System's approach evaluation for decision-making in a lubricant additives package blending plant

Abstract

This thesis applies a systems thinking approach to evaluate decision-making in a lubricant additives package blending plant facing increasing product complexity and demand. The blending unit operates in a high-mix, multi-product environment, with partial automation and strong reliance on human expertise. Nowadays, capacity is more constrained by loading activities than mixing, but future growth and complexity may shift bottlenecks. Current scheduling practices, based on empirical “equivalent quantity” metrics, inadequately capture true capacity constraints or the impact of complexity, particularly when the in-line mixing facility becomes the critical path. The study combines literature review, on-site observation, data analysis, and system architecture modeling—using Object Process Methodology (OPM) and Design Structure Matrices (DSM)—to represent product families, process interdependencies, and stakeholder needs. Operational data are analyzed to compare actual facility utilization with theoretical conservative critical path metrics, revealing difficulty in adequately capturing data. The actual calculation, Equivalent Quantity, and the new proposed one, Theoretical Critical Path, are proposed to be utilized as complementary metrics to improve scheduling validation and bottleneck identification. Simulation models and metrics are developed to evaluate feasibility, anticipate facility saturation, and assess the impact of introducing new packages. These models consider package composition, number of process steps, facility type, and priority rules. The research concludes that integrating system-level representations with improved metrics and simulation can enhance daily scheduling robustness, support long-term investment planning, and maintain operational resilience. Recommendations include refining data capture, adopting dual-metric scheduling, keeping humans for complex decision-making in this moving and uncertain environment, and applying simulation for scenario testing to prepare the blending unit for future demand and complexity.