Industry 5.0 attempts to overcome a key limitation of Industry 4.0: its techno-economic focus often overlooks the human component in smart factories. The new approach fosters the integration of Industry 4.0 technologies with a human-centric and sustainable vision. A notable example is the “Healthy Operator” concept, where wearable devices like smart bracelets or goggles are used to monitor stress-related body signals. Early detection of stress in industrial environments can enhance both operators’ well-being and performance, ultimately benefiting company profitability. This study examines the reliability of wearables and stress-related metrics in detecting acute work content-related stress. First, a systematic literature review explores how stress is elicited and measured with wearables in laboratory or real-life experiments. Findings highlight the broad range of stress-inducing tasks and assessment methods in current research but reveal limited attention to the multidimensional nature of stress, which extends beyond excessive workload to environmental, social, and work pace factors according to health and safety organizations. Second, the study presents the protocol and results of a laboratory experiment involving 28 participants assembling a valve under three conditions: Rest, Stress 1 (primary task with low noise), and Stress 2 (dual task with mental computations and high noise). Objective measures include task performance and several physiological signals (pupil size, blink rate, electrodermal activity - EDA, heart rate, and heart rate variability - HRV) recorded via three wearable devices. Subjective stress levels are assessed through the NAS and NASA-TLX questionnaires. Results indicate that performance and psychological well-being clearly worsen from Stress 1 to Stress 2. On the physiological side, EDA, heart rate, pupil size, and blink rate reliably differentiate conditions and align with subjective perceptions and performance trends, whereas HRV proves far less reliable and dissociated from psychological stress. Suggestions to future research involve the integration of such measuring systems in real industrial environments, and the adoption of a more individual approach to stress assessment, since subjectivity in physiological response proved to be high throughout the participants.
L'Industria 5.0 mira a superare un limite dell'Industria 4.0: il suo focus tecno-economico spesso trascura la componente umana. Questo nuovo approccio promuove l'integrazione delle tecnologie 4.0 con una visione più centrata sull'uomo e sulla sostenibilità. Un esempio concreto è il concetto di "Healthy Operator", in cui dispositivi indossabili vengono utilizzati per monitorare i segnali corporei legati allo stress. L’individuazione precoce dello stress negli ambienti industriali può migliorare sia il benessere che le prestazioni degli operatori, con benefici anche in termini di redditività. Questo studio esamina l'affidabilità dei dispositivi indossabili e delle metriche legate allo stress nella rilevazione dello stress acuto correlato al lavoro. In primo luogo, una revisione sistematica della letteratura esplora come lo stress viene indotto e misurato attraverso dispositivi indossabili in esperimenti di laboratorio o contesti reali. I risultati evidenziano una grande varietà di compiti stressanti e metodi di valutazione nella ricerca attuale, ma rivelano una scarsa attenzione alla natura multidimensionale dello stress. Secondo le organizzazioni per la salute e la sicurezza, infatti, lo stress non dipende solo dal carico di lavoro, ma anche da fattori ambientali, sociali e legati ai ritmi di produzione. In secondo luogo, lo studio presenta il protocollo e i risultati di un esperimento di laboratorio, dove 28 partecipanti hanno assemblato una valvola in tre condizioni: Riposo, Stress 1 (solo assemblaggio e basso rumore) e Stress 2 (doppio compito con calcoli mentali e alto rumore). Le metriche “oggettive” includono le prestazioni e diversi parametri fisiologici (dimensione della pupilla, frequenza di ammiccamento, conduttività cutanea - EDA, frequenza cardiaca e variabilità della frequenza cardiaca - HRV) registrati tramite tre dispositivi indossabili. I livelli soggettivi di stress sono valutati attraverso i questionari NAS e NASA-TLX. I risultati indicano un chiaro peggioramento delle prestazioni e del benessere psicologico passando da Stress 1 a Stress 2. Dal punto di vista fisiologico, EDA, frequenza cardiaca, dimensione della pupilla e frequenza di ammiccamento si sono rivelati indicatori affidabili dello stress e coerenti con le percezioni soggettive, mentre l’HRV si dimostra molto meno affidabile e poco correlata allo stress psicologico. Le indicazioni per la ricerca futura riguardano l’integrazione di tali sistemi di misurazione negli ambienti industriali reali e l'adozione di un approccio più personalizzato alla valutazione dello stress, poiché la risposta fisiologica ha mostrato un’elevata variabilità individuale tra i partecipanti.
Human operator stress detection in industry 5.0: a comparative analysis of physiological, performance, and psychological measurements in an assembly experiment
Biroli, Moreno;MISELLI, FEDERICO
2023/2024
Abstract
Industry 5.0 attempts to overcome a key limitation of Industry 4.0: its techno-economic focus often overlooks the human component in smart factories. The new approach fosters the integration of Industry 4.0 technologies with a human-centric and sustainable vision. A notable example is the “Healthy Operator” concept, where wearable devices like smart bracelets or goggles are used to monitor stress-related body signals. Early detection of stress in industrial environments can enhance both operators’ well-being and performance, ultimately benefiting company profitability. This study examines the reliability of wearables and stress-related metrics in detecting acute work content-related stress. First, a systematic literature review explores how stress is elicited and measured with wearables in laboratory or real-life experiments. Findings highlight the broad range of stress-inducing tasks and assessment methods in current research but reveal limited attention to the multidimensional nature of stress, which extends beyond excessive workload to environmental, social, and work pace factors according to health and safety organizations. Second, the study presents the protocol and results of a laboratory experiment involving 28 participants assembling a valve under three conditions: Rest, Stress 1 (primary task with low noise), and Stress 2 (dual task with mental computations and high noise). Objective measures include task performance and several physiological signals (pupil size, blink rate, electrodermal activity - EDA, heart rate, and heart rate variability - HRV) recorded via three wearable devices. Subjective stress levels are assessed through the NAS and NASA-TLX questionnaires. Results indicate that performance and psychological well-being clearly worsen from Stress 1 to Stress 2. On the physiological side, EDA, heart rate, pupil size, and blink rate reliably differentiate conditions and align with subjective perceptions and performance trends, whereas HRV proves far less reliable and dissociated from psychological stress. Suggestions to future research involve the integration of such measuring systems in real industrial environments, and the adoption of a more individual approach to stress assessment, since subjectivity in physiological response proved to be high throughout the participants.File | Dimensione | Formato | |
---|---|---|---|
2025_04_Biroli_Miselli_Tesi_01.pdf
non accessibile
Descrizione: Testo tesi
Dimensione
3.53 MB
Formato
Adobe PDF
|
3.53 MB | Adobe PDF | Visualizza/Apri |
2025_04_Biroli_Miselli_Executive Summary_02.pdf
non accessibile
Descrizione: Executive summary
Dimensione
729.61 kB
Formato
Adobe PDF
|
729.61 kB | 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/10589/233952