Studying the reliability of complex systems using machine learning techniques involves facing a series of technical and practical challenges, ranging from the intrinsic nature of the system and data to the difficulties in modeling and effectively deploying models in real-world scenarios. This study compares the effectiveness of classical statistical techniques and machine learning methods for improving complex system analysis in reliability assessments. Our goal is to show that in many practical applications, traditional statistical algorithms frequently produce more accurate and interpretable results compared with black-box machine learning methods. The evaluation is conducted using both real-world data and simulated scenarios. We report the results obtained from statistical modeling algorithms, as well as from machine learning methods including neural networks, K-nearest neighbors, and random forests.

Our practical motivation is the analysis of potential correlations between spectral noise current and threshold voltage from common on-wafer MOSFETs. The usual strategy leads to the use of standard techniques based on Normal linear regression easily accessible in all statistical software (both free or commercial). However, these statistical methods are not appropriate because the assumptions they lie on are not met. More sophisticated methods are required. A new strategy based on the most novel nonparametric techniques which are data-driven and thus free from questionable parametric assumptions is proposed. A backfitting algorithm accounting for random effects and nonparametric regression is designed and implemented. The nature of the correlation between threshold voltage and noise is examined by conducting a statistical test, which is based on a novel technique that summarizes in a color map all the relevant information of the data. The way the results are presented in the plot makes it easy for a non-expert in data analysis to understand what is underlying. The good performance of the method is proven through simulations and it is applied to a data case in a field where these modern statistical techniques are novel and result very efficient.

Background Vitamin D status correlates with 25OHD levels which depends on nutritional intake and UVB exposure. These two factors are influenced by friends, as people tend to participate in the same activities or eat a similar diet to their peers. Objectives Investigate how social interactions in a general high school population above the Arctic Circle influence 25OHD levels in the population. Methods The Fit Futures 1 study was performed over 8 months and interview data on social contact among 1038 first-level students in 8 high schools in Northern Norway were collected. Serum levels of 25OHD were measured (n = 890). The participants filled in a questionnaire about nutritional consumption, solarium habits, ethnicity, and chronic diseases. The participant’s BMI was also measured. Results Once high schools’ social biases were accounted for, only UVB radiation levels explained the differences in 25OHD levels. For non-solarium users, logistic regression analysis showed a positive correlation between a person’s 25OHD levels and the 25OHD level of friends in the general population and within the same high school for most of the schools. We saw that women can influence other women into going to the solarium, while this influence was not present for men. Conclusions 25OHD levels can be influenced by social networks. This study can help to add weight to current public health recommendations due to the positive spillover effect in the network.