Deterministic Input Research Articles

Integration of stochastic and robust optimization techniques into DEA model for more accurate and reliable efficiency estimation

Efficiency assessment is vital to assessing decision-making units (DMUs) in numerous sectors. DEA is a prominent non-parametric efficiency assessment tool. Traditional DEA models assume deterministic inputs and outputs, ignoring real-world uncertainties and variability. To improve efficiency estimation, stochastic and robust optimization approaches can be integrated into DEA models. To improve efficiency estimation, we present a stochastic and robust optimization framework incorporating DEA. Probabilistic inputs and outputs allow stochastic optimization to account for uncertainty. The model can capture data variability and create stochastic DMU efficiency scores by adding probability distributions. For data uncertainties and outliers, the DEA model uses robust optimization. Robust optimization considers worst-case scenarios and minimizes extreme observations on efficiency estimation. This makes efficiency scores more resilient to data outliers and noise. DEA models benefit from stochastic and resilient optimization. First, considering data uncertainties and fluctuations improves DMU efficiency representation. Second, eliminating outliers and extreme observations improves efficiency estimation. Third, efficiency scores help decision-makers make better, more informed choices. A case study in a specific industry shows the framework's efficacy. We compare classic and integrated stochastic-robust DEA model outcomes. The integrated model provides more accurate and dependable efficiency estimates, helping decision-makers understand DMU performance. DEA models with stochastic and resilient optimization increase efficiency estimation. By considering uncertainties and outliers, this paradigm helps decision-makers evaluate DMUs in many sectors more accurately and reliably.

Scalable Bayesian Approach for the Dina Q-Matrix Estimation Combining Stochastic Optimization and Variational Inference.

Diagnostic classification models offer statistical tools to inspect the fined-grained attribute of respondents' strengths and weaknesses. However, the diagnosis accuracy deteriorates when misspecification occurs in the predefined item-attribute relationship, which is encoded into a Q-matrix. To prevent such misspecification, methodologists have recently developed several Bayesian Q-matrix estimation methods for greater estimation flexibility. However, these methods become infeasible in the case of large-scale assessments with a large number of attributes and items. In this study, we focused on the deterministic inputs, noisy "and" gate (DINA) model and proposed a new framework for the Q-matrix estimation to find the Q-matrix with the maximum marginal likelihood. Based on this framework, we developed a scalable estimation algorithm for the DINA Q-matrix by constructing an iteration algorithm that utilizes stochastic optimization and variational inference. The simulation and empirical studies reveal that the proposed method achieves high-speed computation, good accuracy, and robustness to potential misspecifications, such as initial value choices and hyperparameter settings. Thus, the proposed method can be a useful tool for estimating a Q-matrix in large-scale settings.

Investigation of Structural Design Optimization of Loaded Cable-Supported Pipe Systems Using Response Surface Method

A Cable-supported pipe system (CSPS) is a composite system of structure involving truss modules and cable to support a pipeline over a large space. The structure is crucial in the transportation of fluid. This paper proposes structural design optimization method using response surface method (RSM) and investigates the results of Multi-Objective Genetic Algorithm and Screening design optimization of the structure. The RSM is implemented using deterministic inputs of the verified experimental results with the aid of genetic aggregation response surface in ANSYS Design Exploration platform. The results of performance response of the loaded CSPS structure considering the uncertainty of the design parameters, including the cross-sectional area of the cable truss, and pipe members were subsequently mined. This revealed that the cable sections and truss modules can effectively and efficiently improve the performance of the structure. The outcome also shows that performance the structure is robust. A set of "Pareto optimum solutions' ' were thus obtained based on the constraints of the design optimization, the yield stress of material properties and limit displacement of the structure. The results illustrate that the proposed optimization method provides a set of Pareto optimum solutions and guide for the robust design of the structure.

Measuring students' learning progressions in energy using cognitive diagnostic models.

This study applied cognitive diagnostic models to assess students’ learning progressions in energy. A Q-matrix (i.e., an item attribute alignment table) was proposed based on existing literature about learning progressions of energy in the physical science domain and the Trends in International Mathematics and Science Study (TIMSS) assessment framework. The Q-matrix was validated by expert review and real data analysis. Then, the deterministic inputs, noisy ‘and’ gate (DINA) model with hierarchical relations was applied to data from three jurisdictions that had stable, defined science curricula (i.e., Australia, Hong Kong, and Ontario). The results suggested that the hypothesized learning progression was consistent with the observed progression in understanding the energy concept. We also found similarities in students’ attribute mastery across the three jurisdictions. In addition, we examined the instructional sensitivity of the selected item. We discuss several curriculum-related issues and student misconceptions that may affect students’ learning progressions and mastery patterns in different regions of the world.

Smart City Construction and Management by Digital Twins and BIM Big Data in COVID-19 Scenario

With the rapid development of information technology and the spread of Corona Virus Disease 2019 (COVID-19), the government and urban managers are looking for ways to use technology to make the city smarter and safer. Intelligent transportation can play a very important role in the joint prevention. This work expects to explore thebuilding information modeling (BIM) big data (BD)processing method ofdigital twins (DTs)of Smart City, thus speeding up the construction of Smart City and improve the accuracy of data processing. During construction, DTs build the same digital copy of the smart city. On this basis, BIM designs the building's keel and structure, optimizing various resources and configurations of the building. Regarding the fast data growth in smart cities, a complex data fusion and efficient learning algorithm, namely Multi-Graphics Processing Unit (GPU), is proposed to process the multi-dimensional and complex BD based on the compositive rough set model. The Bayesian network solves the multi-label classification. Each label is regarded as a Bayesian network node. Then, the structural learning approach is adopted to learn the label Bayesian network's structure from data. On the P53-old and the P53-new datasets, the running time of Multi-GPU decreases as the number of GPUs increases, approaching the ideal linear speedup ratio. With the continuous increase of K value, the deterministic information input into the tag BN will be reduced, thus reducing the classification accuracy. When K = 3, MLBN can provide the best data analysis performance. On genbase dataset, the accuracy of MLBN is 0.982 ± 0.013. Through experiments, the BIM BD processing algorithm based onBayesian Network Structural Learning (BNSL)helps decision-makers use complex data in smart cities efficiently.

Dealing with Missing Responses in Cognitive Diagnostic Modeling

Missing data are a common problem in educational assessment settings. In the implementation of cognitive diagnostic models (CDMs), the presence and/or inappropriate treatment of missingness may yield biased parameter estimates and diagnostic information. Using simulated data, this study evaluates ten approaches for handling missing data in a commonly applied CDM (the deterministic inputs, noisy “and” gate (DINA) model): treating missing data as incorrect (IN), person mean (PM) imputation, item mean (IM) imputation, two-way (TW) imputation, response function (RF) imputation, logistic regression (LR), expectation-maximization (EM) imputation, full information maximum likelihood (FIML) estimation, predictive mean matching (PMM), and random imputation (RI). Specifically, the current study investigates how the estimation accuracy of item parameters and examinees’ attribute profiles from DINA are impacted by the presence of missing data and the selection of missing data methods across conditions. While no single method was found to be superior to other methods across all conditions, the results suggest the use of FIML, PMM, LR, and EM in recovering item parameters. The selected methods, except for PM, performed similarly across conditions regarding attribute classification accuracy. Recommendations for the treatment of missing responses for CDMs are provided. Limitations and future directions are discussed.

Modeling Not-Reached Items in Cognitive Diagnostic Assessments.

In cognitive diagnostic assessments with time limits, not-reached items (i.e., continuous nonresponses at the end of tests) frequently occur because examinees drop out of the test due to insufficient time. Oftentimes, the not-reached items are related to examinees’ specific cognitive attributes or knowledge structures. Thus, the underlying missing data mechanism of not-reached items is non-ignorable. In this study, a missing data model for not-reached items in cognitive diagnosis assessments was proposed. A sequential model with linear restrictions on item parameters for missing indicators was adopted; meanwhile, the deterministic inputs, noisy “and” gate model was used to model the responses. The higher-order structure was used to capture the correlation between higher-order ability parameters and dropping-out propensity parameters. A Bayesian Markov chain Monte Carlo method was used to estimate the model parameters. The simulation results showed that the proposed model improved diagnostic feedback results and produced accurate item parameters when the missing data mechanism was non-ignorable. The applicability of our model was demonstrated using a dataset from the Program for International Student Assessment 2018 computer-based mathematics cognitive test.

An harmonic equivalent excitation method for random vibration response

Many loads on engineering structures had obvious randomness, such as earthquakes, wind, sea waves, uneven pavement, etc. The calculation and analysis of random vibration were complicated and inefficient, which has perplexed the development of engineering technology for a long time. To solve the above problems, this paper proposed a harmonic equivalent excitation method for structural random vibration analysis. The excitation force of multi-frequency action in continuous time interval was converted into harmonic excitation force with the same effect as superposition excitation in each discrete time interval by a harmonic equivalent method. The frequency and amplitude of harmonic excitation force in discrete-time intervals were calculated, which were then taken as the deterministic load input of the structure. The response of the structure was calculated by the conventional structural dynamic analysis method. Finally, the harmonic equivalent excitation method was used to analyze random pavement excitation and its application in the control of time-delay vibration reduction. The results show that the harmonic equivalent excitation method simplifies the calculation form of random vibration analysis without reducing the accuracy, and provides an effective method for analyzing random vibration in the engineering field.

Economic appraisal of an unconventional condensate play based on public pilot data prior to field development: Jafurah Basin case study (Saudi Arabia)

This study uses a stochastic approach to provide a holistic economic evaluation of the Jafurah field development project, the largest unconventional natural gas field in Saudi Arabia, projected to start first production in 2024. The assessment considers deterministic and probabilistic inputs based on data published in the literature, analogy with Eagle Ford reservoir parameters, and engineering-based assumptions. The uncertainty of discounted net cash flow, internal rate of return, and other target metrics for the Jafurah project is quantified (P90, P50, and P10). The results show the most likely (P50) net present value of 47.21 billion US dollars (at 10% discount rate) and a corresponding internal rate of return of 46.8%. Sensitivity analysis highlights the relative importance of the most critical probabilistic inputs (initial production rate, natural gas price, capital and operational expenses). Based on the current economic ranges and uncertainty analysis, it can be concluded that the Jafurah Field can be profitable in the long term. However, frequent reappraisals are recommended to help direct future decisions on capital expenditure programs for the drilling and completion of new wells, especially when new field performance data becomes available after the earlier wells start first production in 2024. Although hundreds of field delineation and production wells have already been drilled and completed over the past decade, major investment and time were needed to construct new regional pipelines (for gathering natural gas and liquids from each well pad), new chemical plants (for processing of the natural gas liquids), and new underground natural gas storage facilities (to buffer seasonal changes in the production supply and demand of natural gas), which is why first production is planned for 2024.

A Multi-level Remedial Teaching Design Based on Cognitive Diagnostic Assessment: Taking the Electromagnetic Induction as an Example

Multi-level teaching has been proven to be more effective than a one-size-fits-all learning approach. This study aimed to develop and implement a multi-level remedial teaching scheme in various high school classes containing students of a wide range of learning levels and to determine its effect of their learning. The deterministic inputs noisy and gate model of cognitive diagnosis theory was used to classify students at multiple levels according to their knowledge and desired learning outcomes. A total of 680 senior high school students from central provinces in China participated in the initial cognitive diagnostic test, and 1,615 high school sophomores from seven high schools in China participated in a formal cognitive diagnosis test. Thirty-six high school students from Southwestern China participated in the think-aloud protocols, and 258 seniors from three high schools in southwest China participated in the remedial teaching experiment. Through an analysis of students’ think-aloud protocols, cognitive errors of students at all levels were determined, and multi-level remedial teaching programs were designed to address these common cognitive errors. The remedial teaching programs were then implemented in three schools and compared with a control group. The results indicated that the students in the experimental group showed a more significant improvement. In this study, the steps of designing multi-level remedial teaching include assessment, classification, and preparing a teaching scheme, which are feasible and can have remarkable teaching effects. This process can be used for reference by teachers of various subjects.

Stall-induced fatigue damage in nonlinear aeroelastic systems under stochastic inflow: Numerical and experimental analyses

This study focuses on characterizing the fatigue damage accumulated in nonlinear aeroelastic systems subjected to stochastic inflows through both numerical simulations and wind tunnel experiments. In the mathematical model, nonlinearities are assumed to exist either in the structure (via a cubic hardening nonlinearity in the pitch stiffness), or in the flow (via dynamic stall condition), or simultaneously in both the structural and aerodynamic counterparts. The aerodynamic loads in the attached flow and dynamic stall conditions are estimated using Wagner’s formulation and semi-empirical Leishman–Beddoes model, respectively. To augment the findings to in-field flow conditions, the oncoming wind flow is considered to be randomly time-varying in nature. The stochastic input flow fluctuations are modeled using a Karhunen–Loeve Expansion formulation. The response dynamics and the associated fatigue damage of the aeroelastic system, possessing different sources of nonlinearities, are systematically investigated under isolated cases of deterministic and stochastic input flows. Specifically, the pertinent role of stochasticity in the input flow is brought out by presenting the response dynamics and the associated fatigue damage accumulation for different values of noise intensity and time scale of the input flow fluctuation. It is demonstrated that under fluctuating flow conditions, the dynamics intermittently switch between attached flow and the dynamic stall regimes even at low mean flow speeds. The intermittent nature of the response varies as the time scale and intensity of the oncoming flow are varied. The role of torsional stresses as the predominant component dictating the fatigue damage accumulation irrespective of the source of nonlinearity is illustrated. Using the rainflow counting method and Miner’s linear damage accumulation theory, it is shown that the accumulated fatigue damage is substantially higher under stochastic flow conditions as compared to deterministic input flows. Importantly, it is observed that different time scales and intensities of the oncoming flow fluctuation play a pivotal role in dictating the fatigue damage in aeroelastic systems. Finally, fatigue damage is observed to be significantly higher for torsionally dominant oscillations in the dynamical stall regime compared to the oscillations at the attached flow regime. The numerical findings are strengthened by drawing comparisons with the preliminary results obtained from wind tunnel experiments performed on a NACA 0012 airfoil undergoing dynamic stall. To the best of our knowledge, this is the first study that systematically bridges the dichotomy between the stall-induced dynamical signatures in stochastic aeroelastic systems and maps the same to the corresponding structural damage.

Enhanced prediction intervals of tunnel-induced settlement using the genetic algorithm and neural network

This paper constructs the prediction intervals (PIs) of the tunnels’ settlement caused by the shielding steering process. The hybrid genetic algorithm -neural network (GA-NN) is developed to obtain the upper and lower bound of the settlement based on a series of shield operating parameters, geological condition parameters, tunnel geometric parameters, and anomalous conditions. An improved prediction interval-based cost function is proposed to enable the consideration of the uncertainty from model misspecification and noise variance. The genetic algorithm optimizes the weighted parameters in the neural network by minimizing the cost function value. This study adopts a metro tunnel construction case in China to verify the effectiveness of the proposed hybrid genetic algorithm-neural network approach. The results based on the study case illustrate the superiority of the proposed hybrid approach in (1) overcoming the limitations of the conventional prediction interval indicator; (2) achieving comparative results with the deterministic estimation based on the least squares support vector machine; (3) providing a probability prediction of the settlement only based on deterministic input multivariables. Overall, this study contributes to (1) the uncertainty assessment of tunnel settlement based on the deterministic variables, (2) developing a new PIs based cost function which is stable and reliable, (3) the engineering practice for a safer assessment based on the prediction intervals.

Deep generative model for probabilistic wind speed and wind power estimation at a wind farm

AbstractThis work introduces a novel method to generate probabilistic hub‐height wind speed forecasts aimed at power output prediction. We employ state‐of‐the‐art convolutional variational autoencoders (CVAEs) trained with historical wind speed observations, multivariable outputs (wind speed, direction, temperature, pressure, and humidity) from a numerical weather prediction (NWP) model and spatio‐temporal encodings. After training, we exploit the CVAE data generating capabilities to produce probabilistic forecasts from the same deterministic dynamical NWP model. The resulting probabilistic forecast provides an insight into the uncertainty of the original deterministic input that compensate for errors due to numerical discretization, inaccuracies in initial/boundary conditions and parameterizations and, most importantly wind speed fluctuations due to complex terrain features. To show the performance of the proposed model, we validate the approach with forecasted and observed data for fifteen sites in a wind farm in Awaji Island, Japan, in a challenging zone with complex topography and highly fluctuating wind patterns. We show that the proposed method provides improved wind speed forecasts from both deterministic (reduced root mean square error) and probabilistic (reduced continuous ranked probability score) standpoints. We also use ensemble forecast validation methods to assess the statistical properties of the CVAE and conclude that in zones with rapidly changing wind speed dynamics, the CVAE ensemble performs in a superior way when compared to the analog ensemble method. Finally, we also provide an insight on how the results can be used to obtain reliable wind power estimations.

Determination of the spatial distribution of wetting in the pore networks of rocks

HypothesisThe macroscopic movement of subsurface fluids involved in CO2 storage, groundwater, and petroleum engineering applications is controlled by interfacial forces in the pores of rocks. Recent advances in modelling these systems has arisen from approaches simulating flow through a digital representation of the complex pore structure. However, further progress is limited by difficulties in characterising the spatial distribution of the wetting state within the pore structure. In this work, we show how observations of the fluid coverage of mineral surfaces within the pores of rocks can be used as the basis for a quantitative 3D characterisation of heterogeneous wetting states throughout rock pore structures. ExperimentsWe demonstrate the approach with water–oil fluid pairs on rocks with distinct lithologies (sandstone and carbonate) and wetting states (hydrophilic, intermediate wetting, and heterogeneously wetting). FindingsFluid surface coverage the within rock pores is a robust signal of the wetting state across varying rock types and wetting states. The wetting state can be quantified and the resulting 3D maps can be used as a deterministic input to pore scale models. These may be applied to multiphase flow problems in porous media ranging from soil science to fuel cells.

Bayesian Cramér-Rao Lower Bound for Magnetic Field-Based Localization

In this paper, we show how to analyze the achievable position accuracy of magnetic localization based on Bayesian Cramér-Rao lower bounds and how to account for deterministic inputs in the bound. The derivation of the bound requires an analytical model, e.g., a map or database, that links the position that is to be estimated to the corresponding magnetic field value. Unfortunately, finding an analytical model from the laws of physics in not feasible due to the complexity of the involved differential equations and the required knowledge about the environment. In this paper, we therefore use a Gaussian process (GP) that approximates the true analytical model based on training data. The GP ensures a smooth, differentiable likelihood and allows a strict Bayesian treatment of the estimation problem. Based on a novel set of measurements recorded in an indoor environment, the bound is evaluated for different sensor heights and is compared to the mean squared error of a particle filter. Furthermore, the bound is calculated for the case when only the magnetic magnitude is used for positioning and the case when the whole vector field is considered. For both cases, the resulting position bound is below 10 cm indicating an high potential accuracy of magnetic localization.

An International Comparative Study of Students' Scientific Explanation Based on Cognitive Diagnostic Assessment.

Scientific explanation is one of the most core concepts in science education, and its mastery level is crucial for a deep understanding of the nature of science. As a new generation of assessment theory, cognitive diagnostic assessment (CDA) can get the knowledge of students' mastery of fine-grained knowledge. Based on the extant research, this research has formed eight attributes of scientific explanation concepts. By coding the Trends in International Mathematics and Science Study (TIMSS) test items, a CAD tool was formed. Data collected from 574 Grade 4 students in Hangzhou, China, combined with the data of the United States, Singapore, Australia, the United Kingdom, and Russia, were used in our study. The Deterministic Inputs, Noisy “And” gate (DINA) model was used to analyze the results from three aspects: the probability of mastery of attributes, the international comparison of knowledge states, and the analysis of learning paths. This study provided a new perspective from a CDA approach on the assessment of scientific explanation.

An Empirical Q-Matrix Validation Method for the Polytomous G-DINA Model.

A number of empirically based Q-matrix validation methods are available in the literature, all of which were developed for cognitive diagnosis models (CDMs) involving dichotomous attributes. However, in many applications, it is more instructionally relevant to classify students into more than two categories (e.g., no mastery, basic mastery, and advanced mastery). To extend the practical utility of CDMs, methods for validating the Q-matrix for CDMs that measure polytomous attributes are needed. This study focuses on validating the Q-matrix of the generalized deterministic input, noisy, "and" gate model for polytomous attributes (pG-DINA). The pGDI, an extension of the G-DINA model discrimination index, is proposed for polytomous attributes. The pGDI serves as the basis of a validation method that can be used not only to identify potential misspecified q-entries, but also to suggest more appropriate attribute-level specifications. The theoretical properties of the pGDI are underpinned by several mathematical proofs, whereas its practical viability is examined using simulation studies covering various conditions. The results show that the method can accurately identify misspecified q-entries and suggest the correct attribute-level specifications, particularly when high-quality items are involved. The pGDI is applied to a proportional reasoning test that measures several polytomous attributes.

Input design for Bayesian frequency response identification via convex programming

This paper proposes a novel approach for the input design of the Bayesian frequency response identification. The input design problem is first formulated by considering deterministic inputs, which leads to a non-convex optimization problem. Restricting inputs to be periodic makes the problem convex but introduces conservativeness. The stochastic consideration facilitates the input design problem to be presented as a convex problem whose decision variables are a finite number of autocorrelation coefficients. A multiplied sampled input turns out to be the good input with a certain probability. Then, the optimal input is selected to be the minimum energy one among the good inputs. Simulations results demonstrate the effectiveness of the proposed method.

Investigating the Relationships Between Listening Skills and Genre Competence Through Cognitive Diagnosis Approach

Although research on listening skills has been frequently conducted to discover the nature of listening comprehension, there is little study about listening genre competence which is related with knowledge about listening text types. In order to find out whether listening skills and listening genre competence are related, cognitive diagnosis, a quantitative method to disclose finer-grained latent attributes, was adopted in this study. The generalized deterministic inputs, noisy “and” gate (G-DINA) model, which takes attribute compensation and attribute interaction into consideration, was used to carry out cognitive diagnostic analysis. The listening comprehension subtest of Band 4 of Test for English Majors (TEM) which is a large scale English proficiency test for English Majors in China was used as the proficiency test for homogenizing the participants. Three genres in the subtest, dialog, lecture, and news, were investigated. The 2,285 subjects were sophomore English major college students and also test-takers of the same TEM4 examination. They were chosen by random sampling from the nationwide test population in China. The study analyzed three types of relationships between listening skills and genre competence. By analyzing how mastery of certain listening text genres goes with mastery of listening skills according to latent class distribution, the coexistence relationship was discovered. By comparing the average number of skills/genres mastered when the number of genres/skills mastered increases through One-Way ANOVA, compensatory and contributory relationships were revealed. The study also found that the subjects mastering Lecture genre got higher listening scores.

Research on the Cognitive Diagnosis of Chinese Listening Comprehension Ability Based on the G-DINA Model.

As a new generation of measurement theory, cognitive diagnosis theory shows significant potential and advantages in educational evaluation in that it combines a cognitive process and a measurement method. The application of the theory not only reveals the potential characteristics of learners in cognitive processing, but also provides targeted remedies and strategic guidance for individuals. Given the difficulties of traditional assessment models in providing an insightful and fine-grained account for individualized and procedural learning, providing personalized learning strategies for learners of Chinese as a second language has been a new goal of teaching and measurement in Chinese listening. This study constructs a cognitive diagnosis model of Chinese listening comprehension for Chinese-as-a-second-language learners through theoretical exploration, model hypotheses, repeated verification, and model modification. The results show that the Q-matrix (Q3) constructed by the experts within modification has the highest fitting degree with the empirical data. The parameter recovery rate, the accuracy of the tested attribute or mode, and the relative fitting index obtained from the simulation study are consistent with the information extracted from the empirical data. The diagnostic reliability and effectiveness of generalized deterministic inputs, noise “and” gate (G-DINA) are higher than those of DINA, deterministic inputs, noisy “or” gate (DINO), and reduced reparametrized unified model (RRUM). In the estimation of the item and subject parameters, the G-DINA model shows good convergence, and the average classification accuracy rate based on attribute level is 0.861.