Quantitative determination of polyvinyl chloride (PVC) concentration by Raman spectrum was studied in the present work. According to partial least squares (PLS) analysis, it was found that scores of PLS factor 1 were proportional to the concentrations of the sample solutions. Meanwhile, the loadings of factor 1 could reflect the contents of PVC and cyclohexanone simultaneously. The PLS regression model for PVC concentration prediction was built. The values of r and root mean square error(RMSE) between predictive results and actual values were 0.996 3 and 2.775, respectively. The Raman characteristic peaks of PVC and cyclohexanone were found, including the C—Cl bond for PVC (620 and 695 cm-1) and the alicyclic ketone for cyclohexanone (1 709 cm-1). By using internal standard method, another model for PVC concentration prediction was established, and the values of r and RMSE were 0.994 1 and 3.151, respectively. The results indicated that it is feasible to use Raman spectrum to detect the PVC concentration, which is of significant importance to PVC recycling.