Resilience of aluminum production processes: analyzing variability propagation with the functional random walker

Industrial production processes require effective management of operational variability to ensure safety and maintain stringent quality standards. In aluminum rolling plants, the grinding of rolling cylinders is not a traditional maintenance task but a continuous, integrated activity essential for preserving the surface precision needed to meet demanding product requirements. This study applies the Functional Random Walker (FRW) approach – an extension of the Functional Resonance Analysis Method (FRAM) enhanced by network theory – to analyze the propagation of variability within the grinding operations of rolling cylinders. By modeling these activities, the FRW identifies how specific operational loops can amplify uncertainty (such as those related to coolant management), while operator routines (such as systematic grinding wheel inspections) can effectively dampen variability. The findings demonstrate the practical value of the FRW in enhancing decision-making and proactive resilience management in complex production environments, particularly where the continuous restoration of rolling cylinder surfaces is vital to maintaining high process quality. This work supports the adoption of system-theoretic approaches for the management of industrial plants where production and maintenance are deeply interwoven.