Machine learning has taken a front seat in railway big data analysis. This is partly due to perpetual data collection and the need for automated systems to expedite maintenance decisions. A case for track defect prognosis in rail track engineering is presented in this paper. Fatigue defects are very common and are influential on rail maintenance. Understanding such defects is essential for optimized maintenance scheduling. The literature is replete with machine learning models developed for defect prediction. Because no single machine learning model is guaranteed to surpass others with every kind of data, each model has its inherent deficiencies. Classifier ensembles such as bagging or boosting aggregate strengths from different models to enhance prediction. The outcome is very effective, although highly correlated. This work proposes a stacking method of combining average learners into powerful learning machines while considering memory, time, computational, structural complexities, and bias-variance trade-offs. Because of the large scale of rail infrastructure considered in this work (35,406 km), this study shows that classical Weibull analysis underestimates annual fatigue defects by at least 25% throughout rail life. The proposed stacking ensemble compensates for this shortfall by aggregating the probability predictions of diverse learners. These predictions were combined from a binary classification ensemble of 0.783 receiver operating characteristic area under curve (ROC-AUC) score with significant room for improvement in computation time and curve fitting.

This work investigates the use of Non-destructive tests as a tool for monitoring the structural performance of concrete structures. The investigation encompassed four phases; the first of which involved the use of destructive and non-destructive mechanisms to assess concrete strength on cube specimens. The second phase research focused on site assessment for a twin engineering theatre located at the Faculty of Engineering, University of Lagos using rebound hammer and ultrasonic pulse velocity tester. The third phase was the use of linear regression analysis model with MATLAB to establish a relationship between calibrated strength as well as ultrasonic pulse velocities with their corresponding compressive strength values on cubes and values obtained from existing structures. Results show that the root-mean squared-R2 values for rebound hammer ranged between 0.275 and 0.742 while ultrasonic pulse velocity R2 values were in the range of 0.649 and 0.952 for air curing and water curing systems respectively. It initially appeared that the Ultrasonic pulse velocity was more suitable for predicting concrete strength than rebound hammer but further investigations showed that the latter was adequate for early age concrete while the former was more suited for aging concrete. Hence, a combined use is recommended in this work.

Track geometry parameters from rail track inspection are regulated within unique safety limits for different track classes. This study focuses on developing an index that combines safety and track quality because of the inefficiency of corrective maintenance activities between routine maintenance cycles when federal geometry limits are violated. This combination is achievable by summarizing multivariate track geometry parameters as an improvement to previous linear approaches to address the problem of inefficient track geometry maintenance programs. The use of nonlinear dimension reduction (T-stochastic neighbor embedding (T-SNE)) for hybrid track quality index development and the influence of time-based parameters on track quality is evaluated in this study. The results show that the probability of geometry defects is correlated with principal components, but T-SNE had the best prediction on train-test splits despite its poor performance on a blind validation set. The absence of an observable correlation between the track geometry and acceleration data requires further investigation.