Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Application of Artificial Intelligence to Enhance sustainability and Transparency in Additive Manufacturing: A Simulation-Based Study using MATLAB

Hatami Majoomerd, Majid LU (2025) MMTM05 20251
Production and Materials Engineering
Abstract
Additive Manufacturing (AM), commonly referred to as 3D printing, has known as a transformative approach in modern manufacturing. It enables the production of complex geometries with reduced material usage, localized fabrication, and improved design flexibility. Despite these advantages, challenges remain in optimizing AM processes to ensure resource efficiency, particularly in minimizing material waste and reducing energy consumption during production.
This thesis investigates how the integration of Artificial Intelligence (AI) can enhance the sustainability of AM, while also contributing to the broader goals of production transparency and traceability through Digital Product Passports (DPPs). The study focuses specifically on Fused... (More)
Additive Manufacturing (AM), commonly referred to as 3D printing, has known as a transformative approach in modern manufacturing. It enables the production of complex geometries with reduced material usage, localized fabrication, and improved design flexibility. Despite these advantages, challenges remain in optimizing AM processes to ensure resource efficiency, particularly in minimizing material waste and reducing energy consumption during production.
This thesis investigates how the integration of Artificial Intelligence (AI) can enhance the sustainability of AM, while also contributing to the broader goals of production transparency and traceability through Digital Product Passports (DPPs). The study focuses specifically on Fused Deposition Modeling (FDM), one of the most widely used AM techniques. Within this context, machine learning methods—namely linear regression and decision tree models and classification learners—are employed to analyze and predict the outcomes of various print parameters. The aim is to develop AI-based strategies that can optimize settings such as layer height, infill density, and print speed in order to reduce both material consumption and production time.
To support this investigation, a series of 3D-printed test components were designed using CREO software. These parts were simulated with different parameter combinations to generate a diverse dataset for analysis. The collected data were then used to train and test predictive models in MATLAB. Although the experimental phase is still in progress, early simulation results indicate that AI-driven optimization has the potential to significantly improve manufacturing efficiency. In addition, the research explores how the insights generated by these models could be structured and recorded within a Digital Product Passport, providing a foundation for improved traceability and informed decision-making in digital supply chains.
Overall, the thesis presents both a technical and conceptual framework for integrating AI into sustainable AM processes. It also reflects on the future role of intelligent manufacturing systems in enabling circular economy principles. The final results, including full model validation and performance analysis, will be documented following the completion of experimental testing. (Less)
Abstract (Swedish)
Additiv tillverkning (AM), vanligtvis kallad 3D-utskrift, erkänns som en transformativ metod inom modern tillverkning. Tekniken möjliggör framställning av komplexa geometrier med ökad materialeffektivitet, lokaliserad produktion och större designfrihet. Trots dessa fördelar kvarstår utmaningar när det gäller att optimera AM-processer för att säkerställa resurseffektivitet – särskilt i att minska materialspill och energiförbrukning under produktionen.
Detta examensarbete undersöker hur integrationen av artificiell intelligens (AI) i AM kan förbättra hållbarheten och samtidigt bidra till ökad transparens och spårbarhet genom Digitala Produktpass (DPP). Studien fokuserar specifikt på Fused Deposition Modeling (FDM), en av de mest använda... (More)
Additiv tillverkning (AM), vanligtvis kallad 3D-utskrift, erkänns som en transformativ metod inom modern tillverkning. Tekniken möjliggör framställning av komplexa geometrier med ökad materialeffektivitet, lokaliserad produktion och större designfrihet. Trots dessa fördelar kvarstår utmaningar när det gäller att optimera AM-processer för att säkerställa resurseffektivitet – särskilt i att minska materialspill och energiförbrukning under produktionen.
Detta examensarbete undersöker hur integrationen av artificiell intelligens (AI) i AM kan förbättra hållbarheten och samtidigt bidra till ökad transparens och spårbarhet genom Digitala Produktpass (DPP). Studien fokuserar specifikt på Fused Deposition Modeling (FDM), en av de mest använda AM-teknikerna. Inom denna ramverk tillämpas maskininlärningsmetoder – inklusive linjär regression, beslutsträd och klassificeringsmodeller – för att analysera och förutsäga resultat baserat på olika utskriftsparametrar. Målet är att utveckla AI-baserade strategier för att optimera lagerhöjd, infill-densitet och utskriftshastighet, och därmed minska materialanvändning och produktionstid.
För att stödja detta arbete utformades en serie testkomponenter i programvaran CREO. Dessa delar simulerades och producerades med varierande parameterkombinationer för att skapa ett dataset för analys. Datan användes därefter för att träna och utvärdera prediktiva modeller i MATLAB. Även om den experimentella valideringen pågår, tyder preliminära resultat på att AI-optimering kan avsevärt förbättra produktionseffektiviteten. Dessutom undersöker arbetet hur AI-genererad insikt kan integreras i DPP:er för att tillhandahålla transparent och spårbar information i den digitala leveranskedjan.
Sammanfattningsvis presenterar detta examensarbete ett tekniskt och konceptuellt ramverk för hur AI kan integreras i hållbara AM-processer. Det reflekterar även över den framtida rollen för intelligenta tillverkningssystem i främjandet av cirkulära ekonomiska modeller. Slutresultat, inklusive prestandamått för modellerna och fullständig validering, kommer att redovisas efter avslutad experimentell fas. (Less)
Popular Abstract
Additive Manufacturing (AM), commonly known as 3D printing, is changing how products are made. Unlike
traditional methods that often waste materials, AM builds objects layer by layer, using only the necessary
amount of material. This makes it an attractive solution for producing parts more efficiently and sustainably.

However, 3D printing can still consume time and energy, especially if the printing settings are not properly
chosen. This thesis investigates how Artificial Intelligence (AI) can be used to make 3D printing smarter. By
using machine learning tools such as regression and decision trees, the research explores how to find the
best combinations of print settings-like layer height, infill density, and speed-that reduce... (More)
Additive Manufacturing (AM), commonly known as 3D printing, is changing how products are made. Unlike
traditional methods that often waste materials, AM builds objects layer by layer, using only the necessary
amount of material. This makes it an attractive solution for producing parts more efficiently and sustainably.

However, 3D printing can still consume time and energy, especially if the printing settings are not properly
chosen. This thesis investigates how Artificial Intelligence (AI) can be used to make 3D printing smarter. By
using machine learning tools such as regression and decision trees, the research explores how to find the
best combinations of print settings-like layer height, infill density, and speed-that reduce material usage and
printing time.

To test this, a simple 3D part was designed and printed many times with different settings. Data from these
prints were used to train AI models in MATLAB, a software commonly used in engineering. The models were
able to predict how much material and time would be needed for each setting-and suggest the most efficient
ones.

The results show that using AI can significantly reduce both material consumption and energy use, making
the production process more sustainable. In some cases, improvements of more than 15-20% were
achieved.

In addition, the study looks at how this optimized data can be stored in a Digital Product Passport (DPP)-a
digital file that records how a product was made. This can help companies prove that their production
methods are more transparent and environmentally friendly.

In summary, this project combines 3D printing and AI to support smarter, greener, and more transparent
manufacturing-and demonstrates that AI can make a real difference in improving efficiency and sustainability. (Less)
Please use this url to cite or link to this publication:
author
Hatami Majoomerd, Majid LU
supervisor
organization
course
MMTM05 20251
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Additive Manufacturing (AM), Artificial Intelligence (AI), Fused Deposition Modeling (FDM), Sustainability, Digital Product Passport (DPP), Machine Learning, Process Optimization, Material Efficiency, Print Time Reduction, Transparency, Circular Economy, MATLAB Simulation
language
English
id
9202002
date added to LUP
2025-06-18 12:43:04
date last changed
2025-06-18 12:43:04
@misc{9202002,
  abstract     = {{Additive Manufacturing (AM), commonly referred to as 3D printing, has known as a transformative approach in modern manufacturing. It enables the production of complex geometries with reduced material usage, localized fabrication, and improved design flexibility. Despite these advantages, challenges remain in optimizing AM processes to ensure resource efficiency, particularly in minimizing material waste and reducing energy consumption during production.
This thesis investigates how the integration of Artificial Intelligence (AI) can enhance the sustainability of AM, while also contributing to the broader goals of production transparency and traceability through Digital Product Passports (DPPs). The study focuses specifically on Fused Deposition Modeling (FDM), one of the most widely used AM techniques. Within this context, machine learning methods—namely linear regression and decision tree models and classification learners—are employed to analyze and predict the outcomes of various print parameters. The aim is to develop AI-based strategies that can optimize settings such as layer height, infill density, and print speed in order to reduce both material consumption and production time.
To support this investigation, a series of 3D-printed test components were designed using CREO software. These parts were simulated with different parameter combinations to generate a diverse dataset for analysis. The collected data were then used to train and test predictive models in MATLAB. Although the experimental phase is still in progress, early simulation results indicate that AI-driven optimization has the potential to significantly improve manufacturing efficiency. In addition, the research explores how the insights generated by these models could be structured and recorded within a Digital Product Passport, providing a foundation for improved traceability and informed decision-making in digital supply chains.
Overall, the thesis presents both a technical and conceptual framework for integrating AI into sustainable AM processes. It also reflects on the future role of intelligent manufacturing systems in enabling circular economy principles. The final results, including full model validation and performance analysis, will be documented following the completion of experimental testing.}},
  author       = {{Hatami Majoomerd, Majid}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Application of Artificial Intelligence to Enhance sustainability and Transparency in Additive Manufacturing: A Simulation-Based Study using MATLAB}},
  year         = {{2025}},
}