| Augustin SEMENESCU1, Ana-Maria NICOLAU2
Abstract. The adoption of human-robot collaboration (HRC) in Industry 4.0/5.0 environments introduces a class of systemic risks that classical work-system analysis methodologies were not designed to capture. This paper presents a critical study of the principal methodological families applicable to HRC safety evaluation – standards-based and checklist approaches, semi-quantitative/MCDM methods, and contemporary systemic models (STAMP/STPA, FRAM) – evaluated through the same six-criterion comparative framework previously applied to general occupational safety methodologies. Two illustrative case studies, from a fixed assembly setting and a mobile logistics setting, show substantial differences between the risks identified by each methodological family. The central finding, extending the “methodological accessibility paradox” identified for occupational safety methodologies in general to the specific context of HRC, is that current standards (e.g., ISO/TS 15066), while accessible, do not capture emergent risks generated by AI-driven dynamic adaptation of robot behaviour. A stratified methodological framework, adapted to the specifics of Industry 5.0, is proposed. Keywords: human-robot collaboration, Industry 5.0, STAMP/STPA, FRAM, collaborative robot safety, ISO/TS 15066, systemic risk, methodological accessibility paradox. DOI 10.56082/annalsarscieng.2026.2.70 Read full article 1Professor, PhD, Eng. Mat. Ec., National University of Science and Technology Politehnica Bucharest, Bucharest, Romania, Member of Academy of Romanian Scientists; (e-mail: augustin.semenescu@upb.ro). 2Assoc. Professor, PhD, Eng., National University of Science and Technology Politehnica Bucharest, Bucharest, Romania, (e-mail: ana.avramescu@upb.ro). |
PUBLISHED in Annals Academy of Romanian Scientists Series on Engineering Sciences
ISSN PRINT 2066 – 6950 ISSN ONLINE 2066 – 8570 |
