Oil containing chemicals and industrial wastes tends to accumulate, forming what is known as oil sludge. This is becoming an increasingly large environmental challenge in certain oil producing regions, including parts of Russia.

These oil sludges contain a mixture of hydrocarbons, minerals and water in soil or sand, and occur across all stages of oil mining and processing.

To understand the complex physico-chemical properties of the waste and depository characteristics, scientists in the Samara region of Russia used multivariate data analysis methods to assess the environmental impact and financial viability of industrial processing of waste containing oil.

Samples from fifty-four depositories provided data for classification of the samples, which was subsequently analyzed with The Unscrambler® multivariate data analysis software using Principal Component Analysis (PCA) and Partial Least Squares regression (PLS).

Principal Component Analysis (PCA), which is an extremely powerful exploratory data analysis tool (sometimes called ‘data mining’), was used to get a deeper understanding of the internal data structure, namely correlations and groupings between variables. The insights gained from the PCA allowed the scientists to recommend a new approach to the classification of oil sludge depositories.

The Unscrambler® software was also used to perform Partial Least Squares (PLS) regression analysis of the data. Based on the results of the Partial Least Squares (PLS) regression analysis, it was possible to make a more accurate assessment of the profitability of processing the depository for specific sites.

To read more about this please see doi:10.1016/j.jenvman.2012.03.041 (V.V. Ermakov et al., Journal of Environmental Management 105 (2012) 144-151).