The thesis proposes a quantitative decision-making model to evaluate the technical and economic viability of demanufacturing for end-of-life products. Developed at the Polytechnic University of Bari, the framework employs specific metrics to calculate the reuse potential of individual components. To ensure industrial scalability, the model has been implemented as an application powered by Artificial Intelligence (LLM). Through rigorous prompt engineering, the AI functions as a pseudo-deterministic parser that automates the triage process: it identifies residual functions, assesses usage constraints, and rapidly routes parts for reuse, upcycling, or recycling. Case studies confirm the system's effectiveness in optimizing analysis times and standardizing value recovery.
La tesi propone un modello decisionale quantitativo per valutare la convenienza tecnica ed economica del Demanufacturing sui prodotti a fine vita. Il framework, sviluppato presso il Politecnico di Bari, utilizza metriche specifiche per calcolare il potenziale di riuso dei singoli componenti. Per garantirne la scalabilità industriale, il modello è stato tradotto in un applicativo basato su Intelligenza Artificiale (LLM). Attraverso un rigoroso prompt engineering, l'IA agisce come un parser pseudo-deterministico che automatizza il triage: identifica le funzioni residue, valuta i vincoli d'uso e smista rapidamente i pezzi verso il riuso, l'upcycling o il riciclo. I casi studio confermano l'efficacia del sistema nell'ottimizzare i tempi di analisi e standardizzare il recupero di valore.
AI E VALORIZZAZIONE DEI COMPONENTI DA DEMANUFACTURING, APPLICAZIONE DI UN MODELLO DECISIONALE A UN CASO STUDIO
SENAPO, FAUSTO GIUSEPPE
2025/2026
Abstract
The thesis proposes a quantitative decision-making model to evaluate the technical and economic viability of demanufacturing for end-of-life products. Developed at the Polytechnic University of Bari, the framework employs specific metrics to calculate the reuse potential of individual components. To ensure industrial scalability, the model has been implemented as an application powered by Artificial Intelligence (LLM). Through rigorous prompt engineering, the AI functions as a pseudo-deterministic parser that automates the triage process: it identifies residual functions, assesses usage constraints, and rapidly routes parts for reuse, upcycling, or recycling. Case studies confirm the system's effectiveness in optimizing analysis times and standardizing value recovery.| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi_Senapo .pdf
File consultabile da postazioni biblioteca
Dimensione
6.14 MB
Formato
Adobe PDF
|
6.14 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/110766