The increasing use of plastics in various applications and the growing demand for environmentally friendly and efficient recycling methods have led to increased interest in identifying and separating plastic types. In particular, multilayer plastics pose a special challenge for recycling due to their complex composition. In this context, modern analysis methods such as near-infrared spectroscopy (NIR) are gaining importance because they enable rapid and precise identification of plastics. This study focuses on the application of NIR spectroscopy and statistical analysis methods for the identification and classification of multilayer plastics using the portable NIR handheld MicroPHAZIR from Thermo Fisher Scientific. The suitability of NIR in combination with principal component analysis (PCA) and different classification algorithms was examined. The aim was to analyze the effects of different backgrounds (aluminum, tile, gold, copper, PVC, silver, and Teflon), film thicknesses, and transparency on identification performance. The results show that both aluminum and PVC as backgrounds in combination with the KNN classification algorithm were able to achieve a flawless identification rate of 100%.