Scientific Programming

This paper investigates single-period inventory management problems with uncertain market demand, where the exact possibility distribution of demand is unavailable. In this condition, it is important to order a reliable quantity which can immunize against distribution uncertainty. To model this type of single-period inventory management problem, this paper characterizes the uncertain demand by generalized interval-valued possibility distributions. We present a novel concept about an uncertain distribution set to describe distribution perturbation characterization. First, we introduce a lambda selection of the interval-valued fuzzy variable, and the uncertain distribution set is a collection of all generalized possibility distributions of lambda selection variables. According to the uncertain distribution set, a new distributionally robust fuzzy optimization method is developed for single-period inventory management problems. Under mild assumptions, the robust counterpart of the proposed fuzzy single-period inventory management model is formulated, which is an optimization program with certain linear objectives and infinitely many integral constraints. We discuss the computational issue of integral constraints and reformulate equivalently the robust counterpart as three deterministic inventory submodels under generalized interval-valued trapezoidal possibility distributions. According to the characteristics of three submodels, a domain decomposition method is designed to find the robust optimal solution that can immunize against uncertainty in our single-period inventory management problem. Finally, some computational results demonstrate the efficiency of the proposed distributionally robust fuzzy optimization method.

Many important computational applications in science, engineering, industry, and technology are represented by PSE (parameter sweep experiment) applications. These applications involve a large number of resource-intensive and independent computational tasks. Because of this, cloud autoscaling approaches have been proposed to execute PSE applications on public cloud environments that offer instances of different VM (virtual machine) types, under a pay-per-use scheme, to execute diverse applications. One of the most recent approaches is the autoscaler MOEA (multiobjective evolutive algorithm), which is based on the multiobjective evolutionary algorithm NSGA-II (nondominated sorting genetic algorithm II). MOEA considers on-demand and spot VM instances and three optimization objectives relevant for users: minimizing the computing time, monetary cost, and spot instance interruptions of the application’s execution. However, MOEA’s performance regarding these optimization objectives depends significantly on the optimization algorithm used. It has been shown recently that MOEA’s performance improves considerably when NSGA-II is replaced by a more recent algorithm named NSGA-III. In this paper, we analyze the incorporation of other multiobjective optimization algorithms into MOEA to enhance the performance of this autoscaler. First, we consider three multiobjective optimization algorithms named E-NSGA-III (extreme NSGA-III), SMS-EMOA (S-metric selection evolutionary multiobjective optimization algorithm), and SMPSO (speed-constrained multiobjective particle swarm optimization), which have behavioral differences with NSGA-III. Then, we evaluate the performance of MOEA with each of these algorithms, considering the three optimization objectives, on four real-world PSE applications from the meteorology and molecular dynamics areas, considering different application sizes. To do that, we use the well-known CloudSim simulator and consider different VM types available in Amazon EC2. Finally, we analyze the obtained performance results, which show that MOEA with E-NSGA-III arises as the best alternative, reaching better and significant savings in terms of computing time (10%–17%), monetary cost (10%–40%), and spot instance interruptions (33%–100%).

Understanding the spatiotemporal pattern of precipitation concentration is important for the assessment of flood and drought risk and utilization of water resources. In this study, the daily precipitation concentration index in the Hai River basin in China was calculated based on the Gini coefficient obtained from the observed data of 51 meteorological stations from 1951 to 2018 and spatiotemporal pattern variations were investigated. The trends and abrupt changes of DPCI were tested by the Mann–Kendall and the Pettitt methods, respectively. The relationships among DPCI, percentage of precipitation contributed by the rainiest days, and disaster losses were discussed by the linear correlation analysis. The results showed that the DPCI value ranges between 0.6471 and 0.7938, decreases westward and northward, which is negatively related to latitude and elevation, and is positively related to longitude. Negative trends of the DPCI were found at most stations, and the PCI trends of more than 80% stations were statistically significant. Abrupt changes of the DPCI have a postponement trend from west to east with time. Daily heterogeneity of the rainfall in a year is highly correlated with the heavy rainfall amount of the 25% rainiest days. For a year, higher DPCI coupled with more precipitation is easy to cause a flood disaster; conversely, higher DPCI along with less precipitation is easy to cause a drought disaster. In the future, the risk of flood disasters would be reduced, but the drought disasters would be increased in the Hai River basin.

To support effectively function-driven 3D model retrieval in the phase of mechanical product conceptual design and improve the efficiency of functional semantics annotation for 3D models, an approach for functional semantics automatic annotation for mechanical 3D product model based on latent functional semantics is presented. First, the design knowledge and function knowledge of mechanical product model are analyzed, and the ontology-based functional semantics for assembly product is constructed. Then, some concept about functional region is defined, and the 3D product model is decomposed into functional regions with different levels of granularity. The similarity of the functional region is evaluated considering multisource attribute information and geometric shape. Subsequently, the similarity based on latent functional semantics annotation model for functional regions is established, which is employed for annotating automatic latent functional semantics in the 3D product model structure. Finally, mechanical 3D models in the model library are used to verify the effectiveness and feasibility of the proposed approach.

In view of the shortcomings of the prediction method of future development measures and indicators of fault block reservoir in the current oilfield practical application, a prediction method of fault block reservoir measures and indicators based on the random forest method and LightGBM is proposed, which can help the oilfield make more effective decisions in the middle and later development. Firstly, using the advantages of random forest (RF) in dealing with high-dimensional data sets, the main controlling factors are selected by feature analysis. Then, the measure prediction model is established by using the 1DCNN-LightGBM algorithm. Firstly, 1DCNN processes the reservoir dynamic data and then trains the LightGBM model with the extracted time series characteristics and static data characteristics as input to predict the measure indexes of fault block reservoir. The evaluation results show that the prediction models proposed in this paper have good performance and can obtain more accurate prediction results and more stable prediction performance. It provides a basis for the future planning and optimization of the oilfield.

Software testing identifies defects in software products with varying multiplying effects based on their severity levels and sequel to instant rectifications, hence the rate of a research study in the software engineering domain. In this paper, a systematic literature review (SLR) on machine learning-based software defect severity prediction was conducted in the last decade. The SLR was aimed at detecting germane areas central to efficient predictive analytics, which are seldom captured in existing software defect severity prediction reviews. The germane areas include the analysis of techniques or approaches which have a significant influence on the threats to the validity of proposed models, and the bias-variance tradeoff considerations techniques in data science-based approaches. A population, intervention, and outcome model is adopted for better search terms during the literature selection process, and subsequent quality assurance scrutiny yielded fifty-two primary studies. A subsequent thoroughbred systematic review was conducted on the final selected studies to answer eleven main research questions, which uncovers approaches that speak to the aforementioned germane areas of interest. The results indicate that while the machine learning approach is ubiquitous for predicting software defect severity, germane techniques central to better predictive analytics are infrequent in literature. This study is concluded by summarizing prominent study trends in a mind map to stimulate future research in the software engineering industry.