The thesis addresses the issue of the assessment of ergonomic risks in manual handling of loads, an activity that is one of the main causes of musculoskeletal pathologies in work environments. The objective is to analyse and compare two distinct approaches: on the one hand, traditional methods based on technical standards such as UNI ISO 11228 and supported by computer software for writing the DVR; on the other hand, innovative models based on artificial intelligence; Capable of analysing real-time biometric data through sensors and artificial vision. The methodology includes a regulatory review, a theoretical analysis of evaluation tools, and a comparison with the experience gained during the internship at ZetaVu. The results show that traditional methods, while accessible and compliant, have limits in objectivity and adaptability. On the contrary, AI systems ensure greater precision, continuous monitoring and predictive analysis, while presenting critical issues related to costs, data management and the need for specific expertise. Finally, the thesis proposes an integrated approach, combining regulatory reliability and technological potential for a more precise, proactive and effective ergonomic risk assessment.
La tesi affronta il tema della valutazione dei rischi ergonomici nella movimentazione manuale dei carichi, attività che rappresenta una delle principali cause di patologie muscoloscheletriche nei contesti lavorativi. L’obiettivo è analizzare e confrontare due approcci distinti: da un lato, i metodi tradizionali basati su normative tecniche come la UNI ISO 11228 e supportati da software informatici per la redazione del DVR; dall’altro, i modelli innovativi basati sull’intelligenza artificiale, capaci di analizzare dati biometrici in tempo reale attraverso sensori e visione artificiale. La metodologia comprende una revisione normativa, un’analisi teorica degli strumenti di valutazione, e un confronto con l’esperienza maturata durante il tirocinio presso l’azienda ZetaVu. I risultati evidenziano che i metodi tradizionali, pur essendo accessibili e conformi alle norme, mostrano limiti in oggettività e adattabilità. Al contrario, i sistemi IA garantiscono maggiore precisione, monitoraggio continuo e analisi predittiva, pur presentando criticità legate ai costi, alla gestione dei dati e alla necessità di competenze specifiche. La tesi propone infine un approccio integrato, che combini l’affidabilità normativa e potenzialità tecnologiche per una valutazione dei rischi ergonomici più precisa, proattiva ed efficace.
METODI TRADIZIONALI E MODELLI BASATI SULL’INTELLIGENZA ARTIFICIALE PER LA VALUTAZIONE DEI RISCHI ERGONOMICI NELLA MOVIMENTAZIONE MANUALE DEI CARICHI
FILOGRASSO, NATHALY
2024/2025
Abstract
The thesis addresses the issue of the assessment of ergonomic risks in manual handling of loads, an activity that is one of the main causes of musculoskeletal pathologies in work environments. The objective is to analyse and compare two distinct approaches: on the one hand, traditional methods based on technical standards such as UNI ISO 11228 and supported by computer software for writing the DVR; on the other hand, innovative models based on artificial intelligence; Capable of analysing real-time biometric data through sensors and artificial vision. The methodology includes a regulatory review, a theoretical analysis of evaluation tools, and a comparison with the experience gained during the internship at ZetaVu. The results show that traditional methods, while accessible and compliant, have limits in objectivity and adaptability. On the contrary, AI systems ensure greater precision, continuous monitoring and predictive analysis, while presenting critical issues related to costs, data management and the need for specific expertise. Finally, the thesis proposes an integrated approach, combining regulatory reliability and technological potential for a more precise, proactive and effective ergonomic risk assessment.| File | Dimensione | Formato | |
|---|---|---|---|
|
FILOGRASSO_tesi.pdf
File consultabile da postazioni biblioteca
Dimensione
2.84 MB
Formato
Adobe PDF
|
2.84 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/108577