Uncertainty Modeling in Knowledge Engineering and Decision Making

The following sections are included:

• FOREWORD

• CONTENTS

The following sections are included:

• FUZZY RELATIONAL CALCULUS AND ITS WIDE RANGE OF APPLICATIONS

• DECISION SUPPORT AND WARNING SYSTEMS FOR BUSINESS INTELLIGENCE

• A REVIEW OF DEVELOPMENTS FROM FUZZY RULE BASES TO FUZZY FUNCTIONS

• CAN FUZZY LOGIC FORMALISM BRING COMPLEX ENVIRONMENTAL ISSUES INTO FOCUS?

In any decision problem where it is necessary to obtain a ranking to know the decision matrix, the weights of the criteria and the method for evaluating the alternatives must be determined. In the present paper the TOPSIS method is the model employed to achieve this because we have linguistic and numerical values such as the assessment of alternatives for each criterion. The weights of the criteria have been obtained by an OWA operator. This theoretical development has been applied to a decision problem in the field of renewable energy.

Natural and human-made disasters cause humanitarian crises all around the world. When help is requested from international instances, Humanitarian logistics play an important role. Mathematical models can help to make strategic and tactical decisions in this context, when decisions must be taken in a multicriteria environment. In this work a multicriteria dynamic flow model is developed, allowing a more detailed planning of aid distribution operations than other existing static models.

This study suggests a future oriented probabilistic approach to the positioning analysis of the videogame console industry. In comparison to the common multivariate positioning techniques, the proposed approach, using Bayesian networks, is less restrictive and attempts to incorporate the potential influences of external factors as well as marketing strategies into the positioning analysis. This provides a better understanding of how the market and its complex environment may evolve in the future and how they can be influenced in favor of one's own product/brand.

The need of organizations to evaluate their environmental practices is increasing, but there are many possibilities to do it. In this contribution an aggregation procedure is proposed to evaluate company's environmental practices. This proposal integrates quantitative and qualitative information. Since the assumption of independence among criteria is rarely verified in an environmental context, the Choquet integral is proposed as aggregation operator. The final aim is to compute partial and global indicators that can be used by the management team to make their decisions regarding the environmental issues.

According to the gap model, service quality is the gap between user's expectations and perceptions. When experiences exceed expectations, the quality of the service is high, and vice versa. Following that idea, we present a fuzzy linguistic quality evaluation model based on the LibQUAL₊ methodology to evaluate the quality of digital libraries according to user's opinions. To derive the gaps essential for measuring perceptions of service quality, users are asked to establish their judgments across three scales for each digital service of the digital library. It allows the identification of digital services in which service levels should be improved and the identification of digital services satisfied outstandingly by the digital library.

Recommender systems (RS) have been successfully used in different personalization issues. One of the most interesting applications is tourism (hotels, restaurants, monuments, etc.) where users can obtain personalized recommendations according to their profiles. Recently, the increasing use of mobile devices drives RSs to a new trend based on context-awareness. In this contribution we take advantage of mobile devices in RSs and propose a location-aware RS for tourism that provides recommendations to users not only based on their profile but also taking into account their current location.

Integral performance appraisal process is used as a key tool for enterprise competitiveness. It considers different indicators to provide evaluation assessments about employees' performance based on the judgment of different groups of reviewers, who socialize with employees. This contribution presents a Web based Support System for Integral Performance Appraisal, so-called WSSIPA, which implements an integral performance appraisal model that offers a heterogeneous framework in which reviewers can provide their judgments within different domains (numerical, interval-valued and linguistic), according to the nature of criteria and the background of each reviewer. In this contribution, we show the functionality of WSIIPA, conducting a simple case study for an integral performance appraisal process.

Measuring consensus level for a set of preferences requires a proper distance defined on the considered domain. We focus on preference-approvals which are extensions of ordinal preferences by the approval information. For any given set of alternatives, a preference-approval is defined by a weak ordering of the alternative set, subsets of approved and disapproved alternatives and a consistency condition. We propose a method of using weighted distances for ordering and approval components. A consensus measure based on this distance is provided which is sensitive to the position of the disagreements on ordering and approval.

Recently, it has been presented the concept of Hesitant Fuzzy Linguistic Term Sets (HFLTS) to manage hesitant situations under qualitative settings in which experts hesitate among different linguistic terms to express their preferences or assessments. It was also defined the concept of envelope for an HFLTS to carry out the computational processes with them. However, this envelope does not keep the fuzzy representation because it is represented by a symbolic linguistic interval. Therefore, in this contribution we propose a new envelope for HFLTS based on Choquet integral that keeps the fuzzy representation in the computational processes with HFLTS.

In this paper we suggest using the fuzzy DEMATEL method to model the energy service contracting market in Turkey and identify factors playing a significant role in its development. The proposed model provides a causative representation of the complex market structure in an uncertain environment. Results are compared to an earlier study applying a fuzzy cognitive mapping approach to explore the critical factors influencing the energy service market in Turkey.

We study reliability centered proactive maintenance of a complex dynamic system consisting of aging components. System reliability can be estimated from the interactions of the components. Maintenance of the system is realized by replacing components in any period. Our aim is to minimize total replacement cost in a discrete horizon such that system reliability never falls below a predetermined threshold value. We represent the problem with dynamic Bayesian networks (DBNs) and develop an algorithm for determining when and what to replace within the DBN framework. We prove that under the existence of a predetermined threshold, this algorithm assures optimum replacement times. Four approaches are proposed to choose the component to replace and are tested on a complex dynamic problem.

One of the most important problems of project management is the decision of task durations in the projects. The duration of each task and cumulative durations of the successive tasks generally results in the success or fail of that project. While the common practice is to use natural numbers for task duration, they generally fail to match the real world results. In order to obtain more realistic results, researchers used fuzzy numbers which resulted in an important improvement of estimations but with the vagueness and uncertainty of the business world, they still require improvement. In our study, we used fuzzy numbers with AHP to determine task durations and adapt them to real world, first we separately find the levels of vagueness and uncertainty of tasks using AHP and then we use these categories to manipulate our fuzzy membership function for better prediction of tasks' durations and certainty levels so we can provide a more accurate project plan with more realistic task durations.

The elicitation of preference information in multi-criteria decision analysis (MCDA) processes and the lack of practical means supporting it is a significant problem in real-life applications of MCDA. The issues at hand are problematical in a multitude of ways, but some of these issues may be remedied by accepting weaker input statements from decision-makers than what is most commonly needed, yet being able to utilize these statements for decision evaluation. In this paper, we propose a fast and practically useful weight elicitation method, which builds on the ideas of rank-order methods, but in addition take imprecise cardinal information into account.

Sales forecasting has a great impact on facilitating effective and efficient management of resources for construction projects. Since, having appropriate sales prediction provides appropriate financial activities, especially for large scaled construction projects. This study proposes a numerical model for construction projects which integrates Fuzzy Rule Based System (FRBS) with Net Present Value (NPV) analysis. In the proposed hybrid model, FRBS is used for sales forecasting and provides input for NPV formulation. The proposed model is applied to a construction project in Istanbul so as to demonstrate its performance and practicality for real life problems.

In this study, synthetic data with 100, 1000 and 2000 records have been produced to reflect the probabilities on the ALARM network. In this study, a medical diagnosis system called as DRCAD is presented. DRCAD system is more innovative and interesting than the classical diagnosis support systems. In other words, DRCAD collects possible diagnosis of the patients from two sub modules. Each of these two sub modules gives the possible diagnosis from symptoms in a specific confidence degree. The proposal of two sub modules are combined linearly and diagnosis decisions are presented as a list. Each of the sub modules consist of Bayesian inference and rule-based inference models respectively. As a result, the methods in which the conclusions are combined in a linear manner are 5% more successful than the “Rule Based Method” when applied individually and 30% more successful than the cases where the “Bayesian Network Based Method” is utilized.

Non-Destructive Testing (NDT) techniques are affected by concrete properties such as porosity, water content, strength, etc. Extracting one concrete property from one NDT measurement may lead to considerable uncertainties. This highlights the benefit of NDT data fusion to evaluate accurately concrete properties.

In this paper, NDT measurements from several NDT techniques were combined to predict more accurately concrete properties such as porosity and water saturation. Two techniques of data fusion were used namely neuro-fuzzy network and theory of possibility. The results obtained show the effectiveness of the statistical modelling to predict the properties of concretes by fusion of NDT measurements.

In this paper, a multi-period decision problem with partially known information is considered. In each stage, a decision maker has one and only one chance to make a decision. The optimal decision in each stage is obtained based on the one-shot decision theory. That is, the decision maker chooses one of states of nature (scenario) of each alternative in every stage with considering the satisfaction of the outcome and its possibility. The selected state of nature is called the focus point. Based on the focus points, the decision maker determines the optimal alternative in each stage by dynamic programming problems.

We propose a new method for ranking alternatives in multicriteria decision-making problems when there is imprecision concerning the alternative performances, component utility functions and weights. We assume decision maker's preferences are represented by an additive multiattribute utility function, in which weights can be modeled by independent normal variables, fuzzy numbers, value intervals or by an ordinal relation. The approaches are based on dominance measures or exploring the weight space in order to describe which ratings would make each alternative the preferred one. On the one hand, the approaches based on dominance measures compute the minimum utility difference among pairs of alternatives. Then, they compute a measure by which to rank the alternatives. On the other hand, the approaches based on exploring the weight space compute confidence factors describing the reliability of the analysis. These methods are compared using Monte Carlo simulation.

The aim of this paper is to develop a fuzzy TOPSIS method with linguistic evaluations by using α–cuts. In this paper, the rating of each alternative and the weight of each criterion are described by linguistic terms which can be expressed in α–cuts. A numerical example is presented to highlight the procedure of the proposed method.

An interactive population-based algorithm called EVALIMCO is presented in the paper. It is designed to solve multi-criteria convex integer optimization problems. A heuristic procedure is used to accelerate the search process. In this way the algorithm performs faster than the usual population-based algorithms. The performance of the algorithm is demonstrated on an illustrative example.

As we search to achieve sustainable development and substance, energy security and efficiency should be viewed not only through the perspective of addressing short-term challenges, but also a necessity for long-term growth of the economy. In this paper, PROMETHEE and Electre approaches are suggested for the selection among energy policies. Also, the methodology is based on these two methods via comparison. In the application of the proposed methodologies, the best energy policy is launched into determining for Turkey.

Energy has a vital role in the development of countries. The situation of global increasing demand for energy, lack of domestic resources, energy prices, etc. force the policy makers to plan energy policies in detail. In this paper, the strategic energy situation of Turkey is analyzed. This paper presents a hybrid model that combines strength, weakness, opportunity and threat (SWOT) analysis with the fuzzy analytic network process (FANP) method. In the evaluation process, it is occasionally impossible to take precise information from decision makers (DMs) who can say their assessments relatively and approximately. It is also needed to use fuzzy logic to alleviate this problem. Since the evaluation and selection of alternative energy policies is a complex multicriteria decision making (MCDM) problem, one of the most popular MCDM methods, ANP, is used under fuzziness in this paper. For this aim, four factors, twenty one sub-factors are determined and seven alternative policies are evaluated. It is computed that the most important three sub-factors are “major companies' interest for offshore drilling in Mediterranean and Black Sea”, “developing economy of the country” and “rapidly increasing energy demand”, respectively. As a result, Turkish energy policy makers are advised to focus on the privatization of electric transmission and distribution facilities, integrating the national electric system with European transmission systems and turning the country into an energy hub and terminal to be able to get the rapid economic growth and mitigate future energy problems.

Many companies operating in construction sector have the projects resulted with failure because of making unsuccessful decisions on project selection. The goal of this study is to propose a decision making approach for project selection by taking strategic goals and organizational expectations into consideration for the firms operating in construction sector. In this study, project selection criteria have been determined by discussing with the firms producing similar type of houses and by considering the related literature. Then a real application has been achieved in a large scale construction company. In the proposed model, a fuzzy technique for order preference by similarity to ideal solution (TOPSIS) method is used for evaluating and prioritizing the candidate construction projects. Consequently, it is determined that proposed model gives successful results and may help decision makers choose the best alternative projects.

In this study, a location selection methodology using an integrated multi-criteria decision making approach for a logistics network model design in the presence of health emergency situations for Turkey is proposed. It is an important issue to urgent response to such unpredictable disasters. The proposed methodology tries to select the locations of temporary emergency centers for affected areas while considering eleven decision making criteria. Decision Making Trial and Evaluation Laboratory (DEMATEL) and Analytic Network Process (ANP) methods are integrated to determine the convenient locations considering the relations between criteria and alternative locations.

In this study, the best location for a bus garage, in which maintenance and repair activities are operated, is tried to determine for public transportation system in Istanbul. An integrated multi-criteria decision making technique is used to obtain reliable results. Firstly, various criteria related with garage location selection are specified and weighted by fuzzy AHP. Then, these weights are used in fuzzy axiomatic design technique to determine the precedencies of the alternative garage locations.

Today's companies, perform a work which aims to enhance individual and corporate efficiency and effectiveness. Companies need to provide establish and provide valid and applicable performance management system in order to become successful on sectoral. In a competitive environment, companies' staff performance evaluation as indicate retrospective performance level, identify prudential potential performance and provide new perspective to studies about enhance performance. The purpose of this study is to specify the recruitment criteria of position that clinical chief of surgery. Research is done on a group including 2 experts that have a career in health sector for many years. In this research, staff recruitment criteria, recruitment process evaluation and the ranking of importance for the corporation are involved for the non-profit organization. This study includes identifying the criteria for selection, some information about criteria are get from the experts which working at hospitals in Istanbul then all in information are combined to build hierarchy of recruitment criteria. By using Analytical Hierarchy Process (AHP) in fuzzy environment criteria are ranked and the importance is identified and this method is used to select appropriate candidate for the position. As a result, according to FAHP methods recruitment criteria and process evaluation for the surgeons are identified.

In this paper, a new method is presented to deal with fuzzy multiple-attribute group decision-making problems based on weights vector of decision makers. In this paper all evaluations that are given by decision makers are interval type-2 fuzzy trapezoidal numbers, since type-2 fuzzy sets involve more uncertainties than type-1 fuzzy sets. We also use a numerical example for location selection problem to utilize the proposed method.

The goal of this research is to present a novel group multi-criteria decision-making (GMCDM) method with alternative vote system (AVS). In the process of solving a GMCDM problem, it is one of most important issue to aggregate the ranking order of alternatives by each decision maker. The best alternative can be selected by minimizing the total unsatisfied points of members in decision group. By considering the importance of each criterion in GMCDM problem, the AVS is applied to minimize the total unsatisfied point of members in decision group and to determine the final ranking order of alternatives. A numerical example is implemented to illustrate the advantage of the proposed method. Finally, conclusions are discussed at the end of paper.

Financial indicators which have been monitoring and announcing to public periodically are indispensable tools particularly in the banking sector. In this study, it has been chosen 22 different Banks in Turkey. The 39 performance related criteria for those banks were evaluated by an integrated AHP approach and their weights were computed. Then, the grey relational degrees of 22 banks were computed by Grey Relational Analysis Method. On the other hand banks were ranked by using Financial Ratio Analysis. The obtained results from both methods were presented comparatively.

The nurse scheduling problems (NSP) deal with assignment of the nurse to the specific shifts in specific days is taking into account both the hospital's need and nurse preference. In the hospitals, nurses are serving under difficult circumstances. To improve the quality of service, work schedules should be prepared in a fair manner and has to provide the desired conditions. In this study, it is aimed at developing a model which can be used in real applications and can be rapidly reached to the solution for the nurse scheduling problem. For this purpose, the formulated problem was solved by using artificial bee colony (ABC) algorithm, which is one of the new meta-heuristic methods. It has been coded a program with a user interface using the Visual Studio programming language for running the model. The designed interface is able to take into account the different level of the demand and the number of staff. The model was tested on a real problem and found better solution than the current situation.

In this paper, a new greedy algorithm is proposed to solve the fuzzy multiobjective flow shop scheduling problem. We use the two approaches; the possibility measure and the area of intersection for multi objective fuzzy flow shop scheduling problem. The proposed new greedy algorithm is tested on the benchmark problems in the literature. The performance parameters of proposed greedy algorithm are determined by full factorial design of experiments (DOE). The performance of the proposed greedy algorithm is compared with the scatter search method.

In this study a priority weight generation method on the intuitionistic fuzzy preference relations is proposed. The method consists of two linear programming models which are generalized to the case of group decision making with the weight information defined by each DM. In the models, the collective preference relations are obtained by using the intuitionistic fuzzy weighted geometric averaging operator and additive consistency is considered. Finally, a numerical example is give to verify the validity and applicability of the models.

Some fuzzy multicriteria methods have recently been developed by using type-2 fuzzy sets. Fuzzy TOPSIS is one of these methods. While these need excessive arithmetic operations with respect to type-1's, they can handle the uncertainty in defining membership functions. We modified Buckley's fuzzy AHP method by using type-2 fuzzy sets.

Wind energy is an important energy source among the renewable energies, in nowadays. Turkey has many wind energy locations due to the appropriate weather conditions. The problem of location selection of wind energy depends on several criteria: wind speed, air density, surface roughness coefficient, and altitude. Therefore, the location selection of wind energy is a difficult multi-criteria decision making problem. In this study, this multi-criteria problem is solved by fuzzy data envelopment analysis (FDEA) with triangular fuzzy numbers (TFN).

This study presents an application of a fuzzy goal programming approach to the problem of end-of-life tyres (ELT). The proposed model attempts to find out the percentage of ELT that must be used in different recovery techniques. The model is solved by using LINDO computer package and yields an efficient compromise solution with the multiple fuzzy goal values.

Concerning the environmental parameters in product development is crucial. However the studies which focus on environmental product developmentput mostly aside cost and quality parameters. In this study, a multi aspect QFD for Environment (QFDE) methodis proposed. In order to define the priorities of the stakeholders' requirements, Fuzzy Analytic Hierarchy Process (FAHP) Extent Analysis technique is used. The method is applied for the product “hand blender”.

This paper presents fuzzy multi attribute valuation of new software projects in banking industry by balanced score card based on the fuzzy TOPSIS model which includes fuzzy real option. Balanced scorecard method is used to determine the decision criteria by applying survey to managers. The economic side of the evaluation is dynamic and risky in nature in software development projects. Fuzzy real options method is used for surmounting these risky situations. Hence the evaluation of the software development projects has many attributes to investigate; fuzzy TOPSIS method based fuzzy real options method is offered in this paper. There are three software projects to be evaluated and investment ranking will be made according to the result of the method offered in this paper.

In any organization there are some main goals and lots of projects for achieving these goals. For any organization, it is important to determine how much these projects affect on achieving the main goals. This paper proposes a new fuzzy multiple attribute-based decision support system (DSS) for evaluating projects in promoting the goals as such a selection may involve both quantitative and qualitative assessment attributes. In addition the proposed DSS has ability to choose the most appropriate fuzzy ranking method for solving given MADM problem. Also it contains sensitivity analysis system which provides opportunity for analyzing the impacts of attributes' weights and project s' performance on achieving organizations' goals, and assess the reliability of the decision making process. The proposed DSS can be applied for solving every FMADM problem which needs to rank some alternatives according to some attributes.

Interest rate is one of the most used tools to determine the time value of money in investment projects. In this paper fuzzy regression analysis method is used to determine the relation among interest rate, inflation and the increases on export and import. The interest rate is formulated by using past data of interest rates, inflation, export and import sizes in Turkey. An illustrative example is given.

The performance evaluation of firms determines capabilities of firms to compete in the sector and has a critical importance for the development of the sector. In this study, a fuzzy multi-criteria decision model is proposed to evaluate the performances of retail firms. The eight Turkish retail firms are analyzed and firms are evaluated in terms of ten different financial ratios in six separate groups. Fuzzy Analytic Hierarchy Process (FAHP), Preference Ranking Organization Method for Enrichment Evaluation (PROMETHEE), Elimination Et Choix Traduisant la Realité (ELECTRE) and Multicriteria Optimization and Compromise Solution (VIKOR) methods are integrated in the proposed model. A hierarchical model of financial performance evaluation is structured by the decision makers. FAHP is used to determine the weights of the criteria. These criteria weights and financial ratio values are input to the PROMETHEE, ELECTRE and VIKOR method to rank the firms. The rankings of methods are compared to each other's in the conclusion.

Selecting the best conceptual aircraft design is a crucial problem during an aircraft development process. To find out a solution for this complex problem, there is only few information available with uncertainty as obtaining more precise measurements and accurate preference information is difficult or sometimes impossible. Therefore, this study proposes a realistic simulation approach based on Fuzzy Axiomatic Design (FAD) method and a novel Random Weight Generation (RWG) algorithm to select an aircraft design that meets the requirements the most. This approach gives an opportunity to the decision-makers to evaluate alternative designs under multiple qualitative and quantitative criteria by considering the operational requirements. Finally, an empirical study is presented to illustrate the applicability of the proposed approach.

Production and consumption relationship shows that marketing plays an important role in enterprises. In the competitive market, it is very important to be able to sell rather than produce. Nowadays, marketing is customer-oriented and aims to meet the needs and expectations of customers to increase their satisfaction. While creating a marketing strategy, an enterprise must consider many factors. Which is why, the process can and should be considered as a multi-criteria decision making (MCDM) case. In this study, marketing strategies and marketing decisions in the new-product-development process has been analyzed in a macro level. To deal quantitatively with imprecision or uncertainty, fuzzy sets theory has been used throughout the analysis.

The Galata district, which is one of the first settlements in Istanbul, has lost its unique architecture and urban texture due to radical changes after the 1950s. Therefore, a need of urban regeneration has emerged. In Istanbul, the concept of ‘urban regeneration’ has gained ground since the early 2000s. In this study, while considering five main criteria, five different urban regeneration strategies for the Galata district are prioritized via the VIKOR method. The weights of criteria are determined according to the judgements of multiple decision-makers.

The aim of the present study is to evaluate the weights of the criteria in green supply chain management to aid the supplier selection process in automotive industry. The problem was considered as a multi-criteria decision making problem. The main and the sub criteria affecting green supply chain management are taken into account in this paper, and their weights are evaluated through fuzzy ANP in order to determine their importance on automotive industry.

Pay structure is the ordering of pay rates for jobs or groups of jobs within an organization and plays an important role in the employment relationships affecting the efficiency of an organization. Design of a pay structure is a complex process requiring the evaluation many factors including both quantitative and qualitative data. Grey Relations Analysis (GRA) a useful mathematical tool for dealing with uncertainty and system analysis with limited information is introduced for creating a pay structure. Fuzzy Analytical Hierarchy Process (FAHP) is also used to determine the factor weights of payment types by pairwise comparisons. The hypothetical case study shows the applicability of the model.

The study developed an integrated approach for facility location problem based on a real life case from the Cement Sector in Turkey. A fuzzy multi-criteria decision making approach integrating Fuzzy AHP and Fuzzy VIKOR was designed in the study by concerning the vague and complicated nature of the information. The application was consisted of two stages: the region selection and site selection.

In this paper, we provide a decision support system for a two-echelon spare parts inventory distribution system that consists of a single central warehouse and a local warehouse. These warehouses are organized as separate entities that make their individual spare parts inventory provisioning decisions independently. The goal is to minimize the total backorders seen by the local warehouse such that the total inventory investment does not exceed a given budget. We provide a decision support system that can be used to coordinate these warehouses in a way to reach this goal. We used a real-life case data obtained from an original equipment manufacturer located in Europe and provided numerical results.

Purchasing activities of a company form a very important part in the operation of the company. The quality of production and services mainly rely on the performance of the suppliers. In addition, a large amount of the product's total cost belongs to the purchasing cost of its materials. Therefore, selecting the most suitable suppliers is very important for the firm. Unfortunately, most of the time, suppliers cannot provide the highest quality parts on-time at the lowest costs in the market. In other words, quality, delivery and cost objectives conflict with each other. In this study a fuzzy analytic hierarchy process approach is employed for the supplier selection problem of a textile company. First the selection criteria according to company's objectives are determined. Then the pair-wise comparisons are made on a fuzzy basis. Similarly, in the next step, the six alternative suppliers are compared by fuzzy means. Finally, the two suppliers which achieved the highest performance are advised to the company.

The word “modeling” comes from the Latin word “modellus”. It describes a typical human way of coping with the reality. So, a model will be a simplified version of something that is real. Before to the study of the typical methods of Fuzzy Logic we expose briefly which have been the historical origins of so forceful tool of the current science and technology. Because this search of their origins and evolution not only possess an intrinsic interest of mere cultural character, it can allow us to open new perspectives to know where it can advance in the future. Fuzzy Modeling will be defined as the construction of the Fuzzy Inference process. We analyze here this process, very useful for many interesting applications in modeling and optimization problems.

Interaction with rapid development of information technology and progress in modern production management science has emerged. In this respect, ERP (Enterprise Resource Planning) and Lean Manufacturing applications are important not only for big companies but also for Small and Medium Size Enterprises (SMEs) to become more competitive. The aim of this study is to analyze the maturity levels of Turkish SMEs on ERP systems and Lean Manufacturing applications. A questionnaire is prepared and applied to White goods enterprises (mostly subcontractors). The derived correlation between these two main function areas is also used to design such new systems. Additionally, ERP and Lean Manufacturing concepts are analyzed together in questionnaire which is applied by regular visits to facilities. Hence, related performance model also gives information about effects of each performance on one another.

In this study, single objective profit maximization problem of a telecommunication intermediary is analyzed under uncertainty. In telecom market, various uncertainties exist in the related cost and pro t margins, which can affect the optimization processes and the decision schemes generated. Therefore, the fuzzy set theory is employed in order to handle the non-deterministic problem parameters rising due to the presence of vagueness and imprecision. Two different pricing policies namely all-you-can send and pay-perbyte are implemented into novel possibilistic mathematical programming model. As solution approach, two commonly used different solution methodologies are applied to resulting integer programming model with fuzzy objective function coefficients. Finally, performances of suggested methods are tested on several randomly generated scenarios. The results obtained indicate that both approaches provide almost identical solutions.

Quality Function Deployment (QFD) is an effective methodology to use product research and development (R&D) and it has lots of applications in the literature. However, none of the presented studies takes into account the dependencies among technical characteristics. Therefore a modified QFD approach by integrating DEMATEL technique into QFD has been proposed to fill in the gap in the literature. To illustrate the proposed algorithm, a numerical example has been presented in the scope of the paper.

Recommender systems are effective approaches to implement personalised e-services. In recent years, they have gained widespread applications in e-commerce. Current recommender systems still need, however, further improvements with respect to the accuracy of prediction and to solve the scalability problem. To this end, an incremental collaborative filtering (InCF) algorithm based on the Mahalanobis distance is presented for recommender systems. Furthermore, the Mahalanobis radial basis function with ellipsoidal shape is employed to determine the decision boundaries of clusters. Experimental results show that the algorithm proposed can lead to improved prediction accuracy and that it turns out to be scalable in recommender applications.

Earthquake early warning (EEW) is very important to earthquake mitigation, but EEW information before an earthquake is usually ignored. China Public Safety Early Warning Information Systems (CPSEWIS) including EEW information before, amid and after an earthquake is introduced. More important is that a method is designed to judge credibility of EEW information for the CPSEWIS to obtain useful EEW information and help people mitigate.

Virtual reality have been used to provide training systems in several areas, in particular in medicine. In that area, user's interactions in a virtual environment are modeled and they are compared with predefined classes of performance to know how much users are prepared to perform that procedure on human beings. In this paper, it is proposed a new approach for online Single User's Assessment System (SUAS) using Possibility and Necessity measures. Those fuzzy measures provide a interval estimator for the compatibility between an user's interactions and previous stored classes of performance and it is the kernel of a decision support system.

The discretization of continuous value is important to machine learning and data mining. In this paper, based on novel class-attribute interdependency criterion, a discretization algorithm (called NCAIC) is proposed, which considers data distribution and the interdependency between all the classes and attributes, and adopts upper approximations as an part of the discretization criterion. In NCAIC algorithm, the class-attribute mutual information is adopted which can automatically control and adjust the extent of the discretization of continuous value, ensure discretization of real value attributes reasonable.

This article describes the task of minimum cost flow finding in the transportation network in fuzzy conditions. Technique for solving the task of minimum cost flow determining in the graph with upper and lower fuzzy flow bounds is examined. This task is observed in fuzzy conditions according to the criteria of real life. The numerical example is presented.

In this paper a set of rules of Answer Set Programming is presented to represent and reason about points-to information in program written in object-oriented language like Java. Especially, path identification problems such as forward and backward path problems in context of aliasing are well solved with these rules. A simple Java program is given to show the effect of our method.

A heuristic is proposed for the Vehicle Routing Problem in this study. The heuristic has two phases that are named as clustering phase and routing phase, and also it has a local search. Firstly customers are assigned to vehicles with proposed multi objective linear programming and fuzzy approaches are used to solve this model in clustering phase. Secondly clusters are routed with TSP classical integer mathematical model in routing phase. Finally local search is used to improve the solution. Some problems from literature are solved and results are given.

Rank reversal is one of the major problems in all ranking techniques. The problems of testing and test results for rank reversals, are discussed for pairwise comparison based non-numerical rankings. The results indicate that some algorithms that use nonnumerical approach produce almost no rank reversals.

The heat transfer during boiling depends on a variety of factors (boiling stage, material parameters, geometrical parameters, state variables and flow conditions). All variations are based on empirical relationships gained from experimental data because there is still no comprehensive theory. With increasing wall superheat (Tw-Ts), the evaporation changes from convective boiling to nucleate boiling and then to film boiling. For each type of boiling, separate equations for the calculation of the heat transfer coefficient do exist. This paper presents possibilities to take account of the existing uncertainties in the dynamic simulation of boiling. For this reason a Takagi-Sugeno-fuzzy-model was developed which includes the fuzzy transitions between the boiling stages.

This paper focuses on the stochastic 2m+1 Point Estimated Method (PEM) for Economic/Emission Load Dispatch (EELD) in a power system considering both thermal and wind turbine units. In solving this problem, the goal is to find optimal generation output of units to serve load demand while both cost and emission are minimized simultaneously. To this end, a novel evolutionary algorithm using Improved Bat Algorithm (IBA) is employed. To enhance the convergence characteristics and avoid being trapped in local optima, a new mutation strategy is suggested. Also, an interactive fuzzy satisfying technique is used to deal with the multi-objective optimization problem. The effectiveness and superiority of the proposed framework is validated with numerical results.

In this paper, an efficient optimization method called Modified Firefly Algorithm (MFA) is proposed to handle complex nonlinear optimization problems. In order to enhance the exploratory ability of the algorithm, a new modification process is integrated sufficiently. The proposed method is examined on several benchmark optimization problems with different characteristics and results are compared to some of the other well-known algorithms in the area.

This paper considers the vehicle routing problem with soft time windows (VRPSTW) where the customers' preferences and tolerances on the latest delivery times are handled by use of a fuzzy programming model. An indirect enumeration strategy is proposed to generate pseudo-efficient (Pareto) solutions to the VRPSTW. The proposed model and the solution approach make easy to implement distributors' various strategies. A simulated annealing (SA) algorithm is implemented to solve the test problems. The importance and advantages of the proposed approach are illustrated by numerical results.

Green or environmental concerns have been drawn more and more attention both in academia and industry. Nowadays, few firms have all the skills and resources to produce complex products entirely in-house and supply chain management has become the dominant vehicle. Thus how to measure and enhance capability of green supply chain management (GSCM) plays an important role in developing an eco-friendly enterprise. The purpose of eco-friendly development planning is to unite the resources of an enterprise to create business value. For an enterprise to enhance GSCM, deploying and starting with green drivers of the business environment, as well as green supply chain (GSC) capabilities and providers, and ultimately transforming all these attributes into strategic competitive edges are all critical. However, the published literature on GSCM hitherto has failed to sufficiently address the relevant perspectives in such analyses. The relationship matrix in the quality function deployment (QFD) method provides an excellent tool for deploying important concepts and linking processes. Furthermore, due to the ill-defined and ambiguous relationships among GSC drivers, capability and provides, the conventional evaluation approaches not can handle such decisions suitably and effectively. This study suggests an operational strategy for deploying the GSC drivers, capabilities as well as providers by using the relationship matrix in the QFD method and fuzzy logic. An illustrative example is presented to demonstrate the applicability of the proposed method.

The understanding of human behaviour in the real world is important aspect of attempts to develop effective tools that simulate humans in areas such as ergonomics, manufacturing, transportation, psychology and architecture. In this research, video observation studies and analysis is applied to understand the individual human behaviour in crowded spaces. Individual human behaviour is modelled and simulated using ‘agents’ with differing characteristics and abilities in a virtual environment. The focus is on developing a simulation method that supports an ‘inclusive’ sustainable environment which allows full participation of the elderly and disabled people in society.

Uncertainties in the design process are investigated in this paper. A formal Bayesian method is presented for designers to quantify uncertainties in design process. The uncertainties are implemented in a decision support system that plays a key role in design of complex projects where a large and multidisciplinary team of engineers are involved. The proposed method produces the probability distribution function of the model score or uncertainty. Two example applications for design and empirical demonstrate application of the method.

An active set of agents is a unifying space being able to act as a “bridge” for transferring information, ideas and results between distinct types of uncertainties and different types of logic. An active set is a set of agents who independently deliver true or false values for a given proposition. Difference between an ordinary or fuzzy set and active set is that the ordinary set has passive elements with values of the attributes defined by an external agent, In the active set any element is an agent that internally defines the value of a given attribute for a passive element. The active set is a set of agents in epistemic modal logic. The difference between active set and modal logic is that all the agents (worlds) in the modal logic are separate but in active set are joined in the evaluation of the given proposition. The agents of the active set evaluate propositions includes the Hilbert multidimensional space where is possible to simulate quantum logic gate.

The problem of data recovery in multi-way arrays (i.e., tensors) arises in many fields such as image processing and computer vision, etc. In this paper we propose a scalable and fast algorithm for recovering a low-n-rank tensor with an unknown fraction of its entries being arbitrarily corrupted. In the new algorithm, the tensor recovery problem is formulated as a mixture convex multilinear RPCA optimization problem by minimizing a sum of the nuclear norm and the ℓ₁-norm. The problem is well-structured in both the objective function and constraints. We apply augmented Lagrange multiplier method which can make use of the good structure for efficiently solving this problem.

In this paper, a renewal-reward process which represents a fuzzy inventory model of type (s, S) with a discrete interference of chance is investigated. By utilizing the expression of the ergodic distribution of this process in terms of a fuzzy renewal function, asymptotic expansions for the α-cuts of this ergodic distribution is obtained when the interferences have a Weibull distribution with a fuzzy parameter.

Corporate Social Responsibility (CSR) has taken an increasingly important role in business. It's observed that the literature for CSR projects has more focus on some parts of the topic such as the effects of the CSR project to the financial performance of the company. In this study, the process chain model originally developed for business logistics is proposed to be adapted to CSR project life cycle for an overall success. This model is aimed to be a backbone model for determining all possible project risks throughout CSR project life cycle. The final objective of the model is to obtain an integrated view to a CSR project for maintaining its sustainability.

In this paper we consider the resource-constrained project scheduling problem with multiple execution modes for each activity. The objective function is the minimization of the project completion time. Heuristics based on priority rule are considered as initial solution procedures for this problem. The proposed Taboo search algorithm (TSA) is computationally compared, the results are analyzed and discussed, and some conclusions are given. Results obtained on six classes of test problems and comparison with other algorithms from the literature show that our algorithm gives better solutions.

It mainly includes the enterprise profit distribution, payment of wages and occupation opportunities and so on of enterprise interests' allocation. Because it relates to many types of stakeholders, such as the shareholders, managers, employees, government and banks, the allocation of business interests is a complex management. In order to achieve the purpose of the enterprise, a power model of business interests' allocation is presented in this paper. The goal function of this model is the power to maximize and the constraint is the total interests limited.

This study aims to provide a decision support tool to improve the quality of the service at an airport terminal. For this purpose, a fuzzy cognitive mapping based model is utilized. The fuzzy cognitive mapping method allows dynamic modeling of problem by considering the complex network structure of model, i.e. the effects of factors on the quality of terminal services. Simulations are performed to analyze the changes in the system.

We present a software implementation of the methods for solving linear programming problems under uncertainty from previous work. Uncertainties about constraint parameters can be expressed as intervals or trapezoidal possibility distributions. The software computes the solutions for the optimality criteria maximin and maximality. For maximality with possibility distributions, only an approximate solution is obtained.

A new C × K nearest neighbor algorithm with a new point of view is proposed. In this algorithm K neighbors from each of the classes are taken into account instead of the well-known K neighbor algorithm in which only the total number of neighbors are considered. After experiments with well-known classification datasets, we conclude that K-NN, weighted K-NN, and average linkage neighbors results are between the single-linkage and complete-linkage algorithms. After the evaluation of the average accuracy results, we realized that the best results are obtained for the values of K between 1 and 10. On the other hand, it is determined that using single-linkage strategy provides high values of the results at most of the times.

This study aims to analyze the automotive industry from competitiveness perspective using a novel cumulative belief degrees (CBD) approach. For this purpose, a mathematical model based on CBD is proposed to quantify the relations among the variables in a system. This model is used to analyze the Turkish Automotive Industry through scenario analysis.

In this paper we present a methodology for rigorous optimization of nonlinear programming problems in which the objective function can be represented using black box functions. The specific application is process design and operation in which the process is modeled using modular process simulators. These models consume large CPU time to converge, derivatives are not available, in some cases generate noise, and they are seen as black box models. Different techniques are available for replacing those models by simpler and computational inexpensive models. In this case a kriging metamodel is used to replace them. Kriging has high prediction accuracy and capabilities to estimate the prediction variance. Coupled to kriging the “Expected improvement” technique (EI) [3] is used to find the global solution for the kriging metamodel. The methodology is tested with two mathematical functions, and finally, applied to the optimization of the operation parameters of a gas turbine case study, which is simulated using AspenPlus.

There are several methods in the literature for finding the shortest path under uncertain constraints. In this study a fuzzy chance constrained model based on genetic algorithm is proposed for shortest path problem with fuzzy constraints. The fuzzy simulation algorithm which checks whether fuzzy total travel time is in time window is presented in detail. A numerical example is given, and solved by using proposed method.

In this paper we develop a method of proactive and reactive scheduling, proposed earlier by Janczura et al. [3] in which fuzzy task durations are used. We construct a tool to create a robust schedule that might also be useful in repairing a schedule during the project realization. One of the benefits of this approach is supporting a Project Manager in project monitoring in the planning and realization phase. In this paper we propose a modification of an algorithm used to defuzzyficate the fuzzy task durations. This approach is based on defining a project security level and allows to control the project risk on the whole project level and not only separate task levels. Moreover, it provides a tool to a build secure schedule during the planning phase. For determining the crisp representations of the activity times the non-linear programming algorithm is used with the criteria of minimization the project duration given a project security level. In this paper we propose a modification of a defuzzyfication algorithm to consider the whole information included in a given fuzzy number.

This paper reports the application of fuzzy set theory for water quality assessment of Wulihu Lake, China. Methodology with ranking feature value (RFV) which transforms fuzzy combination vector into assessment level is used to express the quality of water in the environment of monitored data. The results show that the proposed method gives the certainty water quality levels. The water quality level of Wulihu Lake grew better because of ecological projects. This study could provide a scientific basis for analyzing and evaluating the water quality for environment management.

In this paper, a single-period inventory problem with partially known demand information characterized by possibility distributions is analyzed. Decision models are proposed based on one-shot decision theory. Two choices are used for determining which state of nature (demand) should be considered for each order quantity. Based on the selected states of nature (focus point), the optimal order quantity is obtained. Different kinds of focus points (demands) lead to different results (optimal order quantities) which reflecting a decision maker's attitudes about possibility of demand and satisfaction of decision.

In today's severe competitive environment the selection of appropriate suppliers is a significantly important decision for effective supply chain management. In this study a multiple sourcing supplier selection problem is considered as a multiple objective linear programming problem with fuzzy demand level. Three objective functions are minimization of costs, maximization of quality and maximization on-time delivery respectively. In order to solve the problem, a two-phased additive approach is proposed. In the first phase, ideal solutions are utilized for finding aspiration levels for each objective. In the second phase Chen&Tsai's fuzzy model is adopted for the problem. The adopted model can be efficiently used to obtain a non-dominated solution. The approach is illustrated by a numerical example.

Reverse Logistics focuses on the backward flow of materials from customer to supplier (or alternate disposition) with the goals of maximizing value from the returned item or minimizing the total reverse logistics cost. The selection of best reverse logistics provider becomes the most important issue in the whole supply chain system. In this study, a multi-criteria decision making method in fuzzy environment is developed to guide the selection process of best Reverse Logistics Provider. The proposed decision-making methodology has been applied to a case of paper recycling industry in Tunisia where this material is reused by manufacturing companies.

A growing problem of the supply chain, the bullwhip effect is studied in this paper. In order to develop effective strategies to mitigate the bullwhip effect, we need to understand the factors that cause it and the interactions among them. This paper uses Fuzzy Cognitive Map to analyze the interaction among the causes of the bullwhip effect and determines the leading factors that are the root causes of this problem.

In this paper, we make an attempt to review literature in supply chain (SC) by uncertainty aspect. Using resource based approach [65], main uncertainty sources which are demand, process and supply of supply chain are taken into account for analysis. Uncertainties are given in detail grouping by main uncertainty sides of supply chain. Problem areas that uncertainties occur are classified. Used techniques to define and handle uncertainties in papers also are analyzed. Solving methods of problems which contain uncertainties in supply chain are showed. In conclusion part, suggestions for future researches are provided based on review.

Supply chain management (SCM) has been considered as one of the most important operations strategy for improving competitiveness of organizations in the last decades. While managing the supply chain (SC) effectively one of the essential problems is the uncertainties that should be treated properly. Fuzzy set theory allows for a conceptual and theoretical framework for dealing with uncertainties in SC problems. In this paper we investigate the trend in the SC articles that use fuzzy set theory through the years, and the focus of fuzzy set theory in SC modeling. We find that the number of related articles growing exponentially and it is mostly due to the rising interest of the fuzzy environment researchers to SCM. Additionally we find that the demand is the most studied type of uncertainty in fuzzy SC modeling.

Collaborative product development (CPD) can be defined as two or more firms joining resources for the same goal, where the first phase is the partner evaluation process. The choice of partner affects the performance of the whole CPD, and therefore it must be conducted carefully. This paper presents criteria set defined specifically for CPD partner evaluation and proposes an integrated fuzzy multi-criteria decision methodology for the selection of the right partner. The technique enables decision makers to evaluate criteria weights and the potential partners separately.

We define real extensions f : ℝⁿ ⊃ [0, 1]ⁿ ⟶ [0, 1] ⊂ ℝ of Boolean functions that satisfy Δf = 0 on the whole domain [0, 1]ⁿ, i.e. potential theory as first principle.

We introduce the Real Algebraic Normal Form RANF ∈ ℤ[x₁,…, x_n] of f, a generalization of the Algebraic Normal Form ANF of , and show how the RANF is related to the Disjunctive Normal Form DNF by the Reed-Muller Transform, or the Inclusion-Exclusion Principle.

This paper proposes logical aggregation based on interpolative Boolean algebra as an aggregation tool for problems of multi-expert decision making. Expert opinions are unlikely to be identical, and they are usually either close or conflicting to various degrees, or sometimes include different forms of interaction. In addition, there are many different types of decisions which fit the broad circumstances of multi-expert decision making and decision maker's point of view. Logical aggregation enables multi-expert aggregation based on linguistic requirements described by logical expressions. Logical expressions are uniquely transformed into corresponding generalized Boolean polynomials. Calculation of generalized Boolean polynomial values provides a decision maker with final alternative assessments by multiple experts.

Let L be a countable first-order language such that its set of constant symbols Const(L) is countable. We provide a complete infinitary propositional logic (formulas remain finite sequences of symbols, but we use inference rules with countably many premises) for description of C-valued L-structures, where C is an infinite subset of Const(L) The main goal is to provide a formal framework for reasoning about -valued evaluations of propositional formulas, where is some countable ordered field. The prime examples of are the field of rational numbers ℚ, its real closure and the field of fractions ℚ(ε), where ε is a positive infinitesimal.

In this paper fields of research are modular neural networks and consistent real-valued logic as a technique for aggregation of the modules. Aggregation is done using logical and pseudo-logical aggregation based on interpolative Boolean algebra (IBA). IBA is technically realized through generalized Boolean polynomials. The proposed technique expands the existing base of modular neural network models and shows respectable results in forecasting currency exchange rates

When making a decision on the basis of multiple criteria the traditional AHP method does not take into account the fact that they may be interrelated. It is proposed that the traditional AHP method could be extended by applying fuzzy logic for aggregating the criteria into a single objective function. The application of consistent fuzzy logic is preferred since no conventional fuzzy set theory is in the Boolean frame. The basis of this approach is the Interpolative realization of Boolean algebra (IBA). The proposed approach is illustrated on the web service selection problem.

Let Ω be a complete residuated lattice. Let t SetR(Ω) be the category of sets with similarity relations with values in Ω (called Ω -sets), which is an analogy of the category of classical sets with relations as morphisms. A cut system in an Ω -set (A,δ) in the category SetR(Ω) is a system (C_α)_α∈Ω where C_α are subsets in A×Ω with some special properties. It is well known that in the category SetR(Ω), there is a close relation between cut systems in an Ω-set on one hand and fuzzy sets in the same Ω-set, on the other hand. Moreover, there exists an extension procedure such that any system (C_α)_α of subsets from A×Ω can be extended onto a cut system In the paper we prove that the extension procedure is, in some sense, the best possible. This will be expressed by the theorem which states that the category of such subsets is a full reflective subcategory in the category of cut systems.

We investigate residuated lattices as structures with dualities. We formulated the Principle of Duality for residuated lattice which divides basic properties of this structure into two groups of dual ones.

Among the many desirable properties of fuzzy inference systems not all of them are known to co-exist. For instance, a system based on a monotone fuzzy rule base need not be monotonic and interpolative simultaneously. Recently, Štěpnička and De Baets have investigated and shown the co-existence of the above two properties in the case of a fuzzy relational inference systems and the single-input-single-output (SISO) rule bases. An extension of these results to the multiple-input-single-output (MISO) case is not straight-forward owing to the lack of a natural ordering in higher dimensions. In this work, we study the MISO case and show that similar results are available when the monotone rule base is modeled based on at-most and at-least modifiers.

In this paper, we give a new method for solving a system of fuzzy relation equations. First, we get a derived equation from a system of fuzzy relation equations. Then the relationship between the solution sets of a system of fuzzy relation equations and its corresponding derived equation is investigated, and some necessary and sufficient conditions for the solution sets of these equations to be equal are presented.

We propose a new image reconstruction technique which uses approximation properties of the fuzzy (F-)transform and its ability to remove noise. The proposed technique is based on the inverse F-transform and on a fusion which combines the inverse F-transform with the original image on a non-damaged area. We claim the new technique has less computation complexity than the usually applied technique of interpolation.

In this paper, we turn our attention to model theory of higher-order fuzzy logic (fuzzy type theory). This theory generalizes model theory of predicate logic but has some interesting specificities. We will introduce few basic concepts related to homomorphism, isomorphism, submodel, etc. and show some properties of them.

This paper introduces the concept of tolerable solution set for interval-valued fuzzy relational equations with min-implication composition on complete Brouwerian lattices. Tolerable solution set interval-valued fuzzy relational equations is shown when the righthand sides are meet-irreducible. Necessary and sufficient conditions for the existence of solutions are given.

We deal with the problem of human facial expression recognition. An image is preprocessed for feature extraction using moment descriptors; then, it is represented in the product lattice of Intervals' Numbers (INs). Both learning and generalization are pursued in by a Fuzzy Lattice Reasoning (FLR) classifier based on an inclusion measure function . Preliminary, experiments on a benchmark set have been promising.

The notion of a fuzzy coset of a fuzzy subgroup relative to a fuzzy subgroup is introduced, and some of their properties are investigated. In particular, it is shown that a normal fuzzy subgroup of a fuzzy subgroup can be characterized in terms of fuzzy cosets of a fuzzy subgroup relative to a fuzzy subgroup.

In this paper, we first introduce some important concepts in fuzzy finite automata (FFA). And then we define bifuzzy source and bifuzzy successor operators of FFA, as well as discuss their basic properties and their essential relationships. Afterwards, we study bifuzzy topologies derived from bifuzzy source and bifuzzy successor operators.

Formal concept analysis and rough set theory provide two different methods for data analysis and knowledge processing. Given a fuzzy context(U, A, ), one can get the fuzzy rough concept lattice L_δ(U, A, ), in Will's sense. Further, we apply the attribute reduction method for the concept lattices to the reduction of the fuzzy rough concept lattice. Finally, the experimental results show that the reduction method for the fuzzy rough concept lattice is effective in the reduction of the multiple rules and an important example is given.

In this paper, the n-inequality is studied in the lattice implication algebra. The necessary and sufficient conditions for existence of solution for the n-inequality will be presented respectively. In the case, that b has decomposition of irreducible finite union, the all minimal elements of solutions and detailed solution sets are given.

In this paper we want to establish the multiplication theory of quantum automata(orthomodular lattice-valued automata). According this theory, we kan construct complex quantum automata. First we give the definition of direct product, cascading product and ring product. Then we discuss the basic properties of them. At last, we give the form of the language that can be recognized by product quantum automata.

Similarity measure is an important part of the image matching methods. This paper introduces the medium mathematic system to image matching. It establishes the novel image medium similarity measure between two pixels and that of between two image sets based on the measure of medium truth degree. A new image matching algorithm that is governed by the medium similarity measure is discussed in this paper. The experimental results show that the proposed method is effective.

This paper proposes an integrated scheme for identifying segmented degraded numeral characters in gray images. The scheme consists of distinguishing foreground from background, wiping off mottles, cutting margins, calculating features, and recognizing numerals by the classifier established based on the measure of medium truth degree and several features selected by logistic regression. The experimental results show that the proposed approach performs well on recognition of degraded numeral characters.

In this paper, the uncertainty problem, including randomness and fuzziness as well as the medium mathematics background are introduced firstly. Then the two-phase uncertainty problem is described and two-phase degree (TPD), a measure method to the two-phase uncertainty, is introduced. Next, the SMS (Short Message Service) Spam in the mobile communication is studied as a two-phase uncertainty problem and a TPD of spam based algorithm is applied to make the spam filtering. The algorithm is combined the Naïve Bayes based algorithm with the measure to the n-dimensional distance ratio mean function. The experiment result indicates that the TPD based algorithm is better than the classical Bayes based one and has practical value in current technical conditions.

We continue the study of the residuated operations in the framework of hyperstructures. We focus on the case of a multilattice as underlying algebraic structure and introduce the notions of filter and deductive system. They differ from the analogous concepts in a pocrim due to the connection to congruence relations. Finally, we prove that the set of filters of a residuated multilattice is a complete lattice.

This paper focuses on α-multi ary semantic resolution automated reasoning method based on the general form of α-generalized resolution principle for lattice-valued logic. The definitions of the α-multi ary semantic resolution and α-multi ary semantic resolution deduction in LP(X) are given, the soundness and completeness of this ground case are gotten. This work will provide a theoretical foundation for the more efficient resolution based automated reasoning in lattice-valued logic

Femtocells widely deployed in a macrocell forms hierarchical cell networks, has been regarded as one of the most promising approach. However owing to the absence of coordination between the macro and femtocells, and among femtocells, decentralized spectrum allocation between macro and femtocell users become technically challenging. In this paper, an optimal spectrum allocation scheme based on Stackelberg game is proposed, in which femto and macrocell base stations are players, and the same spectrum is the resource players will choose to assign users to minimize the affected interference among each other. The comparison results show that the proposed scheme might be a solution for efficiently allocating the spectrum in hierarchical cell networks, as the improvement in terms of outage probability and spectrum efficiency had been achieved.

A first study of some properties of the fuzzy D-implications is presented in this paper. Compared to the S-, R-, and QL - implications, D-implications are less used and reported in the literature. This fact constitutes a particular motivation to investigate and analyze some of their properties by providing analytical proofs. Such analysis is of high importance since it constructs a concrete theoretical framework, which can be used in a latter step for applying the D-implications in real life applications.

This paper provides a systematic review on the research development of α-resolution automated reasoning theories, methods, algorithms and prototype in lattice-valued propositional logic and first-order logic based on lattice implication algebra (LIA), linguistic truth-valued lattice-valued propositional logic and first-order logic based on LIA, respectively. In addition, some important problems needed to be further investigated are proposed.

In this paper we propose a novel resolution method for theorem automated proving. General form of α-linear resolution method is investigated in lattice-valued propositional logic system LP(X) based on lattice implication algebra. Both soundness and weak completeness theorems are established. The contribution of this method is to provide a scientific and extended resolution method for automated reasoning based on latticevalued logic system. It also provides a foundation for constructing an α-linear resolution automated reasoning algorithm.

Great achievement have been obtained in the research on the truth degrees of formulae in propositional logic, but now the investigation on the truth degrees of formulae in predicate logic is very lack. Based on this situation, the n-truth degrees of formulae in two-valued predicate logic is proposed in this paper, and some important properties are studied. This paper enriches the content of the theory of quantitative logic, and provides a kind of framework for further approximate reasoning.

In the mid 1990's, the morphological associative memory (MAM) was introduced as a distributive associative memory model that performs certain morphological operations defined in the mathematical theory of minimax algebra. MAM models come in two different versions that are tolerant to different types of noise in the input patterns. To overcome this drawback, we resort to the more recent theory of mathematical morphology (MM) on inf-semilattices whose elementary operators are self-dual and we define an associative memory (AM) model in this framework.

Different variants of variable consistency dominance-based rough set approach (VCDRSA) have been proposed by considering the ordinality and inconsistency of information systems simultaneously. These approaches focus mainly on the criterion selection of rule generation, but rarely consider the effectiveness of decision results. In this paper, an extended dominance relation is induced into the variable-precision rough set (VPRS) and an extended variable consistency model under dominance-based rough sets is proposed. An illustration validates the feasibility of the proposed method.

Filters play a significant role in studying of structure and properties of partially ordered algebraic structures, in particular in the algebraic semantics of formal fuzzy logics and, moreover, it is also very important for the proof of the completeness of these logics. The aim of this paper is to introduce a kind of quasi-filters in lattice implication algebras, which are the algebraic abstractions of the sets of provable formulae and are closed with respect to lifting rules (These rules were defined by Pavelka in Zeitschr. f. math. Logik und Grundlagen d. Math. 25: 447-464, 1979) in the corresponding formal fuzzy logic, and we examine their mutual dependencies and the connections with other different kinds of filters in lattice implication algebras.

A kind of intuitionistic fuzzy logic system LP(S) based on linguistic truth-valued intuitionistic fuzzy algebra is proposed in this paper. The linguistic truth-valued intuitionisitic fuzzy algebra comes from the linguistic truth-valued implication algebra which is fit to express both comparable and incomparable information. The method can deal with the uncertain problem which has both positive evidence and negative evidence at the same time. Some axioms and reasoning rules of linguistic truth-valued intuitionistic fuzzy propositional logic are discussed. The proofs and theorems are also obtained.

This paper presents a fuzzy set construction method by transforming a membership multivalued representation into a scalar one. The method has two steps: firstly, a bifuzzy set is constructed using an aggregator operator from the family of averaging operators; secondly, the bifuzzy set obtained in the first step is transformed into a fuzzy one, preserving some indices, like the net truth or the proportional truth. Furthermore, the paper uses the method described above to create new formulae for entropy calculus of intuitionistic fuzzy sets and then, new formulae for color luminosity calculus.

This paper combines the methodology of belief rule-based (BRB) systems with nonlinear min-max optimisation techniques to develop a novel interval identification method, in which the outputs of an uncertain nonlinear system are represented by belief structures. On a finite set of measured data, the ℓ_∞-norm is used as the optimisation criterion of minimising the identification errors. This identification method is capable of modelling the optimal lower and upper bounds simultaneously, which can be used to approximate the interval output of an uncertain nonlinear system. The proposed method is tested with a family of uncertain nonlinear functions, and numerical results validate the functionality of the proposed identification method for uncertain nonlinear systems.

We present the concept of strong equality index, starting from the definition of strong inclusion given by Dubois and Prade in 1980. We also present a construction method based on the use of implication operators and two specific properties of the implications.

The notion of derivation in a lattice implication algebra is introduced, and some related properties are discussed. The properties of the isotone derivations are mainly researched. Using the concept of derivations as defined, a class of lattice implication algebras is constructed.

Throughout this note, a classical construction by Newman and Read, that arises in Utility Theory, is revisited and analyzed from a modern perspective, relating it to several key principles on representability of ordinal structures.

The article aims at discussing conditional Choquet expectations with respect to capacities. We introduce the concept of conditional Choquet expectation with respect to submodular capacity, and prove some properties. Moreover, we state that C_r inequality, Schwarz inequality, Hölder's inequality, Minkowski inequality and Jensen's inequality still hold for this kind of conditional Choquet expectations.

Ascribing causality amounts to detect causally related events. In this paper, we consider that these events are uncertain and expressed with the belief function formalism. They are either results of observations or interventions which are external actions forcing the variable to take a specific value. We show how the proposed concepts of strong accept/reject can be used, instead of changes in uncertainty, to discriminate between potential causes.

In this paper, we introduce a new hybrid preprocessing method for editing imbalanced data. The algorithm we propose first resamples the training data using the Synthetic Minority Oversampling Technique (SMOTE) method, and subsequently applies an editing technique based on fuzzy rough set theory to the balanced training set. We evaluate the performance of our algorithm in an experimental study, using the C4.5 classifier as the learning algorithm. Statistical tests show the superiority of our method over state-of-the-art resampling methods.

Geolocation applications and software are one of the fields where the configuration of the tools is a very complex task. An interesting challenge is to try to automatize it in proposing natural language interfaces able to interact and configure the low-level devices. This paper shows how such interfaces and fuzzy models can be mixed to facilitate the elicitation of business-level objectives in the geolocation context. Two 2- tuple linguistic representation models are considered and compared in a simple use case.

We present some new induced and heavy aggregation operators in a unified framework between the weighted average and the ordered weighted average. We introduce the induced heavy ordered weighted averaging weighted average operator. It is an aggregation operator that provides a parameterized family of aggregation operators between the minimum and the total operator. It considers complex reordering processes assessed with order inducing variables. We study some of its main properties and particular cases.

Forecast accuracy of economic and financial processes is a popular measure for quantifying the risk in multi-criteria decision making. In this paper, we develop forecasting models based on the statistical (stochastic) methods, sometimes called as hard computing, and on the soft methods using granular computing. To illustrate the forecasting performance of these approaches the learning aspects of RBF networks are presented and an application is included. We show a new approach of function estimation for nonlinear time series model by means of a granular fuzzy logic neural network based on Gaussian activation function with cloud concept. In a comparative study is shown that the presented approach is able to model and predict the high frequency data with reasonable accuracy and more efficient than the statistical methods.

This article discusses statistical, SVM models and fuzzy logic neural network as new managerial tools for finding the input/output function of sales processes for industrial companies. The study reviews experimental results, discusses, analytically and numerically demonstrates the quality and interpretability of the obtained results.

Instance-based Must-Link and Cannot-Link constraints are two forms of background knowledge famously used in semi-supervised clustering. In this work we demonstrate that it is possible to step-up the quality and quantity of constraint sets by adopting a method that combines automated and active constraint selection. This method capitalizes on the cluster feature that most of the data objects in the cluster are actually core points and just small parts are border points. Therefore, considering Must-Link constraints between core points and Cannot-Link constraints between border points adds more effective information because border points would separate each cluster clearly. Our experiments show that our approach provides a competitive edge in identifying informative constraints that would enhance the accuracy of clustering solutions.

Copulas have been frequently used in solving issues about modeling multivariate variables for the significant advantage that it can capture the multivariate dependence of the variables irrespective of their marginal distributions. The parametric families of Archimedean copulas are widely used in hydrologic frequency analysis due to their simple and closed form expressions. This paper mainly deals with the impacts of the length of hydrological data on the uncertainty of the modeling results of the Archimedean copulas. Several estimation methods are implemented including the inference from marginals (IFM) method, canonical maximum likelihood estimator (CMLE) and the estimation based on Kendall's tau. The results show that the length of data set will affect the choice of the best fitted copula and it is more reliable to use CMLE and estimation via Kendall' tau when the length of the data set changes.

Crowd creation is a branch of crowdsourcing that focus on creation of activities such as asking individuals to film TV commercials, perform language translation or solve challenging scientific problems. BurtonStory is a crowd creation project by Tim Burton that was run over twitter between November 20th and December 6th of 2010. The author started a new story with a single sentence, and the rest of the story is created by the crowd. The aim of this paper is to analyze participants based on their profile information using text mining approach.

Continuous change in technology and differentiation in product models in industrial market have an indispensable impact on forecasting demand for spare parts. Inventory managers periodically update their predictions of future demand rates for products. A Bayesian model, using a prior probability distribution for the demand rate, was used to obtain optimal inventory levels over several periods assuming a known cost for surplus and shortage. However, its performance has not been examined under various demand rates such as intermittent demand. Study examines such conditions with a research question.

A new decision making approach is presented where the available information is uncertain and can be assessed with interval numbers. We consider complex environments where the available information is formed by using different sources of information that have to be considered in the analysis. We introduce a new aggregation operator adapted for these situations that we call the uncertain generalized unified aggregation operator (UGUAO). It includes a wide range of classical aggregation operators in the same formulation.

Software cost estimation is one of the most important and complex tasks in software project management. As a result, several techniques for estimating development effort have been suggested. Fuzzy Analogy is one of these techniques suggested to estimate project effort when it is described either by linguistic or numerical values. Based on reasoning by analogy and fuzzy logic, this technique uses fuzzy representation of software project attributes by using expert knowledge or clustering techniques. From this work, we evaluate the accuracy of this approach to estimate the software effort using the International Software Benchmarking Standards Group (ISBSG) repository.

A solution of the ranking problem using kernels is described. As a consequence an elegant probabilistic ranking algorithm is obtained from the pocket algorithm. This removes a restriction from a former paper and simplifies proofs considerably. In addition results concerning class probabilities are exhibited. Some preliminary experimental results are reported and briefly discussed.

This paper proposes a probabilistic approach for energy management of MicroGrids (MGs) under uncertain environment. The proposed framework includes 2m + 1 point estimate method for covering the existing uncertainties in the MGs and a modified optimization algorithm based on the Gravitational Search Algorithm (GSA) to determine the economic assessment of MGs. This paper takes account of uncertainties in load demand, market prices and the available electrical power of wind farms and photovoltaic systems. The weibull and normal distributions are employed to model the input random variables. Moreover, the Gram-Charlier expansion is used to find an accurate distribution of the total operational cost of MGs for the next day-ahead. The effectiveness of the proposed method is validated on a typical grid-connected MG including energy storage and different power generating units.

As the effectiveness of complex networks mainly depend on probability distribution of time between failure and time to repair that involves system operations in different routes, all combinations of system operations are quantitatively described by various running and repairing states. In this paper, we describe a sampling method of random variable's probability distributions of failures and repairs time according to priority and event sequencing. Based on various states of system operations, a detailed simulation procedure has been established through identifying different failure and repair states. Finally, we have carried out the reliability and maintainability evaluation of system operations for a special complex line using Monte Carlo simulation as a case study. The simulation results demonstrate that the reliability and maintainability simulation algorithm is adaptable to measure the effectiveness of system service whose system operations follow to arbitrary failure and repair distributions.

Among existing Bayesian network (BN) parametrizations, conditional Gaussian are able to deal with discrete and continuous variables. Bayesian estimation of conditional Gaussian parameter needs to define several a priori parameters which are not easily understandable or interpretable for users. The approach we propose here is free from this priors definition. We use the Implicit estimation method which offers a substantial computational advantage for learning from observations without prior knowledge and thus provides a good alternative to Bayesian estimation when priors are missing. We illustrate the interest of such estimation method by first giving the Bayesian Expectation A Posteriori estimator (EAP) for conditional Gaussian parameters. We then describe the Implicit estimator for the same parameters. One experimental study is proposed in order to compare both approaches.

Frequently, the set of classes of a supervised classification problem presents an structure related to the specific features of each application context. However, standard classification models does not use to consider such an structure in their learning and reasoning processes. By means of the introduction of a bipolar approach, this paper proposes a revision of the basic notions of supervised classifiers, aimed to extend their generalization power and adaptation to problems with an structured set of classes.

Electroencephalograms (EEG) contain the information important for brain-computer interface (BCI), despite that their earlier use is limited by variations caused by background brain activity of operators as well as by artifacts. Thus achieving high accuracy of EEG classification is an important task for BCI systems. In this paper, we explore possible ways of improving the classification performance by using Group Method of Data Handling (GMDH). Specifically, we aim to find the most suitable selection criteria to be used in GMDH. The experimental results conducted on EEG representing the motor imagery indicates that the proposed criteria are capable of improving the classification performance as compared with Radial Basis networks.

Uncertainty in various domains implies the necessity for data mining techniques and algorithms that can handle uncertain datasets. Many studies on uncertain datasets have focused on modeling, query ranking, discovering frequent patterns, classification models, clustering, etc. However despite the existing need, not many studies have considered uncertainty in sequential data. This paper introduces UAprioriAll, a method to mine frequent sequences in the presence of uncertainty in transactions. UAprioriAll scales linearly in time relative to the size of the dataset.

A “consensus clustering” methodology is applied to long-term (1950-2010) Turkey's meteorological data (temperature, humidity and precipitation) in order to analyze the seasonal variations. The consensus clustering analysis applied is based on the methodology of disturbing the original data using resampling techniques, proposed by Monti et al [2]. We employed four different clustering algorithms (Agglomerative Nesting, Divisive Analysis, Partitioning Around Medoids and K-means) in order to form a consensus solution among different algorithms. Results indicated that Turkey is experiencing longer winter and summer, and rather short spring and fall seasons than usual.

The paper presents an original approach to the analysis of network data flows with the application of statistical spatial analysis methods. An illustrative example has been presented how to identify the correlations in network nodes behaviour and how to support the process of detection of network traffic related problems. The proposed method of data traffic analysis can be used to support various network management and security related tasks for example anomaly detection. The presented example shows how spatial correlations reflect the changes in communication patterns and so how they may be used for the detection of ongoing attacks, data link failures or network traffic routing problems.

Ben-Israel and Iyigun ([1] and [2]) presents a new clustering method which is probabilistic distance clustering (P-D Clustering). In this method, the probability of assignment to cluster for each point is inversely proportional to distances between data point and centers of clusters according to given number of clusters and their centers. In this paper, we study on new uncertainty measure for classification using the assignment probabilities of P-D Clustering. Moreover, the relationship of the new measure with Kullback – Liebner divergence is discussed.

Using a single chart to monitor simultaneously the process mean and variability if found would cut down the time and effort. Some researches have been done in finding such charts. Moreover, some researches have showed that the adaptive control charts could detect process shifts faster than the traditional Shewhart charts. A much easier loss chart with variable design parameters is proposed here to monitor both the deviation from the process target and the shift in process variability. Furthermore, a more efficient optimal variable design parameters loss chart is proposed and performs better than the loss chart with fixed design parameters and the joint and S charts.

In this study, student enrollments to universities of North Cyprus are examined from 2000 to 2009. Future values are forecasted using trend centrality of actual and forecasted enrollments that are obtained by using fuzzy time series.

Depending on decrease in natural resources and increase in perception of people on environment, governments have begun to attach much importance to legal regulations for protecting environment. In consequence of legal enforcements and customer perception, all actors in the business area are forced to operate recovery activities for getting value from used or scrapped products. In reverse logistics (RL) which includes the control of backward move of goods, for the actors it becomes hard to make scheduling regarding to amount and time of returned products. Because in the uncertain nature of RL, it is not easy to know how much products will return at a given time. Therefore, in this study a fuzzy expert system is proposed for forecasting return amount. An application at an electrical and electronic equipment recycling (e-recycling) facility is conducted for clarifying how the method is used in a real decision process.

In this study, rainfall-runoff modeling was carried out in Hajighoshan watershed using artificial neural network (ANN) and adaptive neuro-fuzzy interface system (ANFIS) with different inputs(current day rainfall; current rainfall and pervious day rainfall; current rainfall, pervious day rainfall and two previous day) methods. Root mean squared error (RMSE), mean absolute error (MAE) and correlation coefficient (R) statistics are employed to evaluate the performance of the ANNs and ANFIS models for forecasting runoff. Comparison of the obtained results reveals that the ANFIS model outperforms the ANN models. Based on the results of test stage, ANFIS with RMSE=7.11 m3 s-1, MAE=2.18m3 s-1 and R=0.60 is superior to rainfall-runoff modeling than the ANN with RMSE=6.03m3 s-1, MAE=1.97 m3 s-1 and R=0.39.

In this study, the pricing problem of a transportation service provider company is considered. Our goal is to find optimal prices by using probabilistic dynamic programming. A fuzzy rule-based expert system is used to identify the demand levels under different price levels and other characteristics of the journey. The results obtained by optimal price policies show that the revenue levels and the capacity utilization increase by applying dynamic pricing policy instead of fixed pricing. Thus, the diversification of price policies under different conditions is advantageous for the company.

This article presents a real case application of a neural network used in a production criterion modeling. The goal here is to determine the most adapted configuration of a neural network in a complete predictive model construction for the criterion. Since the evaluated criterion cannot be measured in real time and its influence on the performance of the production system could be important, a model is required to estimate the criterion value as correctly as possible. The analyzed criterion is supposed to be related to a set of independent variables. The employed neural network is a Radial Basis Function Network. Moreover metaheuristics have been used to increase the performance of the proposed method, by modifying the neural network composition.

In this work an alternative method of in-core sensor validation is outlined. Instead of using direct sensor readings, it is based on neuro-fuzzy modeling of residuals between the experimental values and their theoretical counterparts obtained from reactor core simulator calculations. Throughout the fuel cycle, the neural networks are subject to the incremental training on data prior to the current monitoring period. While for thermocouples the method shows very good sensitivity, for self powered neutron detectors the results are not satisfactory. The purpose of this paper is to present the newest results from considerably long time effort in development of this computational intelligence based approach.

Rough sets theory is an important mathematic tool for data mining. Feature selection is a main step in the course of data mining. In this paper, we focus on the attribute reduction by using rough set theory in Set-valued Ordered Information Systems (SOIS) from the view of information theory. Firstly, we introduce the information entropy and relative information quantity to SOIS. Then, we define attribute reduction by means of information entropy. Furthermore, an algorithm for computing reducts is proposed based on the information entropy. Finally, an illustrative example is employed to show the effectiveness of the proposed method for attribute reduction in SOIS.

The buffer allocation problem, i.e. how to distribute a certain amount of buffers among the buffer locations of a production line, is an NP-hard combinatorial optimization problem. In this study, a hybrid approach combining simulated annealing and genetic algorithms is presented to find optimal buffer sizes for unreliable production lines. The objective is to improve the capacity of production line by implementing proposed hybrid approach. To evaluate the capacity of the line, a detailed simulation model is developed. Numerical examples are given for evaluating the performance of the proposed heuristic, traditional genetic algorithm, and simulated annealing procedures. The results show that the proposed hybrid approach gives good solutions in reasonably short computation times.

Keyword suggestion methods are important to effectively match the keywords chosen by the advertisers and the queries in search engine advertising. Existing keyword suggestion methods can only suggest keywords with high complementary relevance based on concurrence analysis which cannot support competitive keyword advertising. This paper proposes a notion of substitute relevance as well as the corresponding definition of the substitute keyword. Moreover, a novel substitute keyword suggestion method named SKS is designed to help advertisers find more appropriately competitive keywords. Examples and experiments demonstrate the efficiency and effectiveness of the proposed method.

Situation awareness (SA) is a critical factor for human decision making and performance in dynamic environments. Actually SA is a mental model of the current state of the environment and includes many types of complex systems such as safety supervisory systems. The current paper employs two focus areas including neural network and expert system for maintaining SA in a safety supervisory system. The neural network components provide adaptive mechanisms for perception, and the expert system offers the ability to support comprehension and projection.

Making successful decisions on image classification can be influenced by the level of present noise. The noise on data can be reduced applying many classical and advanced techniques such as low pass filters or wavelets. However, in some cases the presence of the noise is not global but located only in some particular regions of the image. In these situations the application of global filters over the entire image is not a suitable option since the noise and the information are equally reduced. This article describes an intelligent approach to eliminate noise and avoid affecting the overall image.

Innovation is a competitive obligation for the SMEs but the needs vary by region. In this study, Turkish SMEs in Thrace region are clustered using Fuzzy C-Means which is a strong clustering tool for qualitative data, in order to compare the past and current innovational needs of SMEs. Similar questions are surveyed in 2007 and 2011 and the results are evaluated to emphasize the innovation based improvements.

This paper presents a new technique for model order reduction (MOR) of discrete systems with the advantage of selecting some of the desired frequencies. The proposed technique is achieved using the Artificial Neural Network (ANN) prediction. The ANN is implemented for predicting the unknown elements of the reduced order model while selecting the desired frequencies. Prediction of the ANN architecture is based on minimizing the cost FUNCTION obtained by the difference between the actual and desired system behaviour. The proposed ANN-based model order reduction method is compared to recently published work of MOR techniques. Simulation results verify the validity of the new MOR technique.

Artificial Neural Networks (ANN) has seen an explosion of interest over the last few years. Indeed, anywhere that there are problems of prediction, classification or control, neural networks are being introduced. Hence, the main objective of this paper is to develop a model to predict the response of the soil-structure interaction system without using the calculate code based on sophisticate numerical methods by the employment of a statistical approach based on an Artificial Neural Network model (ANN). In this study, a data base which relates the impedance functions to the geometrics characteristic of the foundation and the dynamic properties of the soil is implemented. This leads to develop a neural network model to predict impedances functions (all modes) of a rectangular surface foundation. Then the results are compared with unused data to check the ANN model's validity.

Biometrics is an emerging field, where technology enhances our ability to identify an individual. The identification by hand geometry is one of the most widely approved by the general public, it is very easy to use and less expensive than other techniques. Our work is based on the realization of a system that is capable for recognizing an individual based solely on the geometry of his hand, gpdsHAND database was used, where a preprocessing has been applied to extract the relevant characteristics of each image, these will be used in the identification phase. Two classification approaches are implemented, the SVMs (support vector machine) and neural networks PMC (Multilayer Perceptron). We then present our experiments and results and we will fix our perspectives.

A new rule-based representation scheme is proposed with belief degrees embedded in the entire consequent terms and in the entire antecedent terms of each rule, which is shown to be capable of capturing uncertainty and nonlinear causal relationships in an integrated way. The overall representation and inference framework offers a further improvement and extension of the recently developed belief rule base inference methodology (called RIMER). Subsequently, a simple but efficient method for automatically generating such extended belief rule-base from numerical data is proposed involving neither time-consuming iterative learning procedure nor complicated rule generation mechanisms, which is mainly attributed to the new features of the ex-tended rule base with belief structures. Then a case study in oil pipeline leak detection is provided to illustrate the proposed approaches.

In this work, we implement a multi-objective genetic algorithm (namely, non-dominated sorting genetic algorithm–II (NSGA-II)) to find the parameters of an ANN which provides in output the lower and upper bounds of the prediction intervals of a quantity of interest. We apply the proposed method on a synthetic case study of literature.

In this paper a new software monitor is presented. Computer systems must be adequately monitored to ensure stability control, reliability and risks to unexpected problems. Classic monitors are based on the analysis of some key variables which are in change during operation of the systems. In this new monitor, some modeling techniques (statistics, reliability and decision making) are used to provide an easy, automated way for the intervention into a system under risk.

The classical method of computing component importance measures (CIM), which are used for decision making on system reliability, is by means of fault tree (FT) analysis. However, the FT method has limitations for complex systems – characterized by the presence of various interrelated aspects like dependencies, uncertainties and knowledge of a reliability problem. The Bayesian Networks (BN) use graph-based representations that define probability distributions over network and over random variables in a concise and flexible way. In this paper, the computation of CIM using BN is investigated for complex railway systems.

This paper use LOGIT model to study financial early-warning by selected ST and normal companies. The in-sample and out-of-sample prediction rate are respectively 97.1% and 94.1%. With revision and re-judgment to logit model prediction result by efficiency indicators, the study result shows that in-sample and out-of-sample correction rate is respectively 95.8% and 96.2%.

This paper aims at integrating fuzzy analytic hierarchy process (FAHP) and data envelopment analysis (DEA) to evaluate relative efficiencies of 69 state universities in Turkey. Since most similar evaluations are multi-objection problems, which both FAHP and DEA are capable of solving, the integration of these two approaches will be shown to be even more powerful. The proposed integrated FAHP/DEA model can evaluate and rank different alternatives. On the first stage, we apply FAHP for determine weights of criterias. On the second stage, we rank alternatives by DEA of universities. A practical case study demonstrates the effectiveness of the proposed methodology.

One of the prominent features of standard and fuzzy data envelopment analysis (DEA) is the representation of each of the participating decision making units (DMUs) in the best possible light. This causes two problems; first, the different set of factor weights with large number of zeros and second a large number of linear programming models to solve. In this paper, we propose an efficient method to address these two problems. In proposed method by solving just one linear programming a Common Set of Weights (CSW) is achieved in fuzzy DEA. Since resulted efficiencies by the proposed CSW are interval numbers rather than crisp values, it is more informative for decision maker. The proposed model is applied to a numerical example to demonstrate the concept.

The assumption in DEA is that all inputs and outputs are definite, but in real situation, this assumption is not true. In this study, we suggest a model with fuzzy data called the fuzzy generalized DEA (FGDEA) model, which can treat the stated basic DEA models with fuzzy data in a unified way. In addition, we linearize this model. And finally with a numerical example we evaluate DMUs with fuzzy data by the fuzzy DEA model and the FGDEA model and compare the results of two models.

Data envelopment analysis (DEA) is a powerful tool for measuring the relative efficiencies of a set of decision making units (DMU) that consume multiple inputs to produce multiple outputs. DEA can also be used in activity planning and economic regulation of decentralized units under a common management. One important management challenge in such systems (e.g. schools, hospitals, utilities) is how to implement reductions in common resources (budget, staff, raw material) as well as centrally allocated outputs (students, patients, service areas) while maintaining a high operational efficiency of the system. We extend earlier work on DEA activity planning under resource and output restrictions for the case in which the centralized planner is facing an imprecise shrinkage factor, here expressed as interval data. The implementation of the proposed method is applied for a numeric example.

Data envelopment analysis (DEA) is a commonly used non-parametric technique for performance measurement of decision making units (DMU) that can also be used in normatively in activity planning, resource allocation and target setting. Whereas earlier work in this line have considered completely defined data, prospective use of DEA in activity planning often involves uncertainty with respect to the feasible ranges for allocation of resources and target settings. In this paper, we present an imprecise DEA-based method with interval inputs and outputs as to address this shortcoming. The proposed model corresponds to reasonable managerial objectives concerning the technical efficiency of the subordinate units after implementing resource allocation and target setting.

A new method of project time overrun risk estimation is proposed. The method takes into consideration various risk levels, corresponding to the project environment understood in a more or less wide manner, and interdependencies between them. It is based on expert estimations expressed in linguistic terms, which are flexible and may be defined in each case accordingly to experts preferences. The method allows to insert buffers protecting the project planned deadline, taking into account the complex, interdependent risk factors – in contrast to the existing methods, which usually treat risk factors as independent.

The principal risks in the railway industry are mainly associated with collisions, derailments and level crossing accidents. An understanding of the nature of previous accidents on the railway network is required to identify potential causes and develop safety systems and deploy safety procedures. Risk assessment is a process for determining the risk magnitude to assist with decision-making. We propose a three-step methodology to predict the mean number of fatalities in railway accidents. The first is to predict the mean number of accidents by analyzing generalized linear models and selecting the one that best fits to the available historical data on the basis of goodness-of-fit statistics. The second is to compute the mean number of fatalities per accident and the third is to estimate the mean number of fatalities. The methodology is illustrated on the Spanish railway system. Statistical models accounting for annual and grouped data for the 1992-2009 time period have been analyzed. After identifying the models for broad and narrow gauges, we predicted mean number of accidents and the number of fatalities for the 2010-18 time period.

Lot Sizing strategies are based on determining periods where production will take place and quantities to be produced in order to satisfy demand pattern imposed to system as parameter. Most of the studies in the literature aim to minimize total cost of lot sizing strategies. It is very well-known that related type of problem structure is very hard to solve optimally. Hence, a number of heuristic approaches are proposed to find a near optimum solution for pre-determined problem characteristics. In this study, one of recent algorithms proposed for problem is evaluated in respect of robustness. The aim of this paper is to test whether algorithmic structure of lot sizing strategy is adequate to fully characterize fluctuations in parameters. A discrete-event simulation model is established to understand performance outcomes of proposed strategy.

In this study, we model provider selection and task allocation problem as an expected cost minimization problem with stochastic chance constraints. Two important parameters of Quality of Service (QoS), delay and jitter are considered as random variables to capture stochastic nature of telecom network environment. As solution methodology, stochastic model is converted into its deterministic equivalent and then a novel heuristic algorithm is proposed to solve resulting nonlinear mixed integer programming model. Finally, performance of solution procedure is tested by several randomly generated scenarios.

In collaborative systems (CS), procedures are performed by a team of professionals acting simultaneously, as in surgical rooms in medicine. Using Virtual Reality is possible to create virtual rooms that join people to simulate those medical procedures. In such cases, collaborative features are used to coordinate the interactions of the multiple participants in the environment. To verify if the group performed the procedure correctly, assessment architectures has been proposed. However, network latency can affect that assessment. This paper presents a new architecture for adaptive assessment systems for training based on virtual reality for multiple users, which is able to solve that problem.

Grassland fire simulation is complex because of many influence input parameters. Based on the field data of grassland fire behavior, the paper simulated the spread process of the grassland fire applying GIS tools. Applying experimental data, the study established initial grassland fire speed in control environment at first, and then calculated grassland fire spread speed rate in different wind speed, wind direction, and slope. The study obtained the grassland fire spread speed model applying composition of forces. Finally, the study calculated grassland fire spread time in large-scale grassland spatial. Take the severe grassland fire disaster of Dongwuqi in May 16, 2005 as a cases, comparing with actual results, this model showed reliable, and this model can provides decision-making for grassland fire management department.

Group buying aggregators collect data from heterogeneous sources on the Internet and fit them into a unified schema. As the amount of data grows, the initial schemes need to be maintained and optimized such as being decomposed into a desired normal form. This paper provides an approach to decompose relation schemes of group buying aggregator in a reverse engineering manner. Functional dependencies with degree of satisfaction that are discovered from massive data are used to decompose relation schemes into 3NF. A preliminary data experiment has been done to illustrate the effectiveness of the approach.

The purpose of this paper is to use bibliometric analysis to forecast 3D TV technology. Forecasting emerging technologies and identifying the rate of diffusion of products based on these technologies is difficult because of lack of data. For this aim, ISI Web of Science publication database is searched and posed with findings appropriate S-curve is composed. Then technology life cycle of 3D TV technology is obtained and results are explained.

In this study an ANFIS model is aimed to achieve the vibration signals decreasing the nonlinearity with moving average technique. This nonlinearity is observed by an aging process of an induction motor. The aging process is identified by vibration data taken from healthy and aged (faulty) cases of the same induction motor. In order to change the nonlinearity degree of the system, moving average method is applied to the input and output pairs for different lagging factors. The success of the model is attached two conditions; these are the progress of the ANFIS and the characteristics of the model output. The gauges of these conditions are training - checking errors of ANFIS and cross correlation coefficient between Power Spectral Densities of the model output and faulty case data (real data), respectively.

The processes of cake forming filtration during a loss of coolant accident in a nuclear power plant are of safety-related importance. Thereby these accretion processes are characterized by high complexity. Thus, precise predications to the operation course and effect are hindered. Hence a dynamic model is reasonable for a broad investigation and understanding of these processes. One submodel is the solid fraction movement in the fluid, which is presented in this paper.

It was proposed in the former research that chaos control can be used to suppress electromagnetic interference (EMI) in DC-DC converters. Here, a practical application in a power supply ATX is to be given to show that inputting an external chaotic signal to a Pulse Width Modulation (PWM) control circuit can reduce EMI by spreading the spectra of switch voltage, transformer current and output inductor current over the whole frequency band.

In this paper, we apply the filled function method to solving a global optimization problem which investigates the maximum microhardness of Fe-Mn binary alloys in different layer pairs. A solution algorithm is also proposed.

This note aims to shed some useful light on the underlying long-run price leadership in market data. A cointegrated vector autoregressive analysis is performed using weekly data for gasoline prices observed in three regions in the US. Statistical evidence revealed in the analysis indicates that one of the prices plays the role of long-run leadership, thus exerting persistent influences on the price determination in the other regions.

Environmental pollution accident is an eternal phenomenon in the process of human social activities. Phenomenon of the public spreading often appears after environmental pollution accident and before the government announces the formal environmental conclusion. It will easily result in distorted information and especially magnified harm. This paper establishes an individual assessment model of the public spreading in the Environmental pollution accident based on fuzzy comprehensive evaluation method and a numerical example shows the application of this model.

An innovative control scheme to regulate the power of the TRIGA Mark III research reactor of the National Nuclear Center of Mexico is proposed. Based on the safety limit value imposed on the power rate of change, a sectionally continuous time profile for the ascent of power from its source level to a desired value is generated. A group of fuzzy systems is used to identify the nonlinear functions of the plant's state-space equations. Using Lyapunov stability theory, the consequent parameters of these fuzzy identifiers are adapted. A supervisor control stage ensures the boundedness of the neutron power error signal. The dynamic behavior of the reactor is obtained from the point kinetics equations given by the Fuch-Hansen model. The controller shows a robust regulation against plant's parameter disturbances.

Due to the increase in both consumption and obsolescence rates for new technologies, but shorter life cycles of electronic products cause a significant rise in the volume of electronic waste (e-waste). E-waste practices mostly focus on the processes for a collection system; however, the amount of e-waste critically affects the network design considering the number of facilities and the capacity assignments, as well. We present a framework proposal to predict the amount of e-waste including IT/telecommunication and consumer equipments in Istanbul metropolitan. The main selection reason of Istanbul is the requirement of an urgent intervention to deal with increasing e-waste amount due to its young and crowded population. The result of this study will strongly assist the authorities to configure well-structured strategies for future e-waste management system.

Options pricing is still an open area for exact results. Monte Carlo integration is the unique solution for especially complicated options. It is desired to control variability while implementing Monte Carlo techniques. In order to supply reliable results variance of simulation trials should be decreased. Importance Sampling is one of the variance reduction techniques commonly used in Monte Carlo applications. This study includes a research to gather the appropriate Importance Sampling density which gives the lowest variance. We illustrate the Importance Sampling method on financial options and calculate the value of options. By the same way, it is possible to calculate any expectation that cannot be calculated analytically. Numerical results indicate that longer tailed proposal distributions provide substantial decrease in the estimated variance.

The capability sets of organizations are the key the organization's success. The learning capabilities are crucial for improving capability sets. In literature although this relation has been examined conceptually there are limited studies which show this relation empirically. The objective of this study is to reveal the relation between organizational learning and organizational capabilities. A statistical technique entitled Structural Equation Modeling will be used in order to achieve this objective. Since it reveals the mechanisms behind capability building this model will help managers in their decision making processes.

In this work, an ensemble of neural networks is built by an algorithm called Learn⁺⁺.NC and applied for fault detection and diagnosis. The algorithm is capable of incrementally learning new classes of faults, while retaining the previously acquired knowledge. The detection of new classes in subsequent data is achieved by thresholding the normalized weighted average of outputs (NWAO) of the neural networks in the ensemble. The unknown classes detected remain unlabeled until their correct labels can be assigned. The proposed method is applied to the identification of simulated faults in the feedwater system of a Boiling Water Reactor (BWR).

A fuzzy reliability approach is a good alternative for a probabilistic reliability approach to assess event reliability data when event historical data is unavailable or inadequate for statistical calculation. It applies possibility theory to represent event failure in qualitative natural languages and fuzzy set theory to represent those qualitative natural languages in mathematical form. In this study, we develop a new area defuzzification technique and propose it to assess nuclear event reliability data from their corresponding failure possibilities. To justify the applicability of the new technique, we define five fuzzy rules to be satisfied by the technique and use five sets of fuzzy subsets to mathematically confirm the technique. The results show that the new area defuzzification technique is suitable for assessing nuclear event reliability data in the fuzzy reliability approach.

This paper describes a multivariable H₂-H_∞ robust control approaches, for an ill conditioned, difficult to control system, given by a plasma chemical reactor CF₄/O₂. The control of such a system resulted, in the past, in a slow, oscillatory coupled response. The goals of applying robust controls, is the achievement of a decent smooth response with minimum oscillations and overshoot. In the pursuit of this goal, we proceed to the quantification of uncertainties affecting the plasma process. Then, the robustness conditions in closed loop, to be achieved, are clearly stated. Only then, the H₂ and H_∞ algorithms are applied, after the construction of the augmented system resulting from the standard process form. To provide the advantages of the H_∞ algorithm, we applied the H₂ one and compare the results obtained between the two controls.

When designing controllers for complex systems, it is not only necessary to control the system but to improve the robustness in order to get a better response. Some indexes allow to measure the performance of the controller, such as the gain and phase margins of the system response. In this paper a computational tool based on genetic algorithms is designed and implemented in order to optimize the robustness of a controller. So it is possible to analyze how the variation of the controller parameters influences the robustness of the system. The tool is applied to the optimization of a LQ controller for a MIMO autonomous vehicle.

In offshore oil platforms, choke valve erosion is a major issue. The difference between the theoretical and the actual values of the valve flow coefficient is commonly used during operation as the indicator of the choke erosion. The actual value of the flow coefficient is analytically calculated as a function of measured and allocated parameters. The latter are typically obtained using a physics-based model proved to be inaccurate for some operation conditions. As a consequence, inaccuracies are introduced in the evaluation of the erosion evaluation and this undermines the possibility of using it for prognostic purposes. To overcome this hurdle, a hybrid ensemble of physics-based and Kernel Regression (KR) models is proposed. The method is verified on real measurements.

This work presents a developed novel optimization process that can be used for the synthesis of mechanical components and systems. One sample application to the design optimization of space frames is presented in detail. Several other applications are also indicated. The main objective in the sample application is to minimize the space frame volume and satisfy stress constraints. The optimization effectively uses the devised Heuristic Gradient Projection (HGP) technique to synthesize the optimum design of the Midi-Bus frame. Results indicate a striking improvement over available designs and remarkably faster convergence over other design and optimization techniques. The technique can moreover be effectively applied to other large 3D-space frame synthesis and optimization. The developed synthesis methodology is also applied to other designs as a tailored optimization paradigm for a better theory of design realization.