Gaussian Process Models for Computer Experiments With Qualitative and Quantitative Factors

We propose a flexible yet computationally efficient approach for building Gaussian process models for computer experiments with both qualitative and quantitative factors. This approach uses the hypersphere parameterization to model the correlations of the qualitative factors, thus avoiding the need of directly solving optimization problems with positive definite constraints. The effectiveness of the proposed method is successfully illustrated by several examples.

This article uses neural networks in forecasting stock market prices. With their ability to discover patterns in nonlinear and chaotic systems, neural networks offer the ability to predict market directions more accurately than current techniques such as technical analysis, fundamental analysis, and regression compared with neural network performance. Proposed intelligent stock market prediction system is based on the quantitative and qualitative factors. Three feedforward neural models can be used to analyze these factors. Input data to the neural network proposed are quantitative factors. Input data to the neural network proposed for qualitative factors can be factors related to the political effect considered. Third neural network consists of decision integration in which input data will be the outputs of above-mentioned neural networks. This facilitates to make right decision whether stock market is influenced by quantitative or qualitative factors

This paper presents a qualitative and quantitative research to explore and understand the agents that determine the level of productiveness and the importance of qualitative factors. A questionnaire was conducted to estimate the outcome of the qualitative agents or factors in improving the productivity of microtunneling projects. Information was collected from 26 competent professionals in microtunneling projects for almost 20years through an expert's opinion, using a semi-structured questionnaire. The Analytical Hierarchy Process and the clustering analysis technique have been used to account for the twenty qualitative agents or factors concerned in developing a proposed assessment model. The present research has developed an assessment model that stems from the result of qualitative factors on microtunneling to investigate the effect induced by the qualitative agents on productiveness. The assessment model is thereafter confirmed, showing an analytical outcome of 79.85% moderately justified percent. This study proved that the qualitative factors affecting the productivity of microtunneling projects can be assessed through the productivity index concept. This paper helps the industry parties to understand the factors influencing the success of microtunneling projects as well as, development of an assessment model to calculate productivity. Most of the past research on productivity models are based on quantitative factors only. This paper presents qualitative research to explore and understand the factors that affect productivity and the importance of qualitative factors. The developed assessment model presented is the first in connecting theory, analytical methods, and experts' opinions to help practitioners to understand and assess the productivity of microtunneling projects.

Agronomic experiments are often complex and difficult to interpret, and the proper use of appropriate statistical methodology is essential for an efficient and reliable analysis. In this paper, the basics of the statistical analysis of designed experiments are discussed using real examples from agricultural field trials. Factorial designs allow for the study of two or more treatment factors in the same experiment, and here we discuss the analysis of factorial designs for both qualitative and quantitative level treatment factors. Where treatment factors have quantitative levels, models of treatment effects are essential for efficient analysis and in this paper we discuss the use of polynomials for empirical quantitative modelling of treatment effects. The example analyses cover experiments with a single quantitative level factor, experiments with mixtures of quantitative and qualitative level factors, polynomial regression designs with two quantitative level factors, split‐plot designs with quantitative level factors and repeated‐measures designs with correlated data and a quantitative treatment response over time. Modern mixed model computer software for routine analysis of experimental data is now readily available, and we demonstrate the use of two alternative software packages, the SAS package and the R language. The main purpose of the paper is to exemplify standard statistical methodology for routine analysis of designed experiments in agricultural research, but in our discussion we also provide some references for the study of more advanced methodology.

The Gaussian process (GP) model provides a powerful methodology for calibrating a computer model in the presence of model uncertainties. However, if the data contain systematic experimental errors, then the GP model can lead to an unnecessarily complex adjustment of the computer model. In this work, we introduce an adjustment procedure that brings the computer model closer to the data by making minimal changes to it. This is achieved by applying a lasso-based variable selection on the systematic experimental error terms while fitting the GP model. Two real examples and simulations are presented to demonstrate the advantages of the proposed approach. This article has supplementary material available online.

The purpose of the study is to form an economic and mathematical instrument for assessing the role of quantitative and qualitative factors in changing the output of dairy products. Objectives: to determine the main theoretical provisions for calculating the indicator of dairy production depending on the resources used; to present a formalized expression for calculating quantitative factors of dairy output growth; to substantiate methodological tools for calculating qualitative, intensive factors of dairy output growth. The object of research is the scientific and technological development of the dairy subcomplex. The economic-mathematical and economic-statistical research methods were used. It has been established that the scientific and technological development of the dairy subcomplex is expressed, first of all, in an increase in the values of qualitative factors, the formal expression of which may be the values of labor productivity, capital productivity, yield or productivity; In periods with a constant level of technical development, quantitative factors receive priority - the number of workers, the cost of basic funds, the acreage, and the number of cows. A toolkit has been developed to assess the cumulative impact of both quantitative and qualitative factors on the resulting indicator of milk production, which will make it possible to reasonably assess the scientific and technological development of the dairy subcomplex. The qualitative factors include the productivity of the workers of the dairy subcomplex, the return on capital and the productivity of cows. Possible combinations of the influence of factors on the resulting indicator of milk production are determined. The key features of changes in influencing variables during periods with a constant level of technical development and in the context of the emergence of new, innovative technological solutions are characterized. Possible directions for further research to assess the scientific and technological de-velopment of the dairy-food subcomplex are outlined.

Exact [formula omitted]-optimal designs for a second-order response surface model on a circle with qualitative factors

The multi-criteria facilities layout problem can be formulated as a quadratic assignment model that can handle multiple qualitative and quantitative factors in the objective function. Some studies have shown that the techniques and tools of facilities layout problems can equally be applied for the layout design of user interface components in human-computer interface. This paper presents an alternate approach, which handles multiple qualitative and quantitative factors in a different manner separately in the objective function to obtain the initial layouts. The proposed approach also consists of a layout procedure, in which the pair of facilities with the least composite criterion value has been selected to be placed far apart in the layout to generate an initial layout in the construction procedure. The results of the proposed approach are compared with that of an existing approach which handles a number of qualitative and quantitative factors in the same manner as in the objective function to obtain the initial layouts for the example task of the user interface components layout problem under consideration.

: This thesis analyzes the effect of quantitative and qualitative factors used in the admissions process at the U.S. Naval Academy in determining the performance of candidates admitted. In determining student performance, graduation, Order of Merit (OOM), cumulative academic Quality Point Ratio (QPR), cumulative military QPR, and striper selection are used as performance outcome measures. The data is from Naval Academy graduation year groups 1995 through 2001. The analysis separates the Naval Academy's Whole Person Multiple into quantitative and qualitative inputs. The Candidate Multiple (CM) is the quantitative input to the admissions process derived from a statistics-based scoring model anchored in proven high school performance measures such as the SAT and high school GPA. The Recommendations of the Admissions Board (RAB) are the qualitative input, which awards points for subjective traits not captured in the CM or that are from various other subjective measures, such as student interviews and essays. This research highlights the properties of the two admissions factors and their estimated impact on student performance. The results show that student performance increased as CM and RAB increased, revealing the importance of a combined quantitative and qualitative admissions process, and emphasizing qualitative input as the value added factor that provides an increased predictability of student success. (26 tables, 22 figures, 26 refs.)

Risk-Value-Cost-based Optimization of Global Value-adding Structures

PurposeThe purpose of this study is to investigate the decision-making process of librarians at the University of Western Ontario who attempted to cancel the Wiley Big Deal. The aim of the study is to reveal the underlying factors that affected their decision-making process. By understanding the decision-making process of librarians, it may be possible to devise a system that takes into consideration not only quantitative factors but also the subjective or qualitative factors that impact librarians’ decisions and thus make it easier to cancel these Big Deals.Design/methodology/approachThe study involved administering an online survey to 25 librarians involved in the cancellation project. Follow-up interviews were conducted with 13 of these librarians to understand at a deeper and more nuanced level the factors that influenced their decisions.FindingsThe main finding was that the librarians who participated in the study could be divided into two groups – a data-driven criteria group and a subjective criteria group – based on their ranking of the factors used to make their cancellation decisions. Most librarians interviewed used a mixture of quantitative factors and qualitative factors when making their cancellation decisions. The authors found that those participants who had greater professional experience and a closer relationship with the faculties in their subject areas had more difficulty in cancelling journals. Very few librarians relied on quantitative data alone.Originality/valueThis study is one of few that have examined the subjective factors that influence librarians’ decisions regarding cancellation of Big Deals. It has implications regarding the movement towards centralized collection management and reliance on quantitative data alone when making collection decisions.

The multifactor facilities layout problem is one of the important cases of quadratic assignment problem (QAP), which deals with the problem of optimum assignment of n facilities to n locations and optimizing multifactor cost objective. The authors present an alternate mathematical model for QAP that combines multiple qualitative and quantitative factors separately, treating them in a different manner for the facilities layout problem. The composite factor is obtained by combining the weighted combined qualitative and quantitative factors. The distances between the locations weigh the composite factor. The objective function is to minimize the total weighted workflow distance between the facilities. The results of the proposed model are compared with that of an existing model that handles multiple qualitative and quantitative factors in the same manner.

The uncertainty of measurement for various testing operations is affected by quantitative and qualitative factors. The effects of quantitative factors can be estimated using statistical approach, but the same for qualitative factors is difficult to achieve. The measurement of uncertainty in case of various testing processes may be affected by qualitative factors like operator, sample, environmental conditions and machine interference. Various sub-factors related to operator like knowledge, experience, training and attitude may affect the overall performance of test results. The analytical hierarchy process (AHP) is the decision making tool used for ranking of these factors. The objective of paper is to rank these factors using AHP, which is used by the management or administrator to set the strategy for quality improvement by decreasing the uncertainty of measurement.

Elevated temperatures limit the peak performance of systems because of frequent interventions by thermal throttling. Non-uniform thermal states across system nodes also cause performance variation within seemingly equivalent nodes leading to significant degradation of overall performance. In this paper we present a framework for creating a lightweight thermal prediction system suitable for run-time management decisions. We pursue two avenues to explore optimized lightweight thermal predictors. First, we use feature selection algorithms to improve the performance of previously designed machine learning methods. Second, we develop alternative methods using neural network and linear regression-based methods to perform a comprehensive comparative study of prediction methods. We show that our optimized models achieve improved performance with better prediction accuracy and lower overhead as compared with the Gaussian process model proposed previously. Specifically we present a reduced version of the Gaussian process model, a neural network–based model, and a linear regression–based model. Using the optimization methods, we are able to reduce the average prediction errors in the Gaussian process from $4.2^\circ$ C to $2.9^\circ$ C. We also show that the newly developed models using neural network and Lasso linear regression have average prediction errors of $2.9^\circ$ C and $3.8^\circ$ C respectively. The prediction overheads are 0.22, 0.097, and 0.026 ms per prediction for reduced Gaussian process, neural network, and Lasso linear regression models, respectively, compared with 0.57 ms per prediction for the previous Gaussian process model. We have implemented our proposed thermal prediction models on a two-node system configuration to help identify the optimal task placement. The task placement identified by the models reduces the average system temperature by up to $11.9^\circ$ C without any performance degradation. Furthermore, these models respectively achieve 75, 82.5, and 74.17 percent success rates in correctly pointing to those task placements with better thermal response, compared with 72.5 percent success for the original model in achieving the same objective. Finally, we extended our analysis to a 16-node system and we were able to train models and execute them in real time to guide task migration and achieve on average 17 percent reduction in the overall system cooling power.

In the process of bridge design, it is necessary to select the best from a number of feasible alternatives by comparison and selection or elimination. The set of criteria used to evaluate and select the satisfactory alternative consists of many qualitative and quantitative factors. Therefore, the selection is a problem of multi-criteria and semi-structural decision making. A new framework and methodology for evaluation of bridge alternatives, which differs from the traditional methods used in semi-structural decision making, is presented in this paper. First the set of criteria for the evaluation of alternatives is established. Next, the method of construction of the relative membership degree matrix, in which both qualitative and quantitative factors are consistent with the uniform calculating standard, is studied. A new weight-assessing method is then developed for calculation of the weightings, based on the relative membership degree matrix and finally a fuzzy optimum model is adopted to select the various alternative bridge designs that satisfy the criteria. A case study is then presented, to show the efficiency and robustness of the new framework and methodology.