Prior to the early 1990s, attempts were made to quantify odour using dynamic olfactometry. However, the methods available at that time for estimating odour concentrations were generally regarded as being too subjective. Since then, the introduction of improved instrument calibration, the adoption of n-butanol as a reference material, and improved panel screening procedures have made it possible to measure odour concentration objectively. The performance of an odour laboratory now can be assessed in terms of measurement accuracy (bias and repeatability) in relation to an agreed reference material such as n-butanol.
Accuracy of a measurement method is described by two terms, "trueness" and "precision" (ISO 5725):
- Trueness (bias) is defined as the closeness of agreement between a test result and an accepted reference value and may be investigated by comparing an accepted reference value with the level of the results given by the measurement method.
- Precision (repeatability and reproducibility) involves the random errors inherent in every measurement procedure. Precision describes how close repeated measurements are to each other. While the term "repeatability" is used to describe precision in the same laboratory under repeatability conditions, the term "reproducibility" is used to describe precision between laboratories.
At standard conditions for olfactometry, the reference value corresponds to a butanol concentration of 0.040 mmol/mol or 40 ppb(w/w). The overall sensory quality criteria for accuracy are: the bias must be equal to or less than 0.217; the repeatability shall be not greater than 0.477. This will result in repeatability limit of 3 (=10 to the power of 0.477). The repeatability limit, expressed as the difference between two single measurements performed on the same testing material in one laboratory under repeatability conditions, should not be larger than a factor 3 in 95% of cases.
An interlaboratory comparison study of olfactometry in Europe was undertaken in 1996. The study demonstrated that individual laboratories following the methods specified in the draft CEN standard for odour concentration measurement (CEN/TC264/WG2) can achieve quality requirements in terms of bias and repeatability as specified in the standard. To assess compliance with the overall sensory quality requirements of the standard a reference value of one odour unit corresponding to 123 mg n-butanol evaporated in one cubic metre of neutral gas has been set.
Over 12 month period of time (April 99 - April 98), it was reported that one laboratory has achieved a bias of 0.182 and a repeatability limit of 2.3 over more than 50 butanol samples tested with an averaged butanol threshold of 61.6 ppb.