Lämkull Dan, Hanson Lars, Ortengren Roland
Volvo Car Corporation, Manufacturing Engineering Dept. 81121, Göteborg, Sweden.
Appl Ergon. 2007 Nov;38(6):713-22. doi: 10.1016/j.apergo.2006.12.007. Epub 2007 Mar 21.
The objective of this study was to investigate whether the appearance of virtual human models influences observers when judging a working posture. A task in which a manikin is manually assembling a car battery was used in the experiment. In total, 16 different pictures were presented to the subjects. All pictures had the same background, but included a unique posture and manikin appearance combination. 24 subjects consisting of manufacturing managers, simulation engineers and ergonomists were asked to rate and rank the pictures. The results showed that the virtual human model appearance influenced subjects when they rated pictures one by one: a more realistic manikin was rated higher than the identical posture visualized with a less natural appearance. This appearance effect was not seen when subjects ranked the pictures while looking at all of them at the same time. The study demonstrates that the human modelling tool used when showing and visually evaluating results makes a difference. To minimize subjective effects, a combination of visualizations and objective ergonomic assessment methods is recommended.
本研究的目的是调查虚拟人体模型的外观在评判工作姿势时是否会对观察者产生影响。实验采用了一个人体模型手动组装汽车电池的任务。总共向受试者展示了16张不同的图片。所有图片背景相同，但包含独特的姿势和人体模型外观组合。24名由制造经理、模拟工程师和人类工效学家组成的受试者被要求对图片进行评分和排序。结果表明，当受试者逐一给图片评分时，虚拟人体模型的外观会影响他们：一个更逼真的人体模型比以不太自然外观呈现的相同姿势获得的评分更高。当受试者同时查看所有图片并进行排序时，这种外观效应并未出现。该研究表明，在展示和视觉评估结果时所使用的人体建模工具会产生影响。为了尽量减少主观影响，建议将可视化方法与客观的人体工程学评估方法相结合。
