Abstract. An accurate prediction of pavement performance results in an effective plan for managing the highway network in cost-effective management and future maintenance strategies. However, in some dry climate countries, the available resources are not enough to conduct a periodic evaluation for their highways and apply a suitable maintenance action at a proper time. Therefore, the objective of this study is to utilize the available data in the Long-Term Pavement Performance (LTPP) program for pavement sections that are in dry-non-freeze zones. The selection of LTPP pavement sections in dry-non-freeze zones attempts to represent the pavement performance in dry climate countries. The International Roughness Index (IRI) was used as a performance indicator because it reflects the level of riding quality, the comfort of road users, and the level of pavement condition. The random forests (RF) and multiple linear regression (MLR) models predict the IRI for flexible pavements from pavement age, traffic and climate data, pavement distress data, and structural properties. The results show that the coefficient of determination (R 2 ) in the MLR model is 0.70, whereas the RF model yields a relatively higher R 2 value of 0.85. Also, the results of the RF model show that the initial pavement roughness was the most significant variable that impacted the pavement roughness, as well as, pavement thickness, pavement age, and truck volume have a high impact on the IRI value.

Abstract. Keeping a microprocessor surface temperature evenly distributed and below a threshold value is difficult to achieve using conventional electronic cooling approaches. Hotspots on the microprocessor surface are the cause of this challenge. The proper approach is to cool these hotspots separately while maintaining the entire microprocessor surface at acceptable temperatures. A thermoelectric cooler module is suggested to be used to cool the hotspot. Thermoelectric generator modules are used to generate/harvest electrical power from the wasted heat of the microprocessor background, and then it is used to run the thermoelectric cooler. In this research, four parameters were used to develop a design tool to achieve the cooling system goals. These parameters are the height of the thermoelectric modules’ legs, the spaces between the thermoelectric legs, the generated heat by the hotspot, and the generated heat by the microprocessor background. Simulated data were generated using a 3-D validated model. A neural network was then used to build the basis of a design tool, which can be improved to cover a wider range of parameters and their values. The results showed that the developed design tool in this study can be used to determine the proper parameters values for a given microprocessor temperature target.

Abstract. As a large and complex system, an electric power system may be subject to a number of disturbances which can cause full or partial failure in its operation. This premise warrants the inclusion of reliability evaluation in the planning process of a power system. Among the several available reliability evaluation techniques, fault-tree analysis is definitely considered a highly suitable one, particularly for dealing with the inherent complexity of power systems. This paper applies of fault-tree analysis for a partial reliability evaluation of a specific power system known as the Roy Billinton Test System (RBTS). The proposed study is to calculate the reliability of power delivery to the largest load point of the RBTS. The steps carried out in the analysis include fault-tree construction, qualitative analysis by establishing the minimal cutsets, and quantitative analysis by calculating reliability utilizing the data of failure probability of each component of the system. The results presented in this study include failure probabilities for the minimal cutsets, failure probability of power delivery to the system’s largest load and the corresponding reliability, as well as the importance measures of various components. The analysis produces a set of twenty minimal cutsets for the power delivery to the largest load point in the RBTS, with a corresponding numerical value of 0.999597 of system partial reliability for the assumed component reliabilities. The importance analyses utilizing Fussell-Vesely importance, risk achievement worth, and risk reduction worth finds that BUS3 is the most important system component for the power delivery to the largest load, while transmission line L5 is the least important system component.

Abstract. Software vulnerability is a weakness and bug that can affect security protocols of the system. The issue at hand has produced an increased software vulnerability within each lifecycle release. A web browser is used by users as the primary method to communicate with each other online via devices. People can find great content on the web. Challenges arise within security protocols when attackers access users via web browsers. Therefore, web browsers have become a target for attackers who use web browser vulnerabilities. This research focuses on Google Chrome vulnerabilities, and proposes a new approach which can provide probabilistic predictions that generate a score level of the Chrome browser vulnerability. The predication model uses artificial intelligence from the "Bayesian Network ".