Master's Thesis
Raw Material Inventory Management Applied to the Plastic Sector
— 2025
Key information
Authors:
Supervisors:
Published in
November 26, 2025
Abstract
The plastics industry has faced numerous challenges from environmental concerns, supply chain disruptions, and regulatory changes, all of which create uncertainty in demand and lead time, complicating raw material ordering. This dissertation develops an inventory management model to support short and long-term decision-making regarding order placement. Key factors were analysed, including: forecasting consumption data, addressing lead time uncertainty through a scenario-based approach, and calculating safety stock using two different approaches. The decision support tool consists of two MILP models: while the Aggregated Model focuses on a mass balance per product, without incorporating plant-specific constraints, the Disaggregated Model adds operational constraints related to material movement and mass balance per storage location types. A real-world case study was conducted using data from the Portuguese rigid plastic packaging company Logoplaste, specifically its Mealhada plant. Insights showed that the machine learning forecasting model, XGBoost, significantly overestimated consumption, resulting in higher stock levels. Conversely, when actual consumption data was used with the cost-minimizing Aggregated Model, planned order quantities were lower than in reality, with reductions of 4% and 9% in scenarios with and without safety stock, respectively. When the goal changed to minimize quantities instead of costs, the models were also financially beneficial for the company, representing savings of 6% in raw material purchasing. Based on its performance, the Aggregated Model is recommended for long-term strategic decisions, while the Disaggregated Model — using outputs from the Aggregated Model — should be applied for short-term operational decisions to efficiently allocate raw materials to appropriate storage locations.
Publication details
Authors in the community:
Mafalda Botelho Simões
ist193654
Supervisors of this institution:
Daniel Rebelo dos Santos
ist167904
Susana Isabel Carvalho Relvas
ist46455
Fields of Science and Technology (FOS)
other-engineering-and-technologies - Other engineering and technologies
Publication language (ISO code)
eng - English
Rights type:
Embargoed access
Date available:
September 10, 2026
Institution name
Instituto Superior Técnico