Risk estimation and management have become critical components of decision-making processes across various industries. As businesses and organizations navigate increasingly complex operational environments, understanding and managing risks is essential for ensuring both safety and productivity. The rapid advancements in technology, particularly in machine learning and artificial intelligence (AI), have opened new avenues for improving risk estimation. Emerging methods such as Digital Twins (DT), Natural Language Processing (NLP), Bayesian Networks (BN), and Deep Learning (DL) offer various benefits to be used for risk estimation. This thesis explores those recent methodologies developed for risk estimation. Each of these methods brings unique strengths to risk analysis but in the meanwhile there are still future directions and opportunities that needs to be addressed.This thesis aims to present a comprehensive understanding of the state-of-the-art in risk estimation. Finally, it concludes by proposing a guidance on selecting the most appropriate method depending on the specific requirements of an industry, the availability of data, and the goals of risk management.
Le stime e la gestione del rischio sono diventate componenti fondamentali dei processi decisionali in vari settori. Man mano che aziende e organizzazioni operano in ambienti operativi sempre più complessi, comprendere e gestire i rischi è essenziale per garantire sia la sicurezza che la produttività. I rapidi progressi tecnologici, in particolare nel campo del machine learning e dell'intelligenza artificiale (AI), hanno aperto nuove possibilità per migliorare le stime del rischio. Metodi emergenti come i Digital Twins (DT), l'elaborazione del linguaggio naturale (NLP), le reti bayesiane (BN) e il deep learning (DL) offrono vari benefici per essere utilizzati nella stima del rischio. Questa tesi esplora queste recenti metodologie sviluppate per la stima del rischio. Ognuno di questi metodi apporta punti di forza unici all'analisi del rischio, ma al contempo ci sono ancora direzioni future e opportunità che devono essere affrontate. L'obiettivo di questa tesi è fornire una comprensione completa dello stato dell'arte nella stima del rischio. Infine, si conclude proponendo una guida per la scelta del metodo più appropriato a seconda delle esigenze specifiche di un settore, della disponibilità dei dati e degli obiettivi della gestione del rischio.
Methods of Risk Estimation in Production Systems- Literature Review
Abdollahi Pour, Mahdi
2023/2024
Abstract
Risk estimation and management have become critical components of decision-making processes across various industries. As businesses and organizations navigate increasingly complex operational environments, understanding and managing risks is essential for ensuring both safety and productivity. The rapid advancements in technology, particularly in machine learning and artificial intelligence (AI), have opened new avenues for improving risk estimation. Emerging methods such as Digital Twins (DT), Natural Language Processing (NLP), Bayesian Networks (BN), and Deep Learning (DL) offer various benefits to be used for risk estimation. This thesis explores those recent methodologies developed for risk estimation. Each of these methods brings unique strengths to risk analysis but in the meanwhile there are still future directions and opportunities that needs to be addressed.This thesis aims to present a comprehensive understanding of the state-of-the-art in risk estimation. Finally, it concludes by proposing a guidance on selecting the most appropriate method depending on the specific requirements of an industry, the availability of data, and the goals of risk management.File | Dimensione | Formato | |
---|---|---|---|
Thesis- Mahdi Abdollahi Pour-2024.pdf
accessibile in internet per tutti
Descrizione: Thesis Text
Dimensione
1.91 MB
Formato
Adobe PDF
|
1.91 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/10589/226751