Drowsy driving poses a significant risk to road safety, necessitating the development of reliable drowsiness detection systems. In particular, the advancement of Artificial Intelligence (AI) based neuroadaptive systems is imperative to effectively mitigate this risk. Towards reaching this goal, the present research focuses on investigating the efficacy of physiological indicators, including heart rate variability (HRV), percentage of eyelid closure over the pupil over time (PERCLOS), blink rate, blink percentage, and electrodermal activity (EDA) signals, in predicting driver drowsiness. The study was conducted with a cohort of 30 participants in controlled simulated driving scenarios, with half driving in a non-monotonous environment and the other half in a monotonous environment. Three deep learning algorithms were employed: sequential neural network (SNN) for HRV, 1D-convolutional neural network (1D-CNN) for EDA, and convolutional recurrent neural network (CRNN) for eye tracking. The HRV-Based Model and EDA-Based Model exhibited strong performance in drowsiness classification, with the HRV model achieving precision, recall, and F1-score of 98.28%, 98%, and 98%, respectively, and the EDA model achieving 96.32%, 96%, and 96% for the same metrics. The confusion matrix further illustrates the model’s performance and highlights high accuracy in both HRV and EDA models, affirming their efficiency in detecting driver drowsiness. However, the Eye-Based Model faced difficulties in identifying drowsiness instances, potentially attributable to dataset imbalances and underrepresentation of specific fatigue states. Despite the challenges, this work significantly contributes to ongoing efforts to improve road safety by laying the foundation for effective real-time neuro-adaptive systems for drowsiness detection and mitigation.

The objective of this research is to apply Quality 4.0 (Q4.0) concept in Agriculture 4.0 (A4.0) to digitize the traditional quality management (QM) system and demonstrate the effectiveness of zero-defect manufacturing (ZDM) in the agricultural part manufacturing industry. An autonomous quality management system was developed based on the ZDM system using the Industrial Internet of Things (IIoT). Both traditional and autonomous quality management systems were evaluated using six-sigma quality indicators and machining and inspection cost analysis. The ZDM resulted in a significant improvement in the quality of CARD148 manufacturing, increasing the manufacturing process from a low level of sigma to a high level of sigma (0.75 to 5.10 sigma). The component rejection rate was reduced by a high percentage, leading to significant economic benefits and a significant reduction in machining cost. The process yield was also increased to a high percentage. The developed ZDM was found to be consistent in improving the quality of the turning process, with notable increases in tool life and reduction in inspection cost. The total component cost was reduced significantly, while the PPM value increased notably. While this study focuses on agriculture-related manufacturing organizations, the developed ZDM has potential for other machining industries to improve sigma levels, particularly in industries such as automotive and medical.

The aim of this study is to investigate the compressive stress on lumbar spine due to whole body vibration (WBV) exposure in real field rotary soil tillage operation. The spinal stress was evaluated in terms of daily equivalent static compression dose (Sed) with respect to various ride conditions (i.e. forward speed, pulling force and tilling depth). Taguchi's L9 orthogonal array was used to form a systematic set of experiments by varying each ride condition over three levels. Signal-to-noise (S/N) ratios were computed to analyse the holistic effect of ride conditions among all the experimental conditions. In addition, the power spectral density (PSD) of the acceleration at the seat pan was measured along the dominant axis for each experiment. A second order regression model was developed to predict the real experimental Sed response. Further, the ride conditions were optimised by using desirability technique with the aim of minimizing Sed response. The magnitude of daily equivalent static compression dose was found between 0.38 and 0.76 MPa which shows a moderate probability of an adverse health effects as per ISO2631-5: 2004. Statistically, forward speed and pulling force are found to significantly affect the Sed response with a contribution of 64.43% and 24.73%, respectively. The PSD indicated that the vibration energy of the acceleration measured on the seat pan depend on the ride conditions. The predicted response of regression model showed 4.56% error in obtaining the actual experimental values. The optimised forward speed, pulling force and tilling depth levels were found to be 0.6 m/s, 2 kN, and 0.10 m, respectively.