Number of observed cycles or element reads. Larger n tightens your estimate of the true mean and σ. Manual time-study rule of thumb: 10–20 cycles, more when variation is high or the cycle is short.

The average. Basis for normal time and standard time. Sensitive to outliers — a single jam or fumble pulls it up.

The middle observation, robust to outliers. If the mean sits well above the median, the data is right-skewed (a few long cycles) — chase those special causes rather than trusting the average.

Best and worst observed cycles. Min approximates entitlement — the best repeatable time. Max usually flags an abnormality (jam, reach, interruption) worth a root-cause look.

Fast measure of spread and the basis of the X̄–R control chart. A wide range means an inconsistent method — a standard-work opportunity.

The core measure of variation around the mean. The n−1 divisor (Bessel's correction) makes s an unbiased estimate of the process σ from a sample. Feeds confidence intervals and capability indices.

Relative variation, unit-free — lets you compare consistency across tasks of very different durations. Field rule of thumb: <10% tight, 10–25% moderate, >25% high (standardize the method). A guide, not a control limit.

Time for one complete cycle — all elements, or one machine cycle. Compare to takt (the customer demand rate) and to your best repeatable cycle (entitlement) to size the gap.

Ranks elements by total active time so you attack the "vital few" first. The cumulative line shows where you pass ~80% of the time — everything left of that point is where the opportunity lives.

Operator or material travel per routine. Multiply by daily repetitions to size the prize. Targets Motion (people) and Transportation (material) — 2 of the 8 wastes.

A leg is one trip between two stations. Path crossings and revisits to the same station signal backtracking and poor flow — aim for one-directional, U- or I-shaped layouts and point-of-use storage.