A psychophysical approach for predicting isometric and isotonic hand muscle strength in the aviation industry.

摘要

In the aviation industry, most operations are accomplished using hands. Hand grip strength is a significant factor that can influence human performance in terms of the amount of force that an individual can apply and their time endurance limit. The main objective of this study is to determine the maximum voluntary contraction and fatigue endurance limits for both types of hand muscles (isometric and isotonic) for workers in the Jordanian aviation industry. Using a psychophysical approach based on human subjective perception of fatigue, a total number of 132 (aged between 20 and 60 years old) subjects from the aviation industry was studied. The experiment investigates the effect of nine different factors on three responses: maximum voluntary contraction (MVC), isometric endurance limit, and isotonic endurance limit, and the relationships between them. In addition, general and specific predictive linear models were developed where not all factors are included simultaneously. The predictor variables are age, hand dominancy, human body posture, grip circumference (GC), forearm circumference (FAC), body mass index (BMI), height, profession (trade) and smoking condition. The isometric endurance limit tested for different percentages of MVC at 20%, 40%, 60% and 80%, which reflects real-life situations. The isometric endurance limit was tested for those between 20% and 60% of the MVC force. In this experiment, digital hand grip dynamometer was used to increase the accuracy of the experiment. The research experiment outputs were analyzed with statistical analysis (e.g., descriptive statistical analysis, interval plots, model adequacy checks, residual plots, MANOVA and ANOVA). Mathematical modeling (linear and nonlinear) and machine learning techniques (Artificial Neural Networks (ANNs), Artificial Neuro Fuzzy Inference System (ANFIS)) were applied. Results show that age and physical factors have significant effects. All predictive models compared on the R-squared values and Root Mean Square Error (RMSE). The machine learning models obtained the lowest RMSE (7.09 e -8 - 9.9 e-1) and provided the better fit for the data than the mathematical models, especially ANFIS methodology; however, linear models were convenient to build for this research. A pilot study was conducted to refine the best framework for the actual experiment. Research findings can be applied to the employment process of aviation industry workers as well as to workers of police, firefighting, and air force to enhance general health of athletic personnel and for better design tasks and related tools in a more economical way.