Key points are not available for this paper at this time.
Mismanagement of plastic waste globally has resulted in a multitude of environmental issues, which could be tackled by boosting plastic recycling rates. Chemometrics has emerged as a useful tool for boosting plastic recycling rates by automating the plastic sorting and recycling process. This paper will comprehensively review the recent works applying chemometric methods to plastic waste sorting. The review begins by introducing spectroscopic methods and chemometric tools that are commonly used in the plastic chemometrics literature. The spectroscopic methods include near-infrared spectroscopy (NIR), mid-infrared spectroscopy (MIR), Raman spectroscopy and laser-induced breakdown spectroscopy (LIBS). The chemometric tools include principal component analysis (PCA), linear discriminant analysis (LDA), partial least square (PLS), k-nearest neighbors (k-NN), support vector machines (SVM), random forests (RF), artificial neural networks (ANNs), convolutional neural networks (CNNs) and K-means clustering. This review revealed four main findings. (1) The scope of plastic waste should be expanded in terms of types, contamination and degradation level to mirror the heterogeneous plastic waste received at recycling plants towards understanding potential application in the recycling industry. (2) The use of hybrid spectroscopic method could potentially overcome the limitations of each spectroscopic methods. (3) Develop an open-sourced standardized database of plastic waste spectra would help to further expand the field. (4) There is limited use of more novel machine learning tools such as deep learning for plastic sorting.
Neo et al. (Fri,) studied this question.