FAIR: Fuzzy-BDI agents for intelligent reasoning on smart cyber-physical systems
EXPERT SYSTEMS WITH APPLICATIONS, vol.299, 2026 (SCI-Expanded, Scopus)
- Publication Type: Article / Article
- Volume: 299
- Publication Date: 2026
- Doi Number: 10.1016/j.eswa.2025.130047
- Journal Name: EXPERT SYSTEMS WITH APPLICATIONS
- Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Compendex, INSPEC, Public Affairs Index
- Keywords: Fuzzy logic, BDI Agents, Multi-agent systems, Smart cyber-physical systems
- Dokuz Eylül University Affiliated: No
Abstract
Traditional engineering often struggles to cope with uncertainty in dynamic systems. To address this limitation, this study presents FAIR, a novel framework that tightly integrates fuzzy logic into the reasoning engine of Jason AgentSpeak to enhance decision-making in complex systems. By extending the Belief-Desire-Intention (BDI) model with fuzzy reasoning capabilities, FAIR enables run-time reasoning with vague or incomplete data under uncertainty at run-time. Moreover, FAIR extends the IEC-61131-7 standard to support fuzzy-BDI syntax. The framework is evaluated across four diverse domains, such as CPS, internet of things (IoT), house robotics, and video gaming. Experimental results show that fuzzy-BDI agents significantly outperform traditional BDI agents. These findings highlight FAIR's potential as a practical engineering approach for developing intelligent, adaptive agents capable of operating under real-world uncertainty.