Incorporating deeply uncertain factors into the many objective search process

This paper proposes an approach for including deeply uncertain factors directly into a multi-objective search procedure, to aid in incorporating divergent quantitative scenarios within the model-based decision support process. Specifically, we extend Many Objective Robust Decision Making (MORDM), a framework for finding and evaluating planning solutions under multiple objectives, to include techniques from robust optimization. Traditional MORDM first optimized a problem under a baseline scenario, then evaluated candidate solutions under an ensemble of uncertain conditions, and finally discovered scenarios under which solutions are vulnerable. In this analysis, we perform multiple multi-objective search trials that directly incorporate these discovered scenarios within the search. Through the analysis, we have created multiple problem formulations to show how methodological choices of severe scenarios affect the resulting candidate planning solutions. We demonstrate the approach through a water planning portfolio example in the Lower Rio Grande Valley of Texas.     

基于优化决策树的化工企业风险监测算法

化工企业生产过程中的数据变化带有很强的随机性和非线性.国内的大型化工企业在生产运行中产生了大量的数据,数据属性众多,对数据属性的监控较为被动和片面.传统的基于主成份分析的数据属性分析方法,在面对大量化工数据时,数据主成份特征不明显,与危险相关的属性很可能被弱化,造成检测不准.为此提出一种区域PSO优化决策的化工企业异常情况检测方法,在使用主成份分析法对影响因素进行综合评定的基础上,在经过PSO优化权重系数后的决策树构造方法对数据的复杂情况进行寻优处理,消除大数据量的影响.实验结果证明,经优化的决策树方法能够将更加准确地对化工企业的生产平稳度进行综合控制与分析,对实际的生产有很好的借鉴作用.     

Consensus-based decision support for multicriteria group decision making

Consensus decision making is complex and challenging in multicriteria group decision making due to the involvement of several decision makers, the presence of multiple, and often conflicting criteria, and the existence of subjectiveness and imprecision in the decision making process. To ensure effective decisions being made, the interest of all the decision makers usually represented by the degree of consensus in the decision making process has to be adequately considered. This paper presents a consensus-based approach for effectively solving the multicriteria group decision making problem. The subjectiveness and imprecision of the decision making process is adequately handled by using intuitionistic fuzzy numbers. An interactive algorithm is developed for consensus building in the group decision making process. A decision support system framework is presented for improving the effectiveness of the consensus building process. An example is presented for demonstrating the applicability of the proposed approach for solving the multicriteria group decision making problem in real world situations.     

Weighted and robust learning of subspace representations

A reliable system for visual learning and recognition should enable a selective treatment of individual parts of input data and should successfully deal with noise and occlusions. These requirements are not satisfactorily met when visual learning is approached by appearance-based modeling of objects and scenes using the traditional PCA approach. In this paper we extend standard PCA approach to overcome these shortcomings. We first present a weighted version of PCA, which, unlike the standard approach, considers individual pixels and images selectively, depending on the corresponding weights. Then we propose a robust PCA method for obtaining a consistent subspace representation in the presence of outlying pixels in the training images. The method is based on the EM algorithm for estimation of principal subspaces in the presence of missing data. We demonstrate the efficiency of the proposed methods in a number of experiments.     

Evaluating project completion time in project networks with discrete random activity durations

Deterministic models for project scheduling suffer from the fact that they assume complete information and neglect random influences, that occur during project execution. A typical consequence is the underestimation of the project duration as frequently observed in practice. This phenomenon occurs even in the absence of resource constraints and has been the subject of extensive research in the scientific community. This paper presents a method for obtaining relevant information about the project makespan for scheduling models, with dependent random processing time available in the form of scenarios.     

The growth of service and the service of growth: Using system dynamics to understand service quality and capital allocation

This paper discusses how system dynamics can help understand a service company's growth potential as well as its limitations. The model discussed here is being built for a European restaurant chain which grew from nothing to over 200 outlets in less than a decade. The model highlights two conflicting pressures: the need to spend on meeting customer expectations and hence build sales versus the need to meet profit targets from headquarters and thus win the capital to fund expansion. We use the model to study how management policies affect the achievable rate of growth. The issues discussed are relevant to any service based company facing the problem of maintaining and improving service quality against the pressure of performance expectations set by shareholders or corporate owners. We also briefly discuss the benefit of using such models for executive training. The model will be used as a basis for educating the emerging generation of managers who will have to cope with the tensions described in the model.     

Scenario inspections

Scenarios help practitioners to better understand the requirements of a software system as well as its interface with the environment. However, despite their widespread use both by object-oriented development teams and human–computer interface designers, scenarios are being built in a very ad-hoc way. Departing from the requirements engineering viewpoint, this article shows how inspections help software developers to better manage the production of scenarios. We used Fagans inspections as the main paradigm in the design of our proposed process. The process was applied to case studies and data were collected regarding the types of problems as well as the effort to find them.

Intelligent decision support for effectively evaluating and selecting ships under uncertainty in marine transportation

This paper presents an intelligent decision support system for evaluating and selecting specific ships under uncertainty. A task-oriented procedure is developed for determining the relative importance of the evaluation and selection criteria with respect to a specific shipping task. A fuzzy multicriteria analysis algorithm is developed for determining the overall performance of each ship across all the selection criteria and their associated sub-criteria. An intelligent decision support system capable of integrating the developments above is proposed for facilitating the ship evaluation and selection process. An example is presented to demonstrate the effectiveness of the proposed intelligent decision support system.     

Designing robust structures - A nonlinear simulation based approach

This paper presents a generally applicable numerical procedure for designing robust structures under uncertainty, which can be coupled with any arbitrary nonlinear computational model for statical or dynamic structural analysis. Based on the results from an uncertain structural analysis several permissible design domains are determined with the aid of cluster analysis methods instead of traditionally computing only one particular set of crisp design parameter values; these represent design alternatives. To identify a preference solution, a discrete three-criteria optimization problem is formulated, which is focused on maximum structural robustness and includes a safety component. A measure for the global robustness of the design alternatives is introduced based on an analog to Shannon’s entropy. The goal of the resulting design is that the structural behavior is only marginally affected by uncertainty and by changes in the design parameters, which further provides comfortable decision margins to the construction engineer.The proposed procedure is demonstrated by means of a numerical example and of an example from engineering practice in vehicle crashworthiness design.     

不确定性多属性决策中的ER方法改进

基于对不确定性多属性决策问题中ER方法的研究,提出一种不确定性多属性决策中的改进ER方法,井证明该方法完全满足证据合成的4个公理．通过实例运算,进一步验证了新方法的有效性和合理性．