Chemical characterization of illicit drugs, like cocaine, is important to provide chemical and physical information to assist police agencies to understand drug trafficking and identify drug origin. In this context, the present work shows the use of an electrochemical sensor containing two working electrodes (pt and GC) to extract voltammetric information about cutting agents added to cocaine (procaine, benzocaine, phenacetin, levamisole and caffeine) for purpose of discrimination and classification. The extract voltammetric information was used as input data for Principal Components Analysis (PCA) and k-nearest Neighbor (k-NN) analysis. Satisfactory discrimination was achieved using PCA, and then, a supervised classification was performed to test the ability of the proposed method to identify four real seized cocaine samples. The supervised model did not misclassify any of the samples, demonstrating the reliability of the proposed method for field analysis in forensic police intelligence.