Traditional production planning and control systems, specifically Workload Control in Dual-Resource Constrained manufacturing, typically rely on a Machine-Centric paradigm, neglecting the dynamic influence of human availability and performance. This oversight leads to "phantom availability" and systemic fragility when faced with realistic volatility. Advancing the principles of Industry 5.0, this thesis investigates the performance and resilience of a Human-Centric Order Release mechanism designed to explicitly model both machine and human capacity constraints. A discrete-event simulation model of a high-load, 5-station General Flow Shop was developed to conduct a full factorial experiment. The study compared the Human-Centric release rule against the traditional Machine-Centric rule across six different intelligent dispatch rules, under systemic disruptions, specifically worker absenteeism and machine downtime. The results demonstrate a complex yet crucial performance profile. The HC rule provided superior flow efficiency and higher delivery reliability across all dispatch rules, validating its efficacy in eliminating phantom capacity. However, a significant trade-off was identified: while the Human-Centric rule reduced the frequency of late orders, it increased the magnitude of the average delay, confirming a strategic choice between reducing failure frequency and mitigating failure severity. Crucially, the analysis confirmed a synergistic interaction between the Human-Centric release rule and high-performing dispatch rules, maximizing flow benefits. Under systemic disruption, the Human-Centric model was fundamentally more resilient, successfully preventing the catastrophic collapse in performance observed in the non-adaptive Machine-Centric model, particularly under high worker absenteeism. This research empirically validates the necessity of Human-Centric control logic for achieving the stability and resilience required by Industry 5.0 in volatile manufacturing environments.
I tradizionali sistemi di pianificazione e controllo della produzione, in particolare il Workload Control in contesti manifatturieri con vincoli doppi di risorse, si basano tipicamente su un paradigma Machine-Centric, trascurando l’influenza dinamica della disponibilità e delle prestazioni umane. Questa mancanza porta a una “disponibilità fantasma” e a una fragilità sistemica di fronte alla volatilità reale. In linea con i principi dell’Industria 5.0, questa tesi indaga le prestazioni di un meccanismo di rilascio ordini Human-Centric, progettato per modellare i vincoli di capacità sia delle macchine che delle risorse umane. È stato sviluppato un modello di simulazione ad eventi discreti di un General Flow Shop a 5 stazioni e ad alto carico, utilizzato per condurre un esperimento fattoriale completo. Lo studio ha confrontato la regola di rilascio Human-Centric con la tradizionale regola Machine-Centric attraverso sei differenti regole di dispatching, in presenza di perturbazioni sistemiche, in particolare assenteismo dei lavoratori e downtime delle macchine. I risultati mostrano un profilo prestazionale complesso ma cruciale. La regola Human-Centric ha garantito una maggiore efficienza di flusso e una più alta affidabilità nelle consegne in tutte le regole di dispatching, validando la sua efficacia nell’eliminare la capacità fantasma. Tuttavia, è stato identificato un compromesso significativo: se da un lato la regola Human-Centric riduce la frequenza degli ordini in ritardo, dall’altro aumenta l’entità media del ritardo, confermando l’esistenza di una scelta strategica tra ridurre la frequenza dei fallimenti e mitigarne la gravità. Inoltre, l’analisi ha confermato una sinergia tra la regola di rilascio Human-Centric e le regole di dispatching ad alte prestazioni, massimizzando i benefici di flusso. In condizioni di perturbazione sistemica, il modello Human-Centric si è dimostrato fondamentalmente più resiliente, prevenendo con successo il collasso delle prestazioni osservato nel modello Machine-Centric, in particolare in presenza di elevato assenteismo dei lavoratori. Questa ricerca fornisce una validazione empirica della necessità di una logica di controllo Human-Centric per raggiungere la stabilità e la resilienza richieste dall’Industria 5.0 in ambienti manifatturieri caratterizzati da elevata volatilità.
Human-centric for industry 5.0: the impact of dispatch rules on dual-resource constrained systems
TORELLI, DANILO
2024/2025
Abstract
Traditional production planning and control systems, specifically Workload Control in Dual-Resource Constrained manufacturing, typically rely on a Machine-Centric paradigm, neglecting the dynamic influence of human availability and performance. This oversight leads to "phantom availability" and systemic fragility when faced with realistic volatility. Advancing the principles of Industry 5.0, this thesis investigates the performance and resilience of a Human-Centric Order Release mechanism designed to explicitly model both machine and human capacity constraints. A discrete-event simulation model of a high-load, 5-station General Flow Shop was developed to conduct a full factorial experiment. The study compared the Human-Centric release rule against the traditional Machine-Centric rule across six different intelligent dispatch rules, under systemic disruptions, specifically worker absenteeism and machine downtime. The results demonstrate a complex yet crucial performance profile. The HC rule provided superior flow efficiency and higher delivery reliability across all dispatch rules, validating its efficacy in eliminating phantom capacity. However, a significant trade-off was identified: while the Human-Centric rule reduced the frequency of late orders, it increased the magnitude of the average delay, confirming a strategic choice between reducing failure frequency and mitigating failure severity. Crucially, the analysis confirmed a synergistic interaction between the Human-Centric release rule and high-performing dispatch rules, maximizing flow benefits. Under systemic disruption, the Human-Centric model was fundamentally more resilient, successfully preventing the catastrophic collapse in performance observed in the non-adaptive Machine-Centric model, particularly under high worker absenteeism. This research empirically validates the necessity of Human-Centric control logic for achieving the stability and resilience required by Industry 5.0 in volatile manufacturing environments.| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi_10710704.pdf
solo utenti autorizzati a partire dal 13/11/2026
Descrizione: Documento principale
Dimensione
1.35 MB
Formato
Adobe PDF
|
1.35 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/10589/247532