Warehouse productivity

Which warehouse KPIs show productivity?

Two do most of the work: labour utilisation, the share of paid hours spent on the work, and productivity, tasks an hour against what a reasonable operator does. Pick rate by zone and order type sits under both. Read them together and only compare like with like: the same job, the same zone and the same kind of order.

Keep the two apart as well: utilisation says whether the paid hours reached the work; productivity says how well the work went once they did.

Why does labour utilisation look so low on our systems?

Because the systems only see the work they record. Replenishment, goods in, housekeeping, training and the time between tasks often never reach the WMS, so a busy shift can read as an idle one. In two warehouses we've worked in, utilisation read around 42 per cent on the systems and 65 per cent once that work was counted.

The systems weren't wrong; they couldn't see half the day. Counting that work is where Fettle starts, because nobody on the floor should be judged by a number that misses what they actually did.

ONE SHIFT

Where the paid hours in a 7.5-hour shift go

Two bars show a 7.5-hour shift. On the systems alone, the tasks the WMS recorded fill around 42 per cent of it, about three hours ten minutes. With the off-system work counted, the bar reaches 65 per cent, about four hours 53 minutes. A shaded band from 85 to 90 per cent marks the realistic level, about six hours 23 minutes to six hours 45 minutes, at least 90 minutes further on.

On the systems alone42%

Tasks the WMS recorded

With the off-system work counted65%

Tasks the WMS recordedOff-system work

90 minutes or more

Realistic: 85-90% utilisation

02467.5 hours

• Tasks the WMS recorded
• Off-system work, once counted
• Paid time that didn't reach the work
• 90 minutes or more to a realistic 85-90%

Two warehouses we've worked in, measured on their own systems and then with the off-system work counted. The hours are those percentages applied to a 7.5-hour shift.

How do you improve warehouse productivity?

Measure it fairly first, since people won't improve against a number they don't believe. Count the off-system work and compare people with others doing the same job. Then look at what slows the work rather than who is slow: stock in the wrong place, pick lists that bunch people together and time lost at the start of a shift.

Sometimes it's deployment: the right number of people are in, but on the wrong job, client or shift for the work that's there. Most of what turns up is method rather than effort. Regrouping the pick lists in one warehouse lifted output by changing how the work was handed out, with nobody asked to work harder.

Will this be used to cut jobs?