A Generative Artificial Intelligence–Driven Industry 5.0 Framework for Adaptive Human-Centric Industrial Transportation Systems

Fazlollahtabar, Hamed; Stevic, Zeljko

doi:10.22068/ijiepr.37.3.2672

Volume 37, Issue 3 (IJIEPR- In Progress 2026) IJIEPR 2026, 37(3): 145-158 | Back to browse issues page

‎ 10.22068/ijiepr.37.3.2672

Mendeley

Zotero

RefWorks

Fazlollahtabar H, Stevic Z. A Generative Artificial Intelligence–Driven Industry 5.0 Framework for Adaptive Human-Centric Industrial Transportation Systems. IJIEPR 2026; 37 (3) :145-158
URL: http://ijiepr.iust.ac.ir/article-1-2672-en.html

A Generative Artificial Intelligence–Driven Industry 5.0 Framework for Adaptive Human-Centric Industrial Transportation Systems

Hamed Fazlollahtabar¹

, Zeljko Stevic²

1- Department of Industrial Engineering, School of Engineering, Damghan University, Damghan, Iran , h.fazl@du.ac.ir
2- 2Faculty of Transport and Traffic Engineering, University of East Sarajevo, Vojvode Mišića 52, 74000 Doboj, Bosnia and Herzegovina, 3Department of Industrial Management Engineering, Korea University, 145 Anam-Ro, Seongbuk-Gu, Seoul 02841, Republic of Korea

Abstract: (112 Views)

The transition from Industry 4.0 to Industry 5.0 has shifted the focus of intelligent manufacturing systems from automation-centric optimization toward human-centricity, resilience, and sustainability. Industrial transportation systems—comprising autonomous mobile robots, automated guided vehicles, collaborative robots, and human operators—operate under significant uncertainty arising from dynamic demand, shared human–robot workspaces, and potential safety-critical disruptions. Existing transportation optimization approaches remain largely reactive and rely on stochastic or worst-case uncertainty modeling, limiting their ability to anticipate complex, correlated disruption scenarios. This paper proposes a novel Generative Artificial Intelligence (GenAI)–driven Industry 5.0 framework for adaptive optimization of industrial transportation systems. The proposed approach integrates a two-phase decision-making model: (i) a performance maximization phase combining reinforcement learning and mixed-integer linear programming to optimize routing, scheduling, and human workload balance, and (ii) a risk minimization phase leveraging GenAI-based scenario generation and Bayesian–fuzzy reasoning to proactively mitigate operational and safety risks. Unlike traditional Monte Carlo simulation, the GenAI module learns the underlying structure of disruption patterns from historical and real-time data, enabling the generation of realistic, high-impact scenarios. A comprehensive case study in a robotic-enabled manufacturing facility, validated through 20 independent simulation runs, demonstrates that the proposed framework increases transportation throughput by 21.7%, improves energy efficiency by 11.7%, and reduces safety incidents by 39.9% compared to a conventional Industry 4.0 baseline. Statistical tests confirm the robustness and significance of these improvements. The results highlight the potential of integrating Generative AI with Industry 5.0 principles to enable proactive, risk-aware, and human-centric industrial transportation systems.

Keywords: Industry 5.0, Generative artificial intelligence, Industrial transportation systems, Risk-aware decision-making, Bayesian–fuzzy modeling

Full-Text [PDF 641 kb] (52 Downloads)

Type of Study: Research | Subject: Logistic & Apply Chain
Received: 2025/12/28 | Accepted: 2026/07/1 | Published: 2026/09/8