INVESTIGATING THE IMPACT OF MATERIAL FATIGUE ON STRUCTURAL INTEGRITY IN AI-ASSISTED MECHANICAL SYSTEMS
Keywords:
material fatigue, how systems hold together, AI being used, mechanical systems, dependability, regression analysis, Cronbach’s Alpha, testing for normalityAbstract
Mechanical structures that include AI-based monitoring may be affected by material fatigue which makes them less reliable. To make maintenance better and reduce failures, it is important to know how different operating and material factors connect to affect the reliability of the system.
Objective: To examine the consequences of material fatigue on AI-assisted mechanical structures and investigate how well the fatigue datasets fit and how the accuracy of AI monitoring systems themselves impacts the results.
Methods: The study used a questionnaire to collect data from workers in the aerospace, automotive, and robotics fields. Our study used responses from 273 total participants. In the instrument, a total of 24 Likert items formed the independent variables (number of load cycles, what the structure is made of, surrounding conditions, quality when made), a mediating variable (material fatigue), a moderating variable (accuracy of AI checks) and dependent variables (how well the structure holds up and the rate of failure). All analysis was conducted with the help of SPSS 24.0. Normality was checked with the Shapiro-Wilk test, reliability with Cronbach’s Alpha, and relationships with Pearson correlation and multiple linear regression.
Results: Researchers found that most of the samples were distributed differently than a normal distribution. Cronbach’s Alpha was as little as 0.036, meaning the test did not show much consistency inside. We found moderate connections between fatigue-related and structural integrity variables during correlation analysis. Results from regression analysis show that the accuracy of AI monitoring and fatigue detection were little predictors of structure safety, while most other predictors had little to no effect.
Conclusion: The report gives an early look at the relationship between material fatigue and how well AI-monitored systems perform. Although the instrument wasn’t very reliable statistically, the results highlight that it is important to track both accuracy and fatigue in AI. Improving the questionnaire and applying advanced methods in modeling are considered important for the progress of future research.
