Statistical Measure
Multivariate distance measurement used to determine the similarity between a sample and a known reference group by accounting for the correlations between variables. Application of mahalanobis distance classification allows laboratories to identify the geographic origin of natural fibers by comparing their chemical profiles to an established database. This method is effective only when the reference data is sufficiently diverse to represent the entire population.
Calculation Logic
Evaluation begins by calculating the covariance matrix of the reference set. The distance is then measured from the sample to the mean of the group in units of standard deviation. Because mahalanobis distance classification treats the data as a multidimensional cloud, it identifies outliers that a simple Euclidean measure would miss.
This sensitivity makes it a preferred tool for detecting fiber blends that attempt to mimic a single origin signature.
Verification Tool
Forensic scientists use the results to confirm or deny the claims made on a certificate of origin. A low distance score indicates that the sample is highly likely to belong to the claimed group. When mahalanobis distance classification produces a high score, the material is flagged for further investigation or rejection.
The technique is commonly applied to cotton and wool to ensure that expensive fibers are not being diluted with cheaper alternatives.
Data Requirement
Success depends on the quality and the quantity of the underlying reference samples. A database must be updated every growing season to account for changes in weather and soil chemistry. If the reference set is too small, the mahalanobis distance classification may return a false positive or an inconclusive result.
This limitation means the system is most reliable for large, well documented production regions.
