- Research Article
1
- 10.1016/j.jmsy.2026.01.007
Machine vision defect segmentation and geometric measurement for real time quality monitoring in friction stir welding
- Apr 01, 2026
- Journal of Manufacturing Systems
- Naveen Loganathan + 2 more +2
Publications from 2021 to 2026
Showing 10 of 1,086 papers
Machine vision defect segmentation and geometric measurement for real time quality monitoring in friction stir welding
"Having surgery is necessary" - a qualitative analysis of the experiences of frail older adults treated with, and recovering from colorectal cancer surgery.
Frail older adults undergoing surgical treatment for colorectal cancer are at increased risk of postoperative mortality and complications compared with non-frail older adults. Even though these risks are established, limited research has explored patient perspectives on surgical treatment and recovery in frail older adults. This study aimed to explore the experiences of frail older adults during diagnosis, treatment, and recovery from surgically treated colorectal cancer. Participants were recruited through purposive sampling from the control group of the randomized controlled trial “Effect of comprehensive geriatric assessment for frail elderly patients operated for colorectal cancer – the Colorectal Cancer Frailty Study.” Patients in the control group receive standardised treatment according to best practice within an ERAS-concept. Sixteen semi-structured interviews were conducted 6–22 months post-surgery. All interviews were audio-recorded, transcribed verbatim, and analysed using qualitative content analysis with a focus on both manifest and latent content. Participants perceived surgery as essential for survival. They expressed considerable trust in health care professionals and showed limited interest in being actively involved in medical decision-making. Even among those who experienced postoperative challenges—such as reduced physical function and increased dependence on others—there was a high level of acceptance and a prevailing sense that the treatment had been worthwhile. Overall, the experience of the treatment was described in a positive sense, and participants stated that they would choose to undergo the procedure again if necessary. The findings of this study revealed that participants sought clear and concrete recommendations and expressed limited interest in influencing medical decision-making. There was a strong will to undergo surgery despite complicating factors such as advanced age, frailty, and comorbidities. A general acceptance of age-related functional decline may have contributed to the sense of being content with treatment outcomes, even in the presence of postoperative challenges. The findings reflect the perspective of surgically treated, mild to moderately frail older adults with colorectal cancer, who survived at least six months after surgery, and the results should be interpreted within this context. OSF registry: https://osf.io/c94xb/, registered May 24, 2025.
Read moreMachine Learning Based Prediction of Surface Quality in Friction Stir Welding
Abstract Ensuring consistent surface quality in friction stir welding (FSW) remains a challenge due to the complex interaction between process parameters such as axial force, rotational speed, and travel speed that govern heat generation and material flow. This study presents a data-driven and physically interpretable machine learning framework to predict weld surface quality directly from these parameters. Thirty robot-assisted FSW trials were performed on sand cast aluminium plates, varying axial force (4–7 kN), rotational speed (700–1700 rpm), and travel speed (180–300 mm min −1 ) at a constant tool tilt of 1°. The resulting welds were classified into four surface-quality categories: good weld, flash/burr, surface roughness, and surface void. Four models logistic regression, support vector machine, random forest, and artificial neural network were trained and validated using stratified five-fold cross validation. The support vector machine achieved the highest performance, with 81.8 % accuracy and a balanced precision recall score of approximately 0.77. Decision-boundary maps and calibrated probability predictions identified an optimal defect-free process window (≈ 4–6 kN, 1200–1500 rpm, ≥ 300 mm min −1 ), consistent with known thermo-mechanical behaviour in FSW. The proposed framework integrates predictive analytics with process physics, offering a deployable foundation for real-time, data-driven quality assurance in automated FSW.
Read moreDigital media use—a magnifying glass for mental health in adolescents: a qualitative systematic review
Microstructural Analysis of Hot Cracking in a 40 mm Thick Alloy 625 Weld Overlay
Co-designing a GenAI chatbot for foreign-born jobseekers
En öppning för konstnärlig undervisningspraktik: förskollärarstudenter och musiker undervisar tillsammans i förskola
I Sverige har alla barn enligt barnkonventionen rätt att fritt delta i det konstnärliga livet där musik ingår som en del. Med förskolor som ofta saknar musikkunnig personal blir det konstnärliga livet i förskolan begränsat. Därför är det av stor vikt att morgondagens förskollärare redan i sin utbildning får möjlighet att utvidga sin kompetens inom musik. Det saknas studier som synliggör vad som kan känneteckna nytänkande musikundervisning genomförd av förskollärarstudenter och musiker under verksamhetsförlagd utbildning. Syftet med studien är därför att ur ett didaktiskt perspektiv bidra med kunskap om vad som kan känneteckna musikundervisning genomförd av förskollärarstudenter och musiker i förskolan. Genom observation av samundervisande studenter och musiker visar huvudresultatet i föreliggande studie hur samverkan mellan studenter och musiker öppnar upp för ett ökat instrumentanvändande i undervisningen med barnen. Det handlar om lyssnande till och prövande av instrumentspel, verbaliserande i form av samtalande, klargörande och instruerande samt komponerande. Studenter och musiker öppnar tillsammans upp för deltagande i ett konstnärligt liv i förskolekontext. ENGLISH ABSTRACT An opening for Artistic Teaching Practice: Early Childhood Education and Care (ECEC) Teacher Students and Musicians Teaching Together in ECEC In Sweden, according to the Convention on the Rights of the Child, all children have the right to freely participate in artistic life where music is included as a part. With Early Childhood Education and Care (ECEC) that often lack music-savvy staff, the artistic life in ECEC is limited. Therefore, it is of great importance thaT tomorrow’s ECEC teachers already in their education have the opportunity to expand their competence in music. There is a lack of studies that make visible what can characterize innovative music teaching carried out by ECEC teacher students and musicians during activity-based training. The purpose of the study is therefore, from a didactic perspective, to contribute with knowledge about what can characterize music teaching carried out by ECEC teacher students and musicians in ECEC. Through observation of co-teaching students and musicians, the main result in the present study shows how collaboration between students and musicians opens up for an increased use of instruments in teaching the children. It involves listening to and trying to play instruments, verbalization in the form of conversation, clarifying and instructing, as well as composing. Students and musicians together open up for participation in an artistic life in an ECEC context.
Read moreOptimization of AA7075–SiC Composite Machining by WEDM Using Biosilica Additives
The study herein discusses the wire-cut electrical discharge machining (WEDM) with the addition of biosilica from maize cobs for the precision cutting of AA7075 aluminum and silicon carbide (SiC) metal matrix composites. The new, harmless addition of biosilica as a dielectric agent not only supports the sustainability of the process but also enhances surface quality and cutting speed. A Taguchi L9 design was used to conduct an experiment investigating the effects of peak current, gap voltage, and pulse-on time on material removal rate (MRR) and surface roughness (Ra). The optimal machining parameters were determined using Grey Relational Analysis and an artificial neural network (ANN). These model developments aimed to predict performance outcomes. The findings indicated that the dielectric fluid with biosilica increased MRR by 25% and simultaneously decreased Ra by 15% when compared with the typical dielectric. SEM and AFM analysis confirmed surface uniformity improvement and reduced microcrack formation. The artificial neural network model, trained using the backpropagation method on the experimental data, produced predictions for material removal rate and average roughness with an R² of 0.96, indicating that the model is highly reliable. In conclusion, the present research not only reveals a non-conventional machining process but also provides an eco-friendly approach to optimizing wire EDM of metal matrix composites with nano-reinforcement.
Read moreCo-workership rebooted for digital transformation: highlighting employees’ reflection and learning
This paper addresses how employees’ perceptions of organizational and technological aspects of digital transformation in manufacturing contexts contribute to understanding co-workership. To answer this, the research design is rooted in an engaged scholarship approach and work-integrated learning in close collaboration with a case company to understand organisational and digital transformations over time. The study consisted of five focus groups with a total of 25 participants and applied the framework of the co-workership wheel to contribute to the understanding of inclusion of co-workers in digital transformation processes. The results show the need for learning and reflection as an additional conceptual pair and thus propose an extended co-workership wheel for the manufacturing context.
Read more‘The danger of a single story’: using Bakhtninian dialogism to explore Identity formation in the writing of Igiaba Scego
ABSTRACT This article employs Bakhtinian dialogism to create reflexive narratives that illustrate identity formation in the context of migration. Starting from the assumption that engaging dialogically with stories is a powerful tool to resist one-dimensional narratives, in this study, we engage dialogically with the writing of Italian author Igiaba Scego. Considering the inherent dialogism of texts, we aim to encourage the active reading of narrative through a polyphony that enmeshes our voices with those of the author of the texts and the characters that inhabit them. We argue that intercultural understanding emerges from this dialogic engagement with the words of others.
Read more