Scholarlink

This paper presents a human‑centric Product Lifecycle Management (PLM) framework aligned with Industry 5.0 that brings together Digital Twins (DTs), Smart Manufacturing, and Generative AI (GenAI) under a PLM backbone. The framework is designed around three pillars—human‑centricity, sustainability, and resilience—and incorporates explainable AI (XAI) and lifecycle assessment to support transparent, accountable decisions. A focused systematic literature review motivates the architecture and identifies gaps in explainability, interoperability, and end‑to‑end lifecycle coverage. The study demonstrates feasibility with a NASA milling case study, integrating sensor data, DT‑based emulation, and a predictive model within PLM workflows to achieve traceability and operator‑in‑the‑loop validation. Results show modest predictive performance, highlighting the need for feature engineering and model generalization; the paper outlines concrete improvement paths and discuss industrial adoption barriers (skills, integration, and governance). Contributions are: (i) a precise definition of PLM 5.0 and its dependencies, (ii) a modular framework integrating DT/GenAI/XAI with PLM and sustainability assessment, and (iii) a reproducible case configuration that maps each operational step to the framework. The paper concludes with open challenges in interoperability, security, and ethical governance and proposes evaluation and benchmarking directions.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Integrating Product Lifecycle Management, Business Innovation, and Sustainability in the Era of Industry 5.0: A Synergistic Framework

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Marcelo Feliciano Filho Calixto,
  • Anderson Luis Szejka,
  • Eduardo Rocha Loures

摘要

This paper presents a human‑centric Product Lifecycle Management (PLM) framework aligned with Industry 5.0 that brings together Digital Twins (DTs), Smart Manufacturing, and Generative AI (GenAI) under a PLM backbone. The framework is designed around three pillars—human‑centricity, sustainability, and resilience—and incorporates explainable AI (XAI) and lifecycle assessment to support transparent, accountable decisions. A focused systematic literature review motivates the architecture and identifies gaps in explainability, interoperability, and end‑to‑end lifecycle coverage. The study demonstrates feasibility with a NASA milling case study, integrating sensor data, DT‑based emulation, and a predictive model within PLM workflows to achieve traceability and operator‑in‑the‑loop validation. Results show modest predictive performance, highlighting the need for feature engineering and model generalization; the paper outlines concrete improvement paths and discuss industrial adoption barriers (skills, integration, and governance). Contributions are: (i) a precise definition of PLM 5.0 and its dependencies, (ii) a modular framework integrating DT/GenAI/XAI with PLM and sustainability assessment, and (iii) a reproducible case configuration that maps each operational step to the framework. The paper concludes with open challenges in interoperability, security, and ethical governance and proposes evaluation and benchmarking directions.