Options
Experimental framework for investigating manual drilling in aircraft assembly
Citation Link: https://doi.org/10.15480/882.16825
Publikationstyp
Conference Paper
Date Issued
2026-02-26
Sprache
English
TORE-DOI
Citation
25th Machining Innovations Conference for Aerospace Industry, MIC 2026
Contribution to Conference
Publisher Link
Peer Reviewed
true
In aircraft structural assembly, a substantial share of rivet holes is drilled manually using handheld machines. In contrast to semi-automatic
drilling machines, operations with handheld tools are subject to process-inherent parameter variations such as fluctuations in feed force and
differences in tool alignment, both of which strongly influence hole quality and tool wear. Moreover, handheld drilling machines typically lack
integrated sensing capabilities, further complicating process monitoring and quality assurance. To address these challenges, a dedicated
experimental test rig for manual drilling is developed and tested. The setup enables systematic investigation of manual drilling operations under
controlled conditions, incorporating external sensors to capture key process signals such as feed force, spindle speed, torque and feed travel.
Based on these measurements, derived parameters such as feed rate and drilling time can be analyzed to study their relationship to hole exit
quality, particularly delamination in carbon fiber reinforced polymer (CFRP) material. Initial investigations using Random Forest classification
and permutation importance to identify significant process characteristics demonstrate that higher feed rates right before tool exit correlate with
increased delamination, consistent with the findings from automated drilling research. The results further reveal a strong influence of operator-
induced variability and tool wear on the measured process parameters, with feed force variations exceeding 40 N within individual drilling
operations. This emphasizes the need for systematic understanding of human–machine interactions in manual drilling and motivates the
development of monitoring and assistance systems capable of compensating operator-induced variations. The presented experimental framework
provides a foundation for quantitative analysis of manual drilling behavior and serves as a data-driven basis for future work on process monitoring,
anomaly detection, and operator feedback control.
drilling machines, operations with handheld tools are subject to process-inherent parameter variations such as fluctuations in feed force and
differences in tool alignment, both of which strongly influence hole quality and tool wear. Moreover, handheld drilling machines typically lack
integrated sensing capabilities, further complicating process monitoring and quality assurance. To address these challenges, a dedicated
experimental test rig for manual drilling is developed and tested. The setup enables systematic investigation of manual drilling operations under
controlled conditions, incorporating external sensors to capture key process signals such as feed force, spindle speed, torque and feed travel.
Based on these measurements, derived parameters such as feed rate and drilling time can be analyzed to study their relationship to hole exit
quality, particularly delamination in carbon fiber reinforced polymer (CFRP) material. Initial investigations using Random Forest classification
and permutation importance to identify significant process characteristics demonstrate that higher feed rates right before tool exit correlate with
increased delamination, consistent with the findings from automated drilling research. The results further reveal a strong influence of operator-
induced variability and tool wear on the measured process parameters, with feed force variations exceeding 40 N within individual drilling
operations. This emphasizes the need for systematic understanding of human–machine interactions in manual drilling and motivates the
development of monitoring and assistance systems capable of compensating operator-induced variations. The presented experimental framework
provides a foundation for quantitative analysis of manual drilling behavior and serves as a data-driven basis for future work on process monitoring,
anomaly detection, and operator feedback control.
Subjects
Aircraft Assembly
Manual Drilling
Experimental Test Rig
Process Monitoring
Rivet Hole Quality
Integreted Sensing
CFRP Drilling
DDC Class
629.1: Aviation
670: Manufacturing
006.31: Machine Learning
620.11: Engineering Materials
Publication version
publishedVersion
Loading...
Name
ssrn-6287318.pdf
Type
Main Article
Size
983.34 KB
Format
Adobe PDF