robustness
/ roh-BUST-ness /
A good family recipe still turns out fine if the oven runs a little hot, you grab a slightly different brand of flour, or you leave it in a minute too long. A fussy recipe collapses at the smallest change. Robustness is that same quality in an analytical method: it keeps giving the right answer even when the everyday conditions wobble a bit.
Robustness is a measure of how little a method's results change when you make small, deliberate variations in its parameters — the temperature, the pH of a buffer, a reagent's age, the flow rate, the exact wavelength. During validation you nudge each of these on purpose and check that the answer barely moves. A robust method shrugs off such tweaks; a fragile one swings noticeably.
It matters because real laboratories are never perfectly steady: thermostats drift, solvents vary between batches, and people read dials a touch differently. A robust method tolerates that ordinary slop, so its results stay dependable day to day. The caveat is that testing robustness also reveals which steps are critical — and those must be controlled tightly and spelled out clearly in the written procedure.
Testing an HPLC method's robustness, the analyst deliberately changes the mobile-phase pH by ±0.2, the column temperature by ±5 °C and the flow rate by ±0.1 mL/min, and confirms the measured drug content shifts by less than 1% in every case.
Deliberately nudging method parameters to confirm the answer barely moves.
Robustness (small, deliberate changes within one lab) is closely related to ruggedness, which usually refers to how well a method survives larger changes such as different analysts, instruments or laboratories; some standards now merge the two ideas.