Abstract
Modern industries make large usage of digital and inter-connected devices to improve daily operations and functionalities. At the company level, the proper design and management of digital technologies ensure greater efficiency, higher coordination, and improved quality levels. Conversely, these technological devices bring inherent additional vulnerabilities at the informative level, with cascading effects on industrial plants. When referring to industrial systems engineering, the impact of cyber vulnerabilities must be studied in larger operating contexts, including the potential consequences on industrial physical processes causing relevant damages. The benefits of industrial plants’ digitalization must be traded-off with the higher chances of successful cyber attacks. To this extent, reactive strategies for managing industrial plants' physical failures must be complemented with proactive ones to model plants' resilience capacities. The incumbent presence of cyber vulnerabilities forces extending the repertoire of available risk preventive strategies for technical failures towards integrated cyber-physical models. When coupled with the actions performed by human operators, engineered aspects must be enlarged to cover an integrated socio-technical perspective, and finally embrace Cyber-Socio-Technical-Systems (CSTS) modelling. This project (RESIST, RESilience management to Industrial Systems Threats) proposes the design and development of an integrated Digital Twin (DT) to assess industrial plants resilience in spite of cyber threats, yet acknowledging human actions. The approach will be tested into an experimental plant simulating a typical oil and gas transportation system, where the cyber-physical DT will be coupled with a human DT. The integrated DT will involve human and hardware in the loop to better reproduce the CSTS in line with Internet of Things (IoT) paradigm. The result of the analysis will include the design of resilience metrics to be used for establishing priorities of interventions aimed at increasing system capacity to withstand and to recover from threats. The project encompasses an industrial systems engineering research dimension to capture system performance via advanced approaches, such as Ministero dell'Università e della Ricerca MUR - BANDO 2022 Systems-Theoretic Accident Model and Processes (STAMP), Motion Capture, and Mixed Reality. Globally, the project will deliver a novel methodology for cyber-socio-technical industrial resilience analyses. The outcomes of the project will be documented in academic publications and conferences usually attended by both practitioners and academics to foster larger dissemination. Furthermore, a set of guidelines about cyber-socio-technical modelling for industrial plants and resilience metrics definition will be developed. RESIST guidelines are meant to support future research in anticipating and testing the effects of cyber vulnerabilities on different industrial plants. RESULTS ACHIEVED The PRIN 2022 RESIST project (Resilience Management to Industrial Systems Threats) was conceived to address one of the most pressing challenges of contemporary industrial systems: how to assess and enhance resilience in Cyber-Socio-Technical Systems (CSTS) exposed to cyber-physical disruptions. The project developed and validated an integrated methodological and technological framework for resilience assessment grounded in systems theory, digital twin technologies, and human-in-the-loop simulation. Since modern industrial plants are no longer purely technical systems but tightly integrated cyber-physical infrastructures where digital artefacts, automation logic, sensors, actuators, human operators, and organizational procedures interact dynamically. Building on the CSTS conceptualization and resilience theory, RESIST aimed to move beyond traditional risk management approaches toward a systemic and quantitative assessment of how industrial systems anticipate, absorb, adapt to, and recover from cyber-induced disturbances. At its core, RESIST designed and implemented a multi-layer Digital Twin (DT) architecture composed of: - A Cyber-Physical Digital Twin of an experimental oil and gas plant, replicating pump-ejector dynamics, vertical tank behavior, and PID-controlled processes. - A Human Digital Twin, developed through motion capture technology and behavioral modelling to represent operator actions within the plant environment. - A Human-Hardware-in-the-Loop (HHIL) simulation framework, integrating plant and human models within a STAMP-based control structure to enable systemic resilience assessment under cyber-attack scenarios. The project adopted STAMP/STPA modelling to define the Safety Control Structure of the system and identify unsafe control actions and potential loss scenarios. Cyber-attack classifications and resilience quantification metrics were systematically reviewed and operationalized. Artificial Intelligence techniques (e.g., neural networks, Gradient Boosted Trees) were incorporated to enhance predictive capabilities and anomaly detection within the Digital Twin. Following these activities, RESIST advanced toward multi-scenario simulations, addressing methodological challenges such as time-series misalignment and performance-area normalization. These simulations enabled the quantitative comparison between nominal and disrupted operational states and supported the definition of redesign actions aimed at strengthening systemic resilience. The resulting guidelines were consolidated through stakeholder engagement and survey-based validation. Overall, RESIST produced a transferable and scalable methodological framework for resilience quantification in CSTSs, combining systemic modelling, AI-enhanced Digital Twins, and human-centric simulation. The project contributes to the advancement of Industry 4.0 and Industry 5.0 paradigms by operationalizing resilience as a measurable and design-oriented property of complex industrial ecosystems.
Dettagli del progetto
Responsabile scientifico: Marco Bortolini
Strutture Unibo coinvolte:
Dipartimento di Ingegneria Industriale
Coordinatore:
"Sapienza" Universita' Di Roma(Italy)
Contributo totale Unibo: Euro (EUR) 63.024,00
Durata del progetto in mesi: 24
Data di inizio
28/09/2023
Data di fine:
28/02/2026
