With the global population aging, healthcare assistance for elderly people is becoming increasingly important. Information technologies can support healthcare by monitoring activities of daily living (ADLs) and providing useful insights into behavioural patterns. However, collecting and analysing large amounts of human behavioural data remains an expensive and time-consuming task. This thesis presents the development of a graphical user interface for NeeMAS (Needs-based Multi-Agent Simulator), an agent-based simulator designed to reproduce human behaviour in virtual environments using digital twins. The simulator models physiological needs and social interactions through a set of configurable parameters that drive agent decision-making and movement within a virtual environment. The graphical interface was designed and implemented to simplify the environment creation, parameter configuration, simulation execution and result analysis. To generate realistic agent profiles, a questionnaire targeting elderly individuals was used. The resulting profiles were then employed to configure and validate the simulator through individual, social and wandering behavior scenarios, demonstrating the effectiveness of both the parameter generation workflow and the graphical interface. The developed framework provides a flexible tool for behavioural simulation and represents a step towards the creation of realistic digital twins for assisted living environments and behavioural analysis applications.
Con l’invecchiamento della popolazione mondiale, l’assistenza sanitaria per le persone anziane sta assumendo un’importanza sempre crescente. Le tecnologie dell’informazione possono supportare il settore sanitario attraverso il monitoraggio delle attività della vita quotidiana (Activities of Daily Living, ADL) e la fornitura di informazioni utili sui modelli comportamentali. Tuttavia, la raccolta e l’analisi di grandi quantità di dati comportamentali umani rimangono attività costose e dispendiose in termini di tempo. Questa tesi presenta lo sviluppo di un’interfaccia grafica per NeeMAS (Needs-based Multi-Agent Simulator), un simulatore basato su agenti progettato per riprodurre il comportamento umano in ambienti virtuali mediante digital twin. Il simulatore modella i bisogni fisiologici e le interazioni sociali attraverso un insieme di parametri configurabili che guidano le decisioni degli agenti e i loro movimenti all’interno dell’ambiente virtuale. L’interfaccia grafica è stata progettata e implementata per semplificare la creazione dell’ambiente, la configurazione dei parametri, l’esecuzione delle simulazioni e l’analisi dei risultati. Per generare profili realistici degli agenti è stato realizzato un questionario rivolto alle persone anziane. I profili ottenuti sono stati successivamente impiegati per configurare e validare il simulatore attraverso scenari individuali, sociali e di wandering, dimostrando l’efficacia sia del processo di generazione dei parametri sia dell’interfaccia grafica. Il framework sviluppato fornisce uno strumento flessibile per la simulazione comportamentale e rappresenta un passo verso la creazione di digital twin realistici per scenari di vita assistita e applicazioni di analisi comportamentale.
Design and implementation of a graphical interface for NeeMAS: a multi-agent behavioral simulator driven by needs
Previtera, Marco Luca
2025/2026
Abstract
With the global population aging, healthcare assistance for elderly people is becoming increasingly important. Information technologies can support healthcare by monitoring activities of daily living (ADLs) and providing useful insights into behavioural patterns. However, collecting and analysing large amounts of human behavioural data remains an expensive and time-consuming task. This thesis presents the development of a graphical user interface for NeeMAS (Needs-based Multi-Agent Simulator), an agent-based simulator designed to reproduce human behaviour in virtual environments using digital twins. The simulator models physiological needs and social interactions through a set of configurable parameters that drive agent decision-making and movement within a virtual environment. The graphical interface was designed and implemented to simplify the environment creation, parameter configuration, simulation execution and result analysis. To generate realistic agent profiles, a questionnaire targeting elderly individuals was used. The resulting profiles were then employed to configure and validate the simulator through individual, social and wandering behavior scenarios, demonstrating the effectiveness of both the parameter generation workflow and the graphical interface. The developed framework provides a flexible tool for behavioural simulation and represents a step towards the creation of realistic digital twins for assisted living environments and behavioural analysis applications.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_07_Previtera.pdf
accessibile in internet per tutti
Dimensione
6.14 MB
Formato
Adobe PDF
|
6.14 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/10589/260583