Labour efficiency variance for shift-based F&B factories

Material variance is about ingredients. Labour variance is about time.

In an F&B factory, time is more elastic than material — operators can work faster or slower, a line can run at 80% or 95% of design speed, a shift can hit its quota by 6pm or 9pm. Labour efficiency variance is the financial readout of all that elasticity.

The formula

labourEfficiencyVariance = (actualHours − standardHoursForActualOutput) × standardLabourRate

The standard hours come from the routing definition. If the routing says “bottling line — 0.5 hours per 1,000 bottles,” and you completed 10,000 bottles, your standard allowed hours are 5. If the actual was 6 hours, your variance quantity is +1 hour, multiplied by the operator’s standard rate.

The standard rate is set at the resource level (bottling line operator, blending crew, QA inspector) — not per person, per role.

How AION posts it

variance-calculation.service.ts:61-79 does the calc per labour transaction per job. The SLA engine generates the journal:

DR  5220 Labour Efficiency Variance — Bottling Line    SAR 35
    CR  1210 Work in Process                                SAR 35

(Example: 1 hour over standard at SAR 35/hour standard rate.)

By month-end, the variance account aggregates all jobs’ labour variance into one line on the financials.

Worked example — bottling line shift

A bottling line operator at Oasis Fresh runs three jobs in an 8-hour shift:

Job	Output (bottles)	Standard hrs (0.5/1000)	Actual hrs	Variance
JO-001	4,000	2.0	2.5	+0.5
JO-002	6,000	3.0	2.8	−0.2
JO-003	5,000	2.5	2.7	+0.2
Total	15,000	7.5	8.0	+0.5

At SAR 35/hour standard rate, the shift’s labour variance is +0.5 × SAR 35 = SAR 17.50 unfavourable.

For one shift, trivial. For 20 operators × 22 working days/month × similar variance = SAR 154/month — still small.

The signal isn’t the absolute number. It’s the pattern across shifts and crews. Three shifts × 5 days × consistently 5–10% slow on the same line = a real productivity issue worth investigating.

What causes labour variance in F&B

Five typical drivers:

1. Line setup time. A juice line switching from mango to orange takes 30–60 minutes for CIP (clean-in-place). If the standard routing assumed 15 minutes, you’re chronically unfavourable.

2. Equipment delays. A capping head jams once per shift. Five minutes per jam × twice = 10 minutes of lost throughput against the standard. Operators don’t book this as a delay; it shows up as labour variance.

3. Quality holds. QA delays releasing a batch by 30 minutes for testing. The line is staffed but not producing. Labour booked to the job; output not yet credited.

4. Crew skill gap. A new operator on a complex line runs at 75% speed for the first month. Variance reflects the learning curve. Once trained up, variance returns to baseline.

5. Outdated standards. The routing was set when the line ran at 8,000 bottles/hour. Hardware was upgraded; line now does 12,000. Your standards are wrong, and you’re posting favourable variance forever.

The fifth one is the dangerous one — chronic favourable variance looks like good news but means you’re underutilising the line.

What to do with the data

A production director who watches labour variance weekly catches issues 3–4 weeks earlier than one who waits for monthly reports. Three questions:

1. Which line / crew is the variance concentrated on? Sort by variance descending. The top three usually account for 60% of total variance.

2. Is the variance correlated with specific operators, shifts, or product changes? Pivot the data by shift and SKU. Patterns become obvious.

3. Is the variance trending or stable? A new chronic issue shows up as a trend; a structural inefficiency shows up as a stable baseline.

The action menu after these questions is usually one of: retraining, process change, equipment maintenance, or standards refresh.

What this is not

Labour efficiency variance is not the same as payroll. Payroll measures what you paid for hours worked. Labour efficiency measures what those hours produced compared to standard. You can have rising payroll and improving labour efficiency simultaneously (more output per dollar) or vice versa.

It’s also not the same as labour rate variance. Rate variance is “we paid SAR 38/hour for someone whose standard is SAR 35” — a payroll question. Efficiency variance is “we used 8 hours for what should have taken 7” — an operations question. AION captures both, but the meaningful weekly signal is efficiency.

Roadmap note

Labour rate variance posting (separate from efficiency) is on the costing audit’s roadmap. Today, if your standard rates are stale or your actual pay rates are creeping up, that drift will fold into the efficiency variance and confuse the signal. Worth knowing.

For the variance category that posts the cleanest journals today — material usage — start with Material usage variance — spotting recipe drift. For the overhead piece of the variance trio, see Overhead absorption — machine-hour vs labour-hour.