This research presents an integrated, data-driven framework for selecting optimal cutting decisions in steel fabrication workflows. The framework combines Building Information Modeling (BIM)-based data extraction, Multi-Objective Optimization (MOO), and Multi-Criteria Decision-Making (MCDM) supported by sensitivity analysis. The methodology enables a systematic evaluation of total cost, production time, and embodied carbon of the structure, balancing economic, temporal, and environmental objectives in fabrication process. Through BIM integration via the Tekla Open API, real fabrication parameters such as material type, geometry and thickness are used to calculate quantitative decision variables. The MOO stage identifies Pareto-efficient alternatives that represent optimal trade-offs among conflicting objectives, while the MCDM process ranks these alternatives according to user preferences. Sensitivity analysis was performed to examine how variations in weight distributions and design parameters influence the optimization outcomes and decision variables. A series of case studies which are covering variations in column thickness, diameter, project scale, and beam profiles were conducted to validate the robustness of the framework and demonstrate the framework’s adaptability in various conditions. Results demonstrate that cost-oriented, time-oriented, and sustainability-oriented priorities lead to distinct optimal configurations and objective trade-offs, highlighting how different decision preferences shape cutting strategies within the optimization framework. The proposed approach contributes to the advancement of data-driven decision-making in construction manufacturing by enabling smarter, more efficient, and environmentally responsive fabrication workflows.
Questa ricerca presenta un quadro integrato e basato sui dati per la selezione delle decisioni di taglio ottimali nei processi di fabbricazione dell’acciaio. Il framework combina l’estrazione dei dati basata sul Building Information Modeling (BIM), l’Ottimizzazione Multi-Obiettivo (MOO) e il Processo di Decisione Multi-Criterio (MCDM), supportato da un’analisi di sensibilità. La metodologia consente una valutazione sistematica del costo totale, del tempo di produzione e del carbonio incorporato della struttura, bilanciando gli obiettivi economici, temporali e ambientali nel processo di fabbricazione. Attraverso l’integrazione BIM tramite la Tekla Open API, i parametri reali di fabbricazione — come il tipo di materiale, la geometria e lo spessore — vengono utilizzati per calcolare variabili decisionali quantitative. La fase MOO identifica le alternative Pareto-efficienti che rappresentano i compromessi ottimali tra obiettivi in conflitto, mentre il processo MCDM classifica tali alternative in base alle preferenze dell’utente. L’analisi di sensibilità è stata eseguita per esaminare come le variazioni nella distribuzione dei pesi e nei parametri di progetto influenzino i risultati dell’ottimizzazione e le variabili decisionali. Una serie di casi di studio — che comprendono variazioni nello spessore e nel diametro delle colonne, nella scala del progetto e nei profili delle travi — è stata condotta per validare la solidità del framework e dimostrarne l’adattabilità in diverse condizioni operative. I risultati dimostrano che le priorità orientate ai costi, ai tempi e alla sostenibilità portano a configurazioni ottimali e compromessi obiettivi distinti, evidenziando come le diverse preferenze decisionali influenzino le strategie di taglio all’interno del quadro di ottimizzazione. L’approccio proposto contribuisce all’avanzamento delle decisioni basate sui dati nella produzione edilizia, consentendo flussi di lavoro di fabbricazione più intelligenti, efficienti e sensibili all’ambiente.
Automated decision-making framework for steel fabrication using Multi-Objective Optimization
Alkan, Huseyin Mert
2024/2025
Abstract
This research presents an integrated, data-driven framework for selecting optimal cutting decisions in steel fabrication workflows. The framework combines Building Information Modeling (BIM)-based data extraction, Multi-Objective Optimization (MOO), and Multi-Criteria Decision-Making (MCDM) supported by sensitivity analysis. The methodology enables a systematic evaluation of total cost, production time, and embodied carbon of the structure, balancing economic, temporal, and environmental objectives in fabrication process. Through BIM integration via the Tekla Open API, real fabrication parameters such as material type, geometry and thickness are used to calculate quantitative decision variables. The MOO stage identifies Pareto-efficient alternatives that represent optimal trade-offs among conflicting objectives, while the MCDM process ranks these alternatives according to user preferences. Sensitivity analysis was performed to examine how variations in weight distributions and design parameters influence the optimization outcomes and decision variables. A series of case studies which are covering variations in column thickness, diameter, project scale, and beam profiles were conducted to validate the robustness of the framework and demonstrate the framework’s adaptability in various conditions. Results demonstrate that cost-oriented, time-oriented, and sustainability-oriented priorities lead to distinct optimal configurations and objective trade-offs, highlighting how different decision preferences shape cutting strategies within the optimization framework. The proposed approach contributes to the advancement of data-driven decision-making in construction manufacturing by enabling smarter, more efficient, and environmentally responsive fabrication workflows.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_12_Alkan.pdf
non accessibile
Descrizione: Text of Thesis
Dimensione
3.02 MB
Formato
Adobe PDF
|
3.02 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/246482