Project details
Title:
A Human-centric IoE-based Framework for Supporting the Transition Towards Industry 5.0
Acronym:
HOMEY
Type:
National (PRIN 2022)
Start Date:
September 28, 2025
End Date:
February 28, 2026
Principal Investigator:
Associate professor, Emanuele Storti (UNIVPM Unit Responsible and sub-PI)
Other Units Involved:
University of Pavia, University of Milan-Bicocca
Keywords:
Industry 5.0; Internet of Everything; Knowledge Graphs; Big Data management; Human-machine interaction
Description:
Industry 5.0 is a novel paradigm identifying the transition from traditional industries towards smart, human-centric, and green-aware industrial ecosystems. Production chains must be always focused on the experience and needs of human beings for both the introduction of creativity and innovation on the production side, and the personalization of offered products on the customer side.
Production environments must be deeply adjusted in such a way to delegate heavy and repetitive tasks to more and more autonomous devices and machines. Interaction between humans and smart devices has been consolidated by the Internet of Things, whose extension, called the Internet of Everything (IoE), involves humans, devices, processes and data into a single strongly interconnected environment.
In this context, the HOMEY project proposes a comprehensive framework that leverages the Internet of Everything (IoE), an evolution of the IoT interconnecting devices, people, data and processes, to create intelligent, adaptive, and human-centric industrial environments.
The proposed solution underlies research objectives focused on specific aspects of the human-centric industrial solution and organized in four main Work-Packages:
– WP1: INDUSTRIAL IoE DEVELOPMENT
Focused on the definition of data models, integration strategies and querying mechanisms to represent and extract knowledge from an industrial IoE, with a human-centric and context-based approach. In particular, processed data will be modeled through a
knowledge graph and the human-centric view will be obtained by projecting suitable context-aware Ego-networks from it.
– WP2: HUMAN-CENTRIC IMMERSIVE WORKING ENVIRONMENT DEVELOPMENT
Devoted to the construction of an immersive digital solution to enable, through Augmented Reality techniques, interactions between
the human resource and the digital industrial environment. Studied solutions will leverage non-invasive and lightweight hardware
equipment.
– WP3: PERSONALIZED RECOMMENDATION FOR TASK EXECUTION AND TEAM BUILDING
Dedicated to build a support system to provide personalized recommendation to human resources inside a factory. In particular,
solutions for activity detection and physical status measurement will be developed by leveraging data possibly gathered from
wearable sensors. The processing of such data will be carried out adopting strategies to minimize privacy impacts.
– WP4: SYSTEM INTEGRATION AND TESTING
Devoted to the integration and testing of all developed solutions with a focus on real-life industrial settings.
The involved Units have a strong expertise in the research fields related to the three objectives of this project and in the practical
implementation of possible solutions related to them.
Objectives:
The project is structured around three main objectives:
– O.1) Design and implementation of an industrial IoE framework that enables seamless interaction among humans, machines, and data. Semantic-based approaches for data management, access and monitoring are based on Knowledge Graphs to ensure interoperability, while context-aware mechanisms for data extraction aim to provide each worker with a personalized view of relevant information.
Security is guaranteed by role-based access and privacy-preserving anomaly detection mechanisms based on behavioral fingerprinting and edge computing, ensuring both security and scalability.
– O.2) Definition and design of human-centric immersive digital working environment through Augmented Reality and zero-touch interfaces.
Wearable devices with sensors and lightweight Machine Learning enable gesture control and real-time feedback for an intuitive, ergonomic experience.
– O.3) Personalized recommendation for task execution and team building.
In the reference scenario, employees are equipped with wearable devices, which can monitor their activities and their physical status during working hours in a privacy-preserving manner.
The system will assess risk levels for tasks and perform optimal tasks assignment based on the monitored worker’s stress level, effort of the task and other organizational constraints. Additionally, a team-building module will recommend suitable agents on contextual data, availability, and required skills. Together, these components will contribute to a safer, more efficient, and more adaptive industrial environment.
Application Contexts:
The HOMEY project lies at the intersection of digital transformation, human-machine collaboration, and industrial innovation. It addresses the critical shift from Industry 4.0, which primarily emphasized automation and data exchange, to Industry 5.0, which focuses on human-centric, resilient, and sustainable industrial ecosystems.
At the core of the application context is the Internet of Everything (IoE) paradigm, which expands the scope of IoT by integrating not only devices, but also people, processes, and data into a single, dynamic, and highly interactive network. In this setting, the HOMEY project aims to transform traditional factories into intelligent, adaptive, and human-friendly environments, where human workers are empowered by digital tools rather than replaced.
The HOMEY framework is applied in industrial contexts where:
– Data-driven decision-making is needed for enhancing productivity and safety;
– Augmented Reality (AR) can support immersive, real-time interaction with digital representations of physical systems;
– Wearable technologies can collect and interpret physiological and positional data from human operators;
– Intelligent recommender systems can optimize task allocation and team formation based on worker status, preferences, and capabilities;
– Knowledge Graphs and Ego-networks can deliver personalized and context-aware access to information.
This framework is particularly relevant for industries that demand high customization, complex human-machine coordination, and continuous adaptation—such as smart manufacturing, logistics, aerospace, and healthcare. The project responds to industrial needs for increased worker well-being and safety, reduced operational costs via data harmonization and context-aware automation, and enhanced decision-making through personalized information access and predictive analytics.
Expected Results:
The activities carried out to cope with the three research objectives will produce the following outcomes:
– Integration of vocabularies/ontologies for knowledge representation of IoE data sources and definition of data acquisition
mechanisms.
– Definition of APIs for IoE graph querying and Ego-network generation.
– Definition of fully distributed behavioral fingerprinting solutions and anomaly detection solutions.
– Prototyping of the IEF and its testing in a real context.
– Definition and standardization of interactions through wearable devices with the IEF and the Ego-network.
– Collection of the subset of interactions (gestures) with wearable devices with the scope of creating training data for Machine
Learning algorithms.
– Design, implementation and testing of Machine Learning techniques for arm gesture recognition using wearable devices.
– Designing a system allowing to recognize, extract and manage a subset of patterns derived from activity and physical status
recognition.
– Development of a context-based risk model for tasks and a personalized task recommendation system.
– Development of a solution for the allocation of human resources through personalized teammate recommendation.
Achivied Results:
(Preliminary results)
The initial results for O.1 include the definition of the metadata model of the IoE network, represented as a Knowledge Graph (KG) based on the SemIoE ontology. SemIoE is an OWL2 ontology built by integrating several modules (e.g., SSN/SOSA, BOT, ORG) designed to provide a structured and standardized description of entities (agents, roles, smart devices, locations, access rights, preferences) and their relations, thereby supporting semantic interoperability at IoE level.
On top of it, a micro-service architecture aims to deliver basic functionalities, such as authentication and authorization, and advanced capabilities. The Data Gathering platform is responsible for collecting heterogeneous data streams from a variety of sources, including traditional IoT sensors, smart objects, wearable devices and IT modules. It supports customized stream pre-processing (e.g., filtering, decryption, decompression), data stream collection, data post-processing (e.g., aggregation) and persistent storage.
Semantic-based Monitoring and Querying services allow users to access both real-time and historical data by formulating request using the KG terminology. These services enforce context-aware and role-based access control to ensure secure access only to data relevant to the user’s location and assigned tasks.
Anomaly Detection is employed to monitor the behavior of devices deployed within the system and includes a privacy-preserving delegation mechanism, identifying the most likely sequences of communication packets and detect anomalies based on the model’s prediction errors.
The O.2 explores secure teleoperation of a robotic arm via IMUs worn by the user, namely a biometric authentication system based on logistic regression ensuring access control, and an arm gesture recognition system running entirely on a consumer-grade smartwatch, eliminating the need for cloud processing.
An outcome of O.3 consists in the definition of a task recommendation module, which dynamically reallocates activities among workers. The module balances efficiency and sustainability through a flexible and periodic negotiation process, allowing workers to refuse an activity if it exceeds a sustainable stress level, as monitored via wearable devices.
The system is modeled using Mixed Integer Linear Programming (MILP) with a hierarchical objective function, aimed at first maximizing the number of assignments and then minimizing the cost due to reassignments, levels of stress and possible overtimes.
Publications:
– M. Arazzi, A. Nocera, E. Storti, “The semioe ontology: A semantic model solution for an ioe-based industry”, IEEE Internet of Things Journal 11 (2024) 40376–40387.
– M. M. Sciarroni, M. Esposito, P. Pierleoni, E. Storti, “Monitoring data streams in industry 5.0: a knowledge graph approach”, in: 2024 IEEE 8th Forum on Research and Technologies for Society and Industry Innovation (RTSI), IEEE, 2024, pp. 566–571.
– M. Esposito, M. Marconi Sciarroni, T. Fava, A. Belli, L. Palma, E. Storti, P. Pierleoni, “Experimental Evaluation of End-to-End Security Protocols for the Internet of Everything”, in: 2024 IEEE 8th Forum on Research and Technologies for Society and Industry Innovation (RTSI), IEEE, 2024, pp. 13-18.
– M. Arazzi, S. Nicolazzo, A. Nocera, “A fully privacy-preserving solution for anomaly detection in iot using federated learning and homomorphic encryption”, Information Systems Frontiers (2023) 1–24.
– I. E. Stan, H. Amrani, P. Napoletano, D. D’Auria, “Authenticated robotic teleoperation with task recognition”, IEEE Consumer Electronics Magazine (2025).
– M. Arazzi, S. Nicolazzo, and A. Nocera. “A novel IoT trust model leveraging fully distributed behavioral fingerprinting and secure delegation”, Pervasive and Mobile Computing 99 (2024): 101889.
– A. Colombo, L. Celona, S. Bianco, A. Nocera, and P. Napoletano, “Arm gesture recognition with smartwatches,” in 2024 IEEE 8th Forum on
Research and Technologies for Society and Industry Innovation (RTSI), IEEE, 2024, pp. 625–629.
– C. Diamantini, O. Pisacane, D. Potena, E. Storti, “Personalized task reassignment in industry 5.0: A milp-based solution approach”, in: Proceedings of the 27th International Conference on Enterprise Information Systems – Volume 2: ICEIS, INSTICC, SciTePress, 2025, pp. 813–820.
