Redução da variabilidade na medição de casca em diferentes pontos de amostragem em uma planta de PVC

Abstract

Polyvinyl Chloride (PVC) is the most versatile thermoplastic among plastics. Due to its need for formulation through the incorporation of additives, PVC's characteristics can be altered within a wide spectrum of properties depending on the final application, ranging from rigid to extremely flexible. The different types of produced resins (grades), to comply with the Product Specification (PS), need to have their quality parameters, like the shell (Gran>500µm), for example, in accordance with the specification limits imposed by the PS. The shell in PVC resins can be defined as the particles with granulometry larger than 500 microns, which are undesirable for the process, as they directly affect the processing of the resin in the transforming industries, possibly causing infused points in the final product. The present work aimed to reduce the variability of the analysis of this quality parameter of PVC between two main sampling points of a chemical industry with a fictitious name "Poliquímica": blower and batches, for the grade SP 600RX. By applying the problem-solving and continuous improvement methodology DMAIC, it was possible to find the root causes of this problem, in a qualitative and quantitative manner, and propose corrective actions so that they could be minimized or remedied, resulting in the reduction of the variability of the quality parameter shell to minimum average and standard deviation values, which were matched between the two sampling points of the plant, as well as predicting the occurrence of these undesirable particles with the PVC still present in the process, which allowed the engineering team to take actions to prevent them from reaching the PVC batches, i.e., the final product.