A numerical and statistical approach for optimization of tab design for non-crimp fabric composites

Kwame Anane-Fenin, Esther T. Akinlabi, Nicolas Perry

Research output: Contribution to journalConference articlepeer-review

2 Citations (Scopus)


In standard practice, testing of composites in tension requires the use of stress-inducing serrated grips. The low transverse compressive strength of unidirectional non-crimp fabric composites limits the application of high clamping forces. Tabs are therefore essential as they ensure a reduction in grip pressure transmission, surface damage and induced stress damage. Tabs, however, tend to introduce induced stress concentrations at the tab termination region. The objective of this study was to minimise stress concentration by varying tab design configurations to determine the optimal design most suitable for tensile testing of non-crimp fabric composites using finite element and statistical tools. Finite element models generated from experimental data were used for accessing the stress concentrations. A two (2)-level full factorial design was adopted and utilised for statistical analysis. Results revealed that tab stiffness, tab taper angle, adhesive thickness and manufacturing process (bonded or molded) were statistically significant for minimising stress concentration. Molded tabs were found to be acceptable if the stiffness of tab was significantly lower than test specimen.

Original languageEnglish
Pages (from-to)820-825
Number of pages6
JournalProcedia Manufacturing
Publication statusPublished - 2019
Event2nd International Conference on Sustainable Materials Processing and Manufacturing, SMPM 2019 - Sun City, South Africa
Duration: 8 Mar 201910 Mar 2019


  • Non-crimp fabric composite
  • Strees concentration
  • Tabs

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Artificial Intelligence


Dive into the research topics of 'A numerical and statistical approach for optimization of tab design for non-crimp fabric composites'. Together they form a unique fingerprint.

Cite this