The construction industry experiences the greatest proportion of workplace injuries globally. Many efforts have gone to improve the effectiveness of safety training using advanced gaming technology and virtual reality (VR) technology. Most existing serious gaming programs are scenario-based which provide trainees, e.g., construction workers, an immersive virtual environment where they may process the construction tasks at the job site. However, the training scenarios are often determined by the developer or designer’s judgment. This is problematic since some important accident-associated details and elements are ignored during the scenario design. To address this problem, this work presents an approach that enables to automatically identify design elements from real world accident reports. The accident reports (n = 232) used in this study are gained from the National Institute of Occupational Safety and Health (NIOSH) for a period of 27 years (1989–2015). The approach integrates human factors engineering and machine learning techniques to gaming technology. This work also provides a case study that adopts the presented method in developing safety training scenarios using BIM and virtual reality headset for home builders. The case demonstrates improved accuracy and efficiency for the scenario design and storybook which are critical to the success of a VR-based training program.

The security in vehicular ad hoc networks (VANETs) has become a large consideration in safeguarding growing applications and intelligent transport systems. A group signature, a popular authentication approach for VANETs, can be implemented to protect vehicular communications against malicious users. However, the issue of securely distributing group keys to fast-moving vehicular nodes arises. The growing size of the certificate revocation list (CRL) has provided the corresponding complication in its management and distribution in VANETs. In this paper, an efficient key management protocol for group signature based authentication is proposed. A group is extended to a domain with various roadside units forming a hierarchical topology. Our proposed scheme provides a secure method to deliver group keys to vehicular nodes, ensuring the security requirements. Similarly, through utilizing the two Bloom filters in our hierarchical topology, an efficient and scalable vehicle revocation mechanism can be achieved that can minimize the CRL size. Our experiment results demonstrate a scalable, efficient, and secure key distribution scheme in vehicular networking. Moreover, an effective CRL management mechanism can be accomplished using the hierarchical topology.

Efficient performance prediction at the source code level is essential in reducing the turnaround time of software development. In this paper, we introduce a new prediction model, which combines several machine learning algorithms, such as KNN, clustering, similarity, sample and attribute weighting with multiple linear regression techniques, to predict the execution time of Digital Signal Processing (DSP) software at the source code level. Prediction at source code level tends to both under-predict the performance for certain testing samples and over-predict for some other samples. Therefore, we propose a new algorithm called MAX/MIN algorithm to select the best-predicted execution time. To validate the new model, we measure experimentally the execution time of a set of functions selected from PHY DSP Benchmark and run them on TIC64 DSP processor. It is observed that the average absolute relative prediction error is less than 10% between the computed performance from the new model and the actual measured execution time.

Three-dimension (3D) NAND flash memory is the preferred storage component of solid-state drive (SSD) for its high ratio of capacity and cost. Optimizing the reliability of modern SSD needs to test and collect a large amount of real-world error data from 3D NAND flash memory. However, the test costs have surged dozens of times as its capacity increases. It's imperative to reduce the costs of testing denser and high-capacity flash memory. To facilitate it, in this paper, we aim to enable reproducing error data efficiently for 3D NAND flash memory. We use a conditional generative adversarial network (cGAN) to learn the error distribution with multiple interferences and generate diverse error data comparable to the real-world. Evaluation results demonstrate it is feasible and efficient for error generation with cGAN.