, a pick zone with more than six hundred light modules. The supervisor told me the first year after go-live was smooth, then from the second summer onward, modules started going dark in patches. It was not a system failure. The batteries were dying on schedule: same installation batch, same battery batch, nearly identical life curves, so they all ran out together. It took us two night shifts to rotate batteries across the whole zone, and peak season nearly started with a hole in it. That left me with one rule: on a battery-powered pick to light system, batteries are not consumables. They are scheduled parts, managed by batch, never left until they fail. (The field story above is an illustrative scenario; its figures are hypothetical.)

Pick to Light Battery Replacement: It Is About Light-Ups, Not Capacity

Start with how the power works. Battery-powered pick to light comes in two flavors: each module carries its own batteries, or a whole row shares a trunk supply that feeds each module. The first saves you the wiring; the second saves you the battery swaps. That is the first dividing line at selection time. The most common battery configuration is two AA cells, and at least one vendor publishes a design life of up to about one year, depending on pick volume, notification settings, and signal conditions. A fast-moving slot can light up several hundred times a day; a slow slot might light up a few times a week. The two can differ by months of battery life. So stop asking how long a battery lasts. Ask how many times a day the slot lights up. Light-ups are the real unit of consumption. Capacity is just the shadow.

There is a second point people miss: batteries suffer more from sitting than from working. A slot idle for months through the slow season lets its batteries self-discharge, and when peak season arrives the lights go dark in whole rows. My habit is to check battery levels in the slow zones during the slow-season stocktake and swap anything below threshold on the spot instead of waiting for peak. The removed cells are not trash yet either. Test their voltage, and move the survivors to even slower slots for one more season. That is cascading use, and battery budgets are won one cell at a time.

Before the Battery Dies, the Light Gives You Three Warnings

First warning: the light dims or flickers. Second: the numeric display drops segments, so an 8 reads as a 0 and the picker takes the wrong quantity. Third: the battery indicator on the management dashboard turns red. The better vendors put a battery indicator on each module and show zone-wide battery status on the dashboard, so you can schedule replacements ahead of time. Two of the three warnings depend on human eyes; only the third depends on the system. A warehouse that relies on eyes alone has handed battery management to the pickers' memory, and that will fail.

Back at that household-goods warehouse, we gave every zone lead a simple sheet: check the dashboard every morning before start-up, and swap any module below twenty percent charge the same day. Do not wait for it to die. Two months in, line stops caused by dead lights fell from several a month to zero. Nothing about the hardware changed. The inspection rhythm changed. Most maintenance holes are inspection-rhythm holes at root; the equipment rarely deserves the blame.

Batch Replacement or Break-Fix: Run the Pick to Light Maintenance Math

This is the question I get most. My answer is to split by frequency: batch replacement for fast zones, break-fix for slow ones. Batch replacement concentrates the labor, one entry to swap a whole row. Break-fix wastes nothing, since you only swap cells that are actually dead. The dividing line is one comparison: the labor cost of swapping a battery versus the cost of a line stop when a light dies.

Here is a worked example, hypothetical figures for a six-hundred-module zone:

Approach Entries per year Labor hours per year Dead-light stop risk
Full-zone batch swap, twice a year 2 about 32 labor hours Low, cells swapped before mass failure
Break-fix only about 40 a month about 120 labor hours High, dark lights found only when pickers report them
Batch for fast zones, break-fix for slow zones about 6 about 50 labor hours Low, fast zones never lose power

The labor hours assume a two-person crew and eight hours per entry. Your warehouse will differ, so re-run it with your own measured numbers. But the direction of the conclusion holds: the denser the slots and the higher the volume, the more batch replacement pays; the sparser and slower the zone, the more break-fix makes sense. Spare stock follows the same logic: hold one quarter of usage plus twenty percent safety stock, and top up once before peak season. Rechargeable cells deserve a serious look too. At least one vendor recommends them to cut waste and maintenance volume, but they need a charging routine and a rotation pool, a charging corner somewhere in the warehouse, or the money saved on cells turns into management cost. For exact models and prices, go by the vendor's latest specifications, not a quote from three years ago.

Swapping Batteries Without Stopping the Pick

The golden rule: pickers never swap their own batteries. The right flow is to swap between waves, in pairs, one person swaps and the other verifies, then light up the module to self-test and reset the battery level in the system. Removed cells go straight into the recycling box. Never leave them lying on the rack. An AA cell that rolls into a slot will be reported as a foreign object the next day. One more step after the swap: update the baseline date for that batch of modules on the dashboard. Otherwise the system keeps calculating life from the old date, and every future warning lands in the wrong week.

Here is the counterintuitive part: the most expensive cost of a battery is not the battery. It is the few seconds a picker spends staring at a dark light, and the minutes of rework after a mispick. Published figures put pick to light accuracy at about 99.6%, and that number assumes the lights are actually on. Two dark modules in a row and the accuracy math has to be redone. The selection framework is three lines: high slot density, high pick volume, and high rewiring cost point to the battery version; fixed slots, steady volume, and easy wiring point to the wired version. Neither side is better. Only fitting or not fitting.

So whether the money was well spent comes down to one number: how many minutes per module per year did dead batteries cost you? More than thirty minutes, and your replacement rhythm is behind. Under ten minutes, and batteries are under control. Before start-up tomorrow, take one look at the battery distribution on the dashboard. That is worth more than any meeting about it.

By Da Liu, Automation Engineer