This thesis investigates how human-centric order release mechanisms improve performance in dual-resource-constrained manufacturing systems, advancing the principles of Industry 5.0. Traditional workload control rules consider only machine capacity, overlooking workforce availability and variability. In contrast, this research embeds worker capacity as a first-class constraint and introduces dynamic workload norm adjustment to respond to absenteeism and machine downtime. A discrete-event simulation of a five-station flow shop was developed to compare machine-centric workload limiting, human-centric workload limiting, and dynamic workload limiting. A full factorial experimental design varied human variability, absenteeism, downtime, and workload norms, with system performance evaluated in terms of Gross Throughput Time (GTT), Shop Floor Throughput Time (SFT), mean tardiness, and tardy percentage. Results show that human variability and absenteeism dominate system performance, while machine downtime exerts a smaller but non-negligible effect. Human-centric release consistently outperforms machine-centric control by preventing infeasible releases to human-constrained stations, producing shorter throughput times and fewer late jobs. Under disruption scenarios, particularly when absenteeism and downtime co-occur, the resilience advantage of human-centric control becomes decisive. Dynamic workload norm adjustment further enhances stability by aligning release decisions with actual available capacity. The main contributions are threefold: (i) extension of workload control theory to dual-resource contexts through a dual-layer feasibility check, (ii) empirical validation of the compound impact of human- and machine-related disruptions, and (iii) operationalization of Industry 5.0 principles by providing transparent, explainable, and worker-aware release mechanisms. The findings offer both theoretical advancement and practical guidance for implementing robust production control in environments characterized by variability and uncertainty.
Questa tesi analizza come i meccanismi di rilascio “human-centric” possano migliorare le prestazioni nei sistemi manifatturieri a doppia risorsa, in coerenza con i principi di Industry 5.0. Le regole tradizionali di workload control considerano solo la capacità delle macchine, trascurando la disponibilità e la variabilità della forza lavoro. In questa ricerca, invece, la capacità umana viene trattata come vincolo primario e viene introdotto un aggiornamento dinamico della Workload Norm (WLN) per adattarsi a eventi di assenteismo e downtime delle macchine. È stato sviluppato un modello di simulazione ad eventi discreti di un flow shop a cinque stazioni per confrontare quattro politiche: regola macchina-centrica, regola human-centrica e variante dinamica con WLN adattiva. Il disegno sperimentale fattoriale ha variato la variabilità umana, i livelli di assenteismo, i livelli di downtime e i valori di WLN; le prestazioni sono state valutate su Gross Throughput Time (GTT), Shop Floor Throughput Time (SFT), ritardo medio e percentuale di ordini tardivi. I risultati mostrano che variabilità umana e assenteismo sono i fattori dominanti, mentre il downtime ha un impatto minore ma comunque significativo. La regola human-centrica supera costantemente quella macchina-centrica, riducendo tempi medi e ritardi grazie alla prevenzione di rilasci non fattibili verso stazioni limitate dalla risorsa umana. In scenari di perturbazione, specialmente congiunzione di assenteismo e downtime, il vantaggio di resilienza della logica human-centrica diventa decisivo. L’aggiornamento dinamico della WLN aumenta ulteriormente la stabilità, allineando le decisioni di rilascio alla capacità effettivamente disponibile. I contributi principali sono: (i) l’estensione della teoria del workload control ai contesti dual-resource, (ii) la validazione empirica degli effetti composti delle perturbazioni umane e meccaniche e (iii) l’operazionalizzazione dei principi di Industry 5.0 tramite un meccanismo di rilascio trasparente, spiegabile e orientato ai lavoratori. I risultati offrono un avanzamento teorico e linee guida pratiche per implementare controlli di produzione resilienti in contesti caratterizzati da variabilità e incertezza.
Dynamism in human-centric workload control: simulation-based assessment
AMIN, CATALINA
2024/2025
Abstract
This thesis investigates how human-centric order release mechanisms improve performance in dual-resource-constrained manufacturing systems, advancing the principles of Industry 5.0. Traditional workload control rules consider only machine capacity, overlooking workforce availability and variability. In contrast, this research embeds worker capacity as a first-class constraint and introduces dynamic workload norm adjustment to respond to absenteeism and machine downtime. A discrete-event simulation of a five-station flow shop was developed to compare machine-centric workload limiting, human-centric workload limiting, and dynamic workload limiting. A full factorial experimental design varied human variability, absenteeism, downtime, and workload norms, with system performance evaluated in terms of Gross Throughput Time (GTT), Shop Floor Throughput Time (SFT), mean tardiness, and tardy percentage. Results show that human variability and absenteeism dominate system performance, while machine downtime exerts a smaller but non-negligible effect. Human-centric release consistently outperforms machine-centric control by preventing infeasible releases to human-constrained stations, producing shorter throughput times and fewer late jobs. Under disruption scenarios, particularly when absenteeism and downtime co-occur, the resilience advantage of human-centric control becomes decisive. Dynamic workload norm adjustment further enhances stability by aligning release decisions with actual available capacity. The main contributions are threefold: (i) extension of workload control theory to dual-resource contexts through a dual-layer feasibility check, (ii) empirical validation of the compound impact of human- and machine-related disruptions, and (iii) operationalization of Industry 5.0 principles by providing transparent, explainable, and worker-aware release mechanisms. The findings offer both theoretical advancement and practical guidance for implementing robust production control in environments characterized by variability and uncertainty.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_10_AMIN.pdf
accessibile in internet per tutti
Dimensione
2.59 MB
Formato
Adobe PDF
|
2.59 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/10589/243920