Much is written on the value of robustness in analytical instruments, but not so much on how it is defined, and even less on the impact it has on lab optimization. In fact, the term robustness (as it relates to lab optimization) may not be defined at all and instead referred to as a feature set, such as temperature control, and grouped without further explanation in product and technical literature.
This blog, and future ones in this series, will take a more critical look at the term robustness, apply a shared definition of the term, and describe how it relates (with quantifiable results) to a portfolio of analytical technologies such as ion chromatography (IC), gas chromatography (GC, GC-MS), liquid chromatography (LC, LC-MS), trace elemental analysis (ICP-OES, ICP-MS), and laboratory information management systems (LIMS).
After spending some time searching the internet for a good working definition of robustness I came up with the following three-part description:
- Measurement robustness: In terms of instrumentation, a robust measurement is one that is both sensitive (able to detect significant changes in the underlying measured parameter) and precise (repeatable with a high signal-to-noise ratio).
- Instrument robustness: The instrumentation itself is considered robust if it is constructed to operate reliably under expected process conditions that can include dust, solvent exposure, vibration, and others once the process is scaled up to the pilot or manufacturing plant.
- Process robustness: In terms of process design and process operation, a robust process means that the system is designed to yield product that meets required specifications in spite of normal process variability.
What Is the Value of Robustness?
While robustness can share a common definition, the value it brings can be unique depending on the product technology and whether it is hardware or software. For example, the value of robustness for instrument hardware may be the confidence in knowing the instrument will run reliably overnight. In contrast, the value of robustness for LIMS software may be the ease of integration and communication with other software.
Like what you are learning?
In addition, robustness can extend beyond the test result it provides and include distinct benefits in installation and maintenance (preventative and unplanned). For example, does replacing a consumable or detector resemble Swiss watch repair or is it more like swapping out a lightbulb?
The following describes a practical example with an IC process robustness study of an analytical procedure aimed at measuring its capacity to remain unaffected by small variations in the method parameters. The study measured the separation and determination of ammonia in a sodium bicarbonate solution using a high capacity carboxylate-based cation exchange column (Dionex Ion Pac CS 16) and a methane sulfonic acid eluent by IC with suppressed conductivity detection.
Results from the robustness study show peak asymmetry ammonia ranged from 1.2 to 1.3 and resolution of ammonia (relative to sodium) ranged from 5.17 to 5.69. The data demonstrate that this is a robust assay for determining ammonia, in spite of the high sodium-to-ammonia ratio.