An accumulative error based adaptive design of experiments for offline metamodeling

A popular method to reduce the computational effort in simulation-based engineering design is by way of approximation. An approximation method involves two steps: Design of Experiments (DOE) and metamodeling. In this paper, a new DOE approach is introduced. The proposed approach is adaptive and samples more design points in regions where the simulation response is expected to be highly nonlinear and multi-modal. Numerical and engineering examples are used to demonstrate the applicability of the proposed DOE approach. The results from these examples show that for the same number of simulation evaluations and according to metamodel accuracy, the proposed DOE approach performs better for majority of test examples compared to two previous methods, i.e., the maximum entropy design method and maximum scaled distance method.     

Improving multi-objective genetic algorithms with adaptive design of experiments and online metamodeling

Applications of multi-objective genetic algorithms (MOGAs) in engineering optimization problems often require numerous function calls. One way to reduce the number of function calls is to use an approximation in lieu of function calls. An approximation involves two steps: design of experiments (DOE) and metamodeling. This paper presents a new approach where both DOE and metamodeling are integrated with a MOGA. In particular, the DOE method reduces the number of generations in a MOGA, while the metamodeling reduces the number of function calls in each generation. In the present approach, the DOE locates a subset of design points that is estimated to better sample the design space, while the metamodeling assists in estimating the fitness of design points. Several numerical and engineering examples are used to demonstrate the applicability of this new approach. The results from these examples show that the proposed improved approach requires significantly fewer function calls and obtains similar solutions compared to a conventional MOGA and a recently developed metamodeling-assisted MOGA.     

An advanced supply chain management tool based on modeling and simulation

The paper presents an advanced modeling approach and a simulation model for supporting supply chain management. The first objective is to develop a flexible, time-efficient and parametric supply chain simulator starting from a discrete event simulation package. To this end we propose and advanced modeling approach. The second objective is to provide a decision making tool for supply chain management. The simulator is a decision making tool capable of analyzing different supply chain scenarios by using an approach based on multiple performance measures and user-defined set of input parameters. Our simulator capabilities as decision making tool are strongly amplified if Design of Experiment (DOE) and Analysis of Variance (ANOVA) are respectively used for experiments planning and simulation results analysis. With regard to supply chain decision making process, we propose an application example for a better understanding of tool potentials. The application example considers a specific supply chain scenario and analyzes the effects of inventory control policies, lead times, customers’ demand intensity and variability, on three different supply chain performance measures.     

Knowledge based formation of re-configurable assembly cells

When developing assembly cells with highly complex modular structures, designers need to translate user requirements into a set of design rules and potential cell configurations. The success in matching user requirements to potential products is dependent on how well the functional and non-functional customer requirements can be understood and translated into cell features (design rules, processes and module types). This paper reports on a knowledge based methodology for forming customisable re-configurable assembly cells. The approach is based on matching user requirements to existing supplier knowledge in terms of design rules and principles, modules offered by different vendors, new emerging technologies and existing own and competitors’ products. The decision making includes requirements analysis, generating assembly processing alternatives and evaluating and selecting assembly modules and cells. The proposed approach aims to assist decision making in assembly system design by enabling users and suppliers to jointly participate in an interactive and iterative process of forming re-configurable assembly cells.     

Using the Perceptual Cycle Model and Schema World Action Research Method to generate design requirements for new avionic systems

The Schema World Action Research Method (SWARM) has previously been used as a means to explore the underlying decision‐making processes involved in retrospective incident reports. The approach has been fruitful in capturing all interacting processes involved in managing incidents. This paper proposes that SWARM may also be used prospectively within the early phases of the design lifecycle for new behavioral‐based intervention strategies. Six pilot interviews were conducted to explore pilots' perceptual cycle processing when faced with a suspected engine oil leak. The aim was to explore whether there may be any deficiencies within current practise and explore ways in which pilots may be better supported in dealing with abnormal system parameters such as this. A number of design recommendations are proposed for a new avionic system capable of supporting and guiding pilots through the decision‐making process.     

Hesitant 2-tuple linguistic Bonferroni operators and their utilization in group decision making

Hesitant 2-tuple linguistic variable realizes a graded information approach to characterize the uncertainty of human cognition. This study is concerned with the development of new aggregation operators and aims to design a new group decision making approach to address the information fusion involving the interrelationship between aggregated terms and the prioritization relationship among decision makers under hesitant 2-tuple linguistic situation. Firstly, hesitant 2-tuple linguistic Bonferroni mean (H2TLBM) operator and prioritized weighted hesitant 2-tuple linguistic Bonferroni mean (PWH2TLBM) operator are established. Subsequently, some pertinent properties and special forms of the developed operators are studied in detail. To employ the proposed operators to solve group decision making problems, a novel TODIM (an acronym in Portuguese of interactive and multiple attribute decision making) method based on possibility degree is developed under the situation of hesitant 2-tuple linguistic information. The designed decision making method not only inherits the merits of the traditional TODIM approach, but also characterizes the interrelationship of criteria. The detailed process of solving problems is exemplified to highlight the practicality and feasibility of the designed method. Furthermore, comparative analysis with other methods is carried out to further offer insights on the designed decision method.     

An effective approach to teach design of experiments (DOE) using calculation-and-analysis worksheets and computerized spreadsheets

The École de technologie supérieure (ÉTS) of the University of Québec is offering Design of Experiments (DOE) courses for more than a decade within its undergraduate certificate program of Quality Management and Assurance (QMA). Previous teaching experiences showed that the conceptual aspects of DOE has been somehow difficult to be rapidly and efficiently assimilated by the part-time adult students. Although DOE have proved to be very effective in improving quality throught process parameters' optimization, many engineers rejected the idea of applying the technique on the floor level because of the conceptual and statistical barriers. For these reasons, the authors have developed and introduced a new teaching approach in order to clarify the conceptual aspects and simplify the mechanics underlying the DOE techniques and, therefore, help the students to better understand and implement DOE at the work place. The approach is mainly based on the use of simple and straight forward calculation-and-analysis worksheets that have been developed for the most commonly used experimental designs (full factorial, screening factorial, screening fractionnal factorial and Taguchi arrays). In this paper, the authors present the approach with a sample of the most pertinent calculation-and-analysis worksheets used along with a practical example.     

Prediction of product design decision Making: An investigation of eye movements and EEG features

Design decision making is happened in every design node and iteration, and the expert decision-making bias and personal preference will ultimately affect the success or failure of the product reaching the market. In this paper, we try to predict the design decision making by investigating the relations between design decision making and subjects’ eye movements and Electroencephalogram(EEG) response. Four different methods were applied and compared to classify the different EEG features and two methods were used for EEG feature selection to correspond the design decision making results. In this study, the authors applied a multimodal fusion strategy for design decision making recognition where the authors used eye tracking and EEG response data as input dataset. According to the experiment results, the performance of the fusion strategy combined with EEG signals and eye movement characteristics is well in fitting the expert decision making results. The multimodal fusion combining eye tracking data and EEG has a strong potential to be a new design decision method to guide the design practice and provide supportive and objective data to reduce the effects of subjectivity, one-sidedness and superficiality in decision making. These results show that it is possible to create a classifier based on features extracted from eye movements and EEG response for the design decision making behaviour.     

On decentralized estimation and control with application to freeway ramp metering

This paper presents an approach to the suboptimal design of time invariant, infinite time horizon, L-Q-G regulator in which decision making and state estimation are decentralized into "local" subsystems. For a prespecified information pattern (set of admissible intercommunications) the problem of finding a best linear control law and state estimator is converted to a parameter optimization which can be solved off line. A modified Davidon-Fletcher-Powell algorithm is used to obtain numerical solutions to the static optimization. A design is illustrated by means of a 12 state freeway corridor ramp metering problem; behavior of centralized and decentralized designs in a nonlinear macroscopic simulation of the freeway are provided.     

Cross-validation based single response adaptive design of experiments for Kriging metamodeling of deterministic computer simulations

A new approach for single response adaptive design of deterministic computer experiments is presented. The approach is called SFCVT, for Space-Filling Cross-Validation Tradeoff. SFCVT uses metamodeling to obtain an estimate of cross-validation errors, which are maximized subject to a constraint on space filling to determine sample points in the design space. The proposed method is compared, using a test suite of forty four numerical examples, with three DOE methods from the literature. The numerical test examples can be classified into symmetric and asymmetric functions. Symmetric examples refer to functions for which the extreme points are located symmetrically in the design space and asymmetric examples are those for which the extreme regions are not located in a symmetric fashion in the design space. Based upon the comparison results for the numerical examples, it is shown that SFCVT performs better than an existing adaptive and a non-adaptive DOE method for asymmetric multimodal functions with high nonlinearity near the boundary, and is comparable for symmetric multimodal functions and other test problems. The proposed approach is integrated with a multi-scale heat exchanger optimization tool to reduce the computational effort involved in the design of novel air-to-water heat exchangers. The resulting designs are shown to be significantly more compact than mainstream heat exchanger designs.     

Group decision making problems in a linguistic and dynamic context

The aim of this paper is to present a new model of decision support system for group decision making problems based on a linguistic approach and dynamic sets of alternatives. The model incorporates a mechanism that allows to manage dynamic decision situations in which some information about the problem is not constant in time. We assume that the set of alternatives can change during the decision making process. The model is presented in a mobile and dynamic context where the experts’ preferences can be incomplete. The linguistic approach is used to represent both the experts’ preferences about the alternatives and the agreement degrees to manage the change of some alternatives. A prototype of such mobile decision support system in which the experts use mobile devices to provide their linguistic preferences at anytime and anywhere has been implemented. In such a way, we provide a new linguistic group decision making framework that is mobile and dynamic.     

Decision making process: typology, intelligence, and optimization

Decision making is concerned with evaluating and/or ranking possible alternatives of action. In this paper, we develop a model for the process of decision making. Understanding the decision process can provide insights into how humans make decisions, understand their decision making approaches, and how they differ from each other. We believe that decision makers who are conscious of their decision process types can make more effective and balanced decisions. In this paper, we present a new decision process model based on the following four dimensions where each dimension is defined by two opposing types: Information Processing (Concrete and Abstract), Alternative Generation (Adaptive and Constructive), Alternative Evaluation (Moderate and Bold), and Decision Closure (Organized and Flexible). Furthermore, an approach for assessing each of the four decision process types by a mathematical function is presented. In a much boarder scope than decision making, these assessed functions can be used to evaluate and rank alternatives. The decision process model can also be used in conjunction with multiple criteria decision making and multiple objective optimization. The model can also be used to explain the reasons that the classical decision making models fail to describe real decision makers?? behavior, and mistakenly label such behavior as irrational. The proposed decision process model can be used for developing new behavioral, rational, and intelligent decision making theories and approaches. Extensions of this work may include group decision making, organizational decision making, team formation, and risk behavior analysis. Experimental results of over four hundred engineering students are reported. A web site has been developed for users (http://car.cwru.edu/decision/).     

Modeling decisions in intelligent user interfaces

In this paper, we are concerned with the run-time communication decisions which need to be made by an intelligent user interface. We model communication decisions as a decision-making process, where a selection among a set of alternative constituents is needed, in order to satisfy specific design goals. Based on techniques from the domains of multiple-criteria decision making and decision making under doubt, we propose the use of a model which takes into account the significance of each constituent toward the satisfaction of each design goal, as well as the consideration of the decision maker (interface designer) regarding the significance of each design goal. Following the proposed approach, a design strategy can be represented by a design vector, and thus, we can define properties of, and relationships between, different design strategies, based on their respective vectors. A specific example deploying the proposed model is presented, based on data from the relevant literature. © 1997 John Wiley & Sons, Inc.     

综合集成与决策

在探索“从定性到定量的综合集成法”提出的动机与基本原理的基础上，讨论了综合集成与决策之间的关系，认为从定性到定量的综合集成体系的研究不能局限于决策与决策支持系统，而应将其视为科学创新的方法论．面对当前众多复杂的重大决策问题，应以综合集成理论作为决策研究的指导，建立综合集成型决策支持系统．     

A balanced solution of a fuzzy soft set based decision making problem in medical science

The purpose of this paper is two folded. Firstly, the concept of mean potentiality approach (MPA) has been developed and an algorithm based on this new approach has been proposed to get a balanced solution of a fuzzy soft set based decision making problem. Secondly, a parameter reduction procedure based on relational algebra with the help of the balanced algorithm of mean potentiality approach has been used to reduce the choice parameter set in the parlance of fuzzy soft set theory and it is justified to the problems of diagnosis of a disease from the myriad of symptoms from medical science. Moreover the feasibility of this proposed method is demonstrated by comparing with Analytical Hierarchy Process (AHP), Naive Bayes classification method and Feng's method.     

An enhanced Support Vector Machine classification framework by using Euclidean distance function for text document categorization

This paper presents the implementation of a new text document classification framework that uses the Support Vector Machine (SVM) approach in the training phase and the Euclidean distance function in the classification phase, coined as Euclidean-SVM. The SVM constructs a classifier by generating a decision surface, namely the optimal separating hyper-plane, to partition different categories of data points in the vector space. The concept of the optimal separating hyper-plane can be generalized for the non-linearly separable cases by introducing kernel functions to map the data points from the input space into a high dimensional feature space so that they could be separated by a linear hyper-plane. This characteristic causes the implementation of different kernel functions to have a high impact on the classification accuracy of the SVM. Other than the kernel functions, the value of soft margin parameter, C is another critical component in determining the performance of the SVM classifier. Hence, one of the critical problems of the conventional SVM classification framework is the necessity of determining the appropriate kernel function and the appropriate value of parameter C for different datasets of varying characteristics, in order to guarantee high accuracy of the classifier. In this paper, we introduce a distance measurement technique, using the Euclidean distance function to replace the optimal separating hyper-plane as the classification decision making function in the SVM. In our approach, the support vectors for each category are identified from the training data points during training phase using the SVM. In the classification phase, when a new data point is mapped into the original vector space, the average distances between the new data point and the support vectors from different categories are measured using the Euclidean distance function. The classification decision is made based on the category of support vectors which has the lowest average distance with the new data point, and this makes the classification decision irrespective of the efficacy of hyper-plane formed by applying the particular kernel function and soft margin parameter. We tested our proposed framework using several text datasets. The experimental results show that this approach makes the accuracy of the Euclidean-SVM text classifier to have a low impact on the implementation of kernel functions and soft margin parameter C.     

Generalized ordered weighted logarithm aggregation operators and their applications to group decision making

We present the generalized ordered weighted logarithm averaging (GOWLA) operator based on an optimal deviation model. It is a new aggregation operator that generalizes the ordered weighted geometric averaging (OWGA) operator. This operator adds to the OWGA operator an additional parameter. controlling the power to which the arguments are raised. We further generalize the GOWLA operator and obtain the generalized ordered weighted hybrid logarithm averaging (GOWHLA) operator. We next introduce a nonlinear objective programming model for determining GOWHLA weights and an approach to group decision making based on the GOWHLA operator. Finally, we present a numerical example to illustrate the new approach in human resource management problem. © 2010 Wiley Periodicals, Inc.     

Design matters for decision makers: Discovering IT investment alternatives

The literature has traditionally focused on how decision makers choose among alternatives based on rational behaviors, organizational politics, and organized anarchies. However, there is little research on decision making as a creative process where managers discover and evaluate alternatives. We address this gap by proposing a design approach to transform decisions in organizations. The approach views decision making as a creative and adaptive process in which managers recursively collect and interpret heterogeneous evidence, explore and test different ideas, and discover and evaluate alternatives. While the approach is rooted in the know-how of the involved stakeholders, its underpinning design attitude invites managers to better understand possible alternatives based on analyses of structured as well as unstructured data. To assess its utility, we applied the approach to IT investment evaluations through a Collaborative Practice Research project involving Swedish public organizations. Building on the project findings, we discuss the proposed approach and spell out its implications for transforming decision making in organizations.     

Model-based multi-objective decision making under deep uncertainty from a multi-method design lens

Several approaches within the exploratory modelling literature—each with strengths and limitations—have been introduced to address the complexity and uncertainty of decision problems. Recent model-based approaches for decision making emphasise the advantage of mixing approaches from different areas in leveraging the strengths of each. This article shows how a multi-method lens to the design of decision-making approaches can better address different characteristics of multi-objective decision problems under deep uncertainty. The article focuses on interactions between two broad areas in model-based decision making: exploratory modelling and multi-objective optimisation. The article reviews this literature using a specific multi-method lens to analyse previous researches and to identify the knowledge gap. The article then addresses this gap by demonstrating a multi-method approach for designing adaptive robust solutions. The suggested approach uses a Pareto optimal search from multi-objective optimisation for enumerating alternative solutions. It also uses Robust Decision Making and Dynamic Adaptive Policy Pathways approaches from exploratory modelling for analysing the robustness of enumerated solutions in the face of many future scenarios. A hypothetical case study is used to illustrate how the approach can be applied. The article concludes that a new lens from a multi-method design perspective is needed on exploratory modelling to provide practical guidance into how to combine exploratory modelling techniques, to shed light on exiting knowledge gaps and to open up a range of potential combinations of exiting approaches for leveraging the strengths of each.     

Accelerating Assay Development using Experimental Design and Integrated Liquid Handling

Molecular biologists developing assays face complex challenges inherent with the rate of new target identification and their adaptation to automation-ready assays. In addition, the exponential growth of compound libraries over the last several years has required that assays be miniaturized as a means of containing costs.

Using AAO, the assay developer can take advantage of statistical design of experiment (a.k.a. DOE or experimental design) to create more robust, cost-effective assays. Making assays more robust for automated systems no longer has to be a bottleneck because AAO lets the scientist concentrate on the science, not the mundane tasks of pipetting or writing complex liquid handling methods.

AAO combines the power of statistically designed experiments with the flexibility of the Biomek 2000 and post-run data collection and analysis. No longer is the DOE approach to assay development cumbersome and impractical.