Analysis of wear in extruder machines screws and cylinders

The Michelin Group is a leading company in the tire production sector, specializing in the manufacturing, distribution, and sale of tires. Specifically, the Michelin Cuneo site focuses on producing semi-finished products and tires for cars and trucks. The extrusion process in Service O of the plant is a critical part of the manufacturing process, as it transforms rubber compounds into profile rubber sheets, which are later used in various tire components, such as the tread, sidewalls, and carcass. The working conditions of the extruder machines directly impact both product quality and productivity. This thesis aims to identify the causes of wear in the screws and cylinders of the extrusion system, such as aggravation of wear due to abrasion caused by foreign metal objects, or inaccuracies due to the data acquisition procedure. Additionally, it seeks to analyze the current rate of tool degradation, using clearance measurements between the screw and cylinder over the past eight years. The wear in each extruder is correlated with the rheology of the processed rubber compounds. This behavior is analyzed under two temperature conditions: low temperature (qualitative classification of rigidity) and high temperature (Mooney viscosity), as the temperature of the compound is different depending on its location inside the barrel. Since the monitoring and measurement techniques currently employed at the plant are not standardized, resulting in non-continuous and unevenly spaced data acquisition, an optimal predictive model (Condition-based Maintenance) cannot be developed at this time. Such models could offer accurate forecasting for the replacement of screws and cylinders, allowing for optimal ordering of new tools. Furthermore, they could support the implementation of advanced maintenance technologies, such as Additive Manufacturing, to repurpose already degraded screws and cylinders. As a result, the thesis concludes the analysis by proposing future techniques that would facilitate the plant's transition towards Industry 4.0 methodologies, by identifying any extruder in critical condition to perform predictive maintenance and to benefit the most from the technology advancements.