Computational Intelligence in Decision and Control

The following sections are included:

• Foreword

• CONTENTS

The following sections are included:

• Introduction

• Soft computing defined by essential properties

• Soft computing as the mixture of several preexisting techniques

• Soft computing as opposite to hard computing

• Soft computing as a tool to cope with uncertainty and imprecision

• Soft computing at the European Centre for Soft Computing

• References

In this paper we consider the problem of how to derive mass functions systematically from data samples, and the ensuing problem of how to combine different mass functions thus derived. We show that a mass function can be efficiently, systematically derived from multivariate data space. We also show that combining mass functions thus derived appears trial, but this approach opens door for a potentially completely new rule for combining mass functions.

In this paper, we improve a very famous algorithm Apriori for data mining. Algorithm Apriori is to mine motifs, i.e., maximal frequent patterns or itemsets. The algorithm assumes that all items are totally ordered, e.g., A < B < C < D < for items A,B,C,D etc. Furthermore, items within an itemset are sorted in lexicograghic order. We improve the algorithm, present a new one without assuming any ordering. The new algorithm looks sound and more efficient.

Some preliminary classes have been developed in MATLAB to interface with the JET PPF system which stores the data elaborated after the discharge. These classes simplify the most common operations required during an interactive analysis and visualization of JET data, while keeping all the power of a full programming language. The first class is ppfs which given a shot number retrieves a list of all the main family (DDAs) related to that shot, the second is ddas which lists all the DTYPEs belonging to a DDA. The last two classes are used for retrieving the actual data and perform the most common operation, such as basic plotting slicing and arithmetic operation.

Free unfolding in neutron spectroscopy means reconstructing energy spectra from experimental data without a priori assumptions regarding their shape. Due to the ilI-conditioned nature of the problem, this cannot be done analytically. Neural networks (NN) are here applied to this task and synthetic data is used for training and testing. Results showed very consistent performance especially in the region of low and medium counts, where they fall near the Poisson statistical boundary. Comparison with other unfolding methods validated these results. Application time on the order of μs makes NN suitable for real-time analysis. This approach can be applied to any instrument of which the response function is known.

Integrated Data Analysis (IDA) offers a unified way of combining data from different experiments to obtain improved results. IDA meets with typical issues arising in the analysis of data from magnetic confinement fusion experiments. Heterogeneous and complementary experimental data as well as physical prior information are integrated employing Bayesian probability theory. The concept of IDA is compared to the traditional approach for data analysis where sequential analysis and iterative schemes are usually employed. In contrast to classical backward inversion techniques, IDA needs only forward modeling and a thorough error assessment. In practice, the probabilistic description of systematic measurement and model uncertainties are of major importance to resolve data inconsistencies, which is a prerequisite for the integration of different diagnostics results. Complex error propagation is an inherent part of a concise probabilistic one-step analysis.

Techniques for plasma diagnostic data processing that are relatively new to fusion research are described. The following problems are considered: plasma boundary reconstruction using video images, optimization of real or numerical experiments with artificial neural networks, application of support vector machine to the classification of plasma pulses, processing of magnetic diagnostics data using hidden Markov models, clustering and navigation in the database of graphical information with Kohonen self-organizing maps, reconstruction of the distribution of light source using high-resolution plasma images. Examples of successful solution of different problems in fusion are given, which prove the high efficiency of the methods and motivate further applications.

Sales forecasting has a great impact on facilitating effective and efficient allocation of scarce resources. However, how to best model and forecast sales has been a long-standing issue. There is no best method of forecasting in all circumstances. Therefore, confidence in the accuracy of sales forecasts is derived by corroborating the results using two or more methods. This paper evaluates the relative performance of Linear Multiple Regression, Artificial Neural Networks and Adaptive Neuro Fuzzy Networks by applying them to the problem of sales forecasting for a Turkish paint producer firm. The results indicate that Adaptive Neuro Fuzzy Networks yields better forecasting accuracy in terms of Root Mean Square Error and Mean Absolute Deviation.

A new methodology for analysis of periodical time series is proposed. It is based on application of special soft computing techniques: fuzzy transform and perception-based logical deduction.

When time series of observed distributions arise in real-life, they are usually represented as classic single-valued time series, which supposes a substantial loss of information. Histogram time series (HTS) represent time series of distributions avoiding this loss. This paper adapts smoothing methods to HTS. These methods serve well for forecasting and removing fluctuations. The forecasting performance of these methods is shown by a financial example.

One of the pernicious master issues of supply chain is the bullwhip or whiplash effect; i.e. the variability of the demand information between stages of the supply chain and the increase of this variability as the information moves upstream through the chain. In this paper, the reaction of bullwhip effect to grey and fuzzy grey GM(11) forecasting models is analyzed by quantifying the bullwhip effect with variance estimation and Lyapunov exponent using a simple supply chain Matlab simulation (i.e. beer game model) and results derived from the simulation compared with the ones obtained from using selected crisp forecasting model.

There are several studies in the literature that assumes fuzzy demands in supply chain or production planning models but most of them do not mention about how to derive the fuzzy demands from statistical and judgmental forecasts. In this study we propose a methodology to aggregate the forecasts coming based on different sources; namely statistical methods as well as the experts judgments, and to obtain an aggregated demand forecast that is represented by a possibilistic distribution. Results of the statistical and judgmental forecasts are represented by triangular possibilistic distributions. Subsequently, those results are combined by using weights of each input forecast. An illustrative example is also provided.

In this paper, we propose a forecasting model suitably developed to forecast monthly anchovy catches in the north area of Chile. This forecaster is developed in order to combine in the most effective way the capability of the quantum particle swarm optimization algorithm and a reduced multivariate polynomial. The performance evaluation of this model achieves an approximately determination coefficient equal to 93%. Beside, the advantages of the proposed model are reduced parsimony and increasing accuracy precision.

In the classical logic the inferential process is based on the modus ponens rule for which when the rule that connects the premise with the consequence is true and the premise is true, we know that also the consequence is true. Now in the fuzzy inferential process [1] the premise and the consequence are not logic propositions that assume the logic values true or false but are fuzzy sets that can assume values between zero and one. The classical logic rule is replaced with the fuzzy rules R. When a fuzzy set A (the antecedent) is known, then one needs a way to infer the consequence B from A and R. In fuzzy set theory a lot of different methods are given to solve the previous problem. Now in this paper a new conceptual frame denoted Morphogenetic System is used in place of the traditional approach to the inferential process. The main idea in the morphogenetic system is the separation of the fuzzy inferential process into two parts. One is given by the rules that represent the fuzzy system and the other is the environment that represents all the possible premises A. Now given a premise A or a generic fuzzy set, we define the space of the fuzzy objects: in this space we represent the elements of the fuzzy rules. Now we create a special projection operator that projects the premise A into the fuzzy rules to obtain consequence B which is the part of A that is coherent with the fuzzy rules. When the premises are orthogonal to the rules we have not consequence B. The definition of the projection operator and the properties of the projection operator are the aim of this paper.

Given a set of points in the plane, we find the fitting ellipse by considering the sum of the Euclidean distances of the given points to the ellipse. Our approach to this optimization problem is of a hybrid kind using two heuritics: Differential Evolution (DE) and classical Nelder-Mead algorithm. DE is used as a coarse search method that gives an initial point to Nelder-Mead, and then it performs an efficient local search. We test our approach with several simulations and we illustrate better performances against DE alone.

We study the problem of warning in the oceanic data processing. Due to high-dimensional and dynamic oceanic data, we develop a maximum entropy approach for this problem. First, we describe the framework of the oceanic data processing. Then, we address dimensionality reduction by clustering so as to attempt to minimize the data complexity. And, we address the inherent dynamic nature of the problem by explicitly modeling the data as a time series. We show how this representational expressivity fits naturally into a maximum entropy framework. Final, we conduct experiments, and show that our maximum entropy formulation outperforms several algorithms in data processing for oceanic warning system.

Learning systems in massive databases are key elements for the development of both efficient data retrieval methods and data-driven theories. Typically in fusion, waveforms have a very high dimensionality and the construction of classification systems (mainly through unsupervised techniques) without the help of visual techniques is very difficult. The Grand Tour is a 2D visual exploratory method that can be used in the clustering process. Applications to data retrieval and disruption classification in JET are presented.

In this paper, we deal with the problem of learning decision rules from partially uncertain data based on rough sets. The uncertainty exists in the decision attribute and not in condition attribute values of the decision system. This latter is represented by the belief function theory. So, we will adapt the basic concepts of rough sets in order to generate rules, denoted belief decision rules.

In this paper we present a profile-based approach to information filtering by an analysis of the content of text documents. The Wikipedia index database is created and used to automatically generate the user profile from the user's document collection. The problem-oriented Wikipedia subcorpora are created (using knowledge extracted from the user profile) for each topic of user interests. The index databases of these subcorpora are applied to filtering information flow (e.g., mails, news). Thus, the analyzed texts are classified into several topics explicitly presented in the user profile. The paper concentrates on the indexing part of the approach. The architecture of an application implementing the Wikipedia indexing is described. The indexing method is evaluated using the Russian and Simple English Wikipedia.

This work briefly analyses the difficulties to adopt the Semantic Web, and in particular proposes systems to know the present level of migration to the different technologies that make up the Semantic Web. It focuses on the presentation and description of two tools, DigiDocSpider and DigiDocMetaEdit, designed with the aim of verifYing, evaluating, and promoting its implementation.

In order to model natural phenomena and their relationships, fuzziness and vagueness should be considered. Topological relations are one of the most important definable characters between two spatio-temporal objects. This paper proposes a new way for modeling and constructing topological relationships between spatio-temporal objects under the umbrella of rough topological relationships in space and time. And as a case study, land covers as spatio-temporal objects considered. Results show that rough set theory models uncertainty of spatio-temporal objects in a transparent and explainable way.

We introduce a survey on different techniques that have recently been issued in the search for a characterization of the representability of typical (non trivial) semiorders by means of a real-valued function and a threshold of discrimination.

We focus on a possible generalisation of the theory of congruences on a lattice to a more general framework. In this paper, we prove that the set of congruences on an m-distributive multilattice forms a complete lattice and, moreover, show that the classical relationship between homomorphisms and congruences can be adequately adapted to work with multilattices.

A method to get the transitive closure, a transitive opening and a transitive approximation of a reflexive and symmetric fuzzy relation is presented. The method builds at the same time a binary partition tree for the output similarities.

For representing the concepts related to a given domain with uncertain or incomplete information about a set of objects, Zadeh's fuzzy set theory and Pawlak's rough set theory have been most influential on this research field. However, a lot of relations determined by the attributes of objects do not satisfy the transitivity, and the information about the symmetry of a relation is mostly uncertain as well. This motivates us to explore a new class of relations, so-called a class of fuzzy semi-equivalent relations. The present work introduces the notion of fuzzy semi-equivalent relation and present approaches to the fuzzification of indistinct concepts approximated by semi-equivalent relations.

A fast method to compute a T-indistinguishability from a reflexive and symnetric fuzzy relation is given for any left-continuous t-norm, taking O(n³) time complexity, where n is the number of elements in the universe. It is proved that the computed fuzzy relation is a T-transitive opening when T is the minimum t-norm or a strictly growing t-norm. As far as we know, this is the first known algorithm that computes T-transitive openings preserving the reflexive and symmetric properties.

We investigate interpretations of formulas ψ of a first order fuzzy logic in models which are based on objects of a category SetF(Ω) which consists of Ω-sets, i.e. sets with similarity relations with values in a complete MV-algebra Ω. These interpretations are fuzzy sets in an Ω-set (A, δ), i.e. a morphism (A, δ) → (Ω, ↔) in a category SetF(Ω). We show that if is a strong homomorphism between two models then there are also strong relationships between interpretations and . Finally, for any model based on an Ω-set (A, δ) we construct another model based on a set F(A, δ) of all fuzzy sets in (A,δ) and such that a singleton map {-} is a strong model homomorphism .

Determining concept similarity in heterogeneous ontologies is a vital problem in the area of the semantic web. Current approaches normally consider single hierarchial concept relations, only, which fail to express rich and implied information. Moreover, concept similarity under multiple relations as required in many application scenarios of semantic web services has not been investigated, yet. Basing on a method to merge heterogenous ontologies into an application ontology, here a representation model for application ontologies is elaborated. It has the form of a semantic net with multiple weighted concept relations, for which a novel algorithm to assess concept similarity is presented.

We can use the temporal and spatial properties of data for integrating data coming from different data sources more efficiently and according to the requirements of the Data Warehouse administrator. For this integration process we propose a fuzzy algorithm in order to obtain more precise data in the Data Warehouse. The data source schemas as well as the Data Warehouse schema are expressed using an extension of an ontology definition language which allows the incorporation of metadata to support the integration process.

This paper traces the historical origins of perceptual computing and credits Tong and Bonissone [9] as being the first to originate it but under a different name. It also explains why interval type-2 fuzzy sets, and not type-1 fuzzy sets, should be used in perceptual computing.

This paper tries to show, only from a theoretical perspective, the importance of well designing the representation of fuzzy systems whose behaviour is known by a linguistic description of it. The process of designing the representation by means of fuzzy sets, connectives, and relations, marks an actual distinction between the fuzzy and the formal logic methodologies, two different disciplines whose agendas are not coincidental.

Based on the theory of fuzzy logic in sense of Parvelka's theory, in this paper, we investigate the semantical theory of propositional fuzzy logic based on Lukasiewicz algebra on [0,1]. The tautology theory is established by extending concept of tautology in classical setting to generalized tautology, the mutual relation among generalized tautologies is investigated.

This paper focuses on investigating the inference rules with generalized quantifiers in a lattice-valued first-order logic system ℓF(X) with truth-values in a linguistic truth-valued lattice implication algebra (L-LIA). Since the qualifier constraint may become more important when considering the semantic of natural language, so in using multi-valued logic as a tool to model approximate reasoning, how to control the truth-value transfer during the inference process for some rule with qualifiers is very important, which is not the case in classical logic in which the logic deduction system is symbolic reasoning with strict syntactical proof, the semantics are too simple to be considered. This work put more effort on semantic interpretation and truth-valued transfer, especially present and investigate some reasoning rules with generalized quantifiers (rather than universal quantifiers and existential quantifier only) for lattice-valued first-order logic system ℓF(X). We prove the satisfiability and validity of these inference rules, where inference rules are interpreted by the semantic truth value transfer, which shows the control of truth-value level during the inference process.

For crisp relations the concept of interval order can be written in different equivalent ways and it satisfies that its strict preference relation is transitive and the associated large preference relation is complete. In this contribution we study the previous implications for each of the (non-equivalent for fuzzy relations) possible definitions of fuzzy interval order.

The goal of this work is to extend the concept of linguistic variable to the interval-valued case.

After a brief introduction on fuzzy numbers and linguistic variables in [0,1], we will define the interval-valued linguistic variables and we will study their behavior through two properties. After that, we will show their utility to replace the absent values in an L-Fuzzy Context.

Triangle Logic is a formal fuzzy logic with intervals as truth values. Its construction is based on triangle algebras: equationally defined structures that are equivalent with certain residuated lattices on a set of intervals, which were called interval-valued residuated lattices (IVRLs). We prove that the so-called pseudo-prelinear triangle algebras are subdirect products of pseudo-linear triangle algebras. This can be compared with MTL-algebras (prelinear residuated lattices) being sub direct products of linear residuated lattices. Using this result, we prove an analogue of the chain completeness of MTL for Pseudo-prelinear Triangle Logic. It also enables us to prove properties of pseudo-prelinear triangle algebras more easily. We give some examples.

We show here some of our results on intuitionistic fuzzy topological spaces. In 1983, K.T. Atanassov proposed a generalization of the notion of fuzzy set: the concept of intuitionistic fuzzy set. D. Çoker constructed the fundamental theory on intuitionistic fuzzy topological spaces, and D. Çoker and other mathematicians studied compactness, connectedness, continuity, separation, convergence and paracompactness in intuitionistic fuzzy topological spaces. Finally, G.-J Wang and Y.Y. He showed that every intuitionistic fuzzy set may be regarded as an L-fuzzy set for some appropriate lattice L. Nevertheless, the results obtained by above authors are not redundant with other for ordinary fuzzy sense. Recently, Smarandache defined and studied neutrosophic sets (NSs) which generalize IFSs. This author defined also the notion of neutrosophic topology. We proved that neutrosophic topology does not generalize the concept of intuitionistic fuzzy topology.

This paper explores the relationship between object level intuitionistic fuzzy sets and predicate based intuitionistic fuzzy sets. Mass assignment uses a process called semantic unification to evaluate the degree to which one set supports another. Intuitionistic fuzzy sets are mapped onto a mass assignment framework and the mass assignment semantic unification operator is generalised to support both mass assignment and intuitionistic fuzzy sets. Transfer of inconsistent and contradictory evidence is also dealt with. As a consequence, by conjoining the mutual semantic unification of two sets a similarity measure emerges.

There are various probability concepts on IF-events (RIEČAN¹). The paper deals with the Gödel connectives max-min (RIEČAN²). In the framework a special type of ergodic theorem is presented.

The purpose of this paper is to commence studying the incompatibility in the Atanassov's intuitionistic fuzzy sets framework. In order to do this, firstly we deal with the concept of -incompatible sets, where is an intuitionistic t-norm, relating it with the -contradictory sets, where is a intuitionistic fuzzy negation.

Next, an axiomatic model for measuring -incompatibility is introduced, and finally some methods for obtaining families of such measures are provided.

This paper is devoted to introduce an axiomatic model to distinguish what functions are suitable for measuring the degree of contradiction between two Atanassov's intuitionistic fuzzy sets. After stating the needed background, in section 2, we justify and present the axioms that a contradiction measure must satisfy, and the first examples are set out. After motivating the necessity of achieving some definition for modelling the continuity, in the next section we introduce the concepts of semicontinuity from below and semicontinuity from above for contradiction measures. Finally, in section 4, some families of contradiction measures are constructed.

In this paper we construct an expression for calculating the total contrast of an image from Atanassov's intuitionistic fuzzy S-implications and from the fuzzy expected value.

We propose a new method for ranking alternatives represented by Atanassov's intuitionistic fuzzy sets (A-IFSs) which takes into account the amount of information related to an alternative (expressed by a distance from the ideal positive alternative) and the reliability of information (how sure the information is).

A lattice-valued logical algebra–lattice implication algebra (LIA) and its corresponding logic and reasoning systems have been extensively investigated in the last few years. There have been also some research works with an attempt to establish the linguistic truth valued logical algebra for decision making purpose, including the linguistic truth valued logical algebra based on LIA. In this paper, in order to characterize the feature of linguistic truth-valued algebra based on LIA, firstly the product-irreducible and sum-irreducible elements of LIA are introduced and their properties are discussed,. Then some decomposition theorems in linguistic truth-valued LIA are given, these results will place a support for the further investigation of linguistic truth-valued algebra structure and modeling.

The main result of the paper is a submeasure extension theorem. As a domain a multivalued algebra is considered and the range is an lattice ordered group with some properties.

We introduce Restriction Level relations (RL-relations) to represent imprecise relations. The proposal is based on the notion of RL-representation of imprecise properties proposed by the authors as an alternative to fuzzy representations, with the advantage that the ordinary properties of the crisp case are preserved. On this basis we define RL-preference structures and discuss about the properties of this new approach.

When we deal with group decision making (GDM) situations with incomplete preference relations, there exist cases in which the classical selection procedure (aggregation and exploitation) could not be applied satisfactorily. For example, we could find that some preference degrees of the collective preference relation cannot be computed in the aggregation phase and consequently, the ordering of some alternatives cannot be computed in the exploitation phase. To overcome this problem, we present a selection process for GDM with incomplete 2-tuple fuzzy linguistic preference relations that requires three phases: (1) estimation phase of missing values, (2) aggregation phase and (3) exploitation phase.

Decision making with linguistic information is a research hotspot now. This paper introduces a linguistic truth-valued concept lattice based on lattice-valued logic for dealing with decision making under uncertain linguistic information. For dealing with the different linguistic terms more formally, a unified set of linguistic terms is defined according to the features of linguistic terms, and this set is verified to be a bounded lattice and then to be a lattice implication algebra through the definitions of relevant operations on it. Based on the unified set of linguistic terms, a linguistic truth-valued formal context as the premise of linguistic truth-valued concept lattice is defined and the linguistic truth-valued decision method is provided.

We present a multi-valued temporal reasoning framework which can be used to analyse problems with uncertainty under dynamic environment. We show that it is feasible to combine temporal logic and multi-valued logic and we illustrate the framework with a simple but realistic example.

The aim of this paper is to compare two common methods used to solve multiple criteria decision making (MCDM) problems: the general fuzzy method and a method based on influence diagrams. The first method is based on the Possibility Theory while the other one is based on the Probability Theory. The two methods are used to evaluate and choose the best means of transport in order to show advantages and disadvantages. Expert knowledge has been applied in this work to model the preferences and show how the final decision changes according to these weights. This problem has been broadly used in previous literature as a prototype problem to evaluate MCDM methods. Some interesting conclusions regarding this comparison have been drawn.

Usually customers' preference on product varies with personal character, feeling, aesthetic and so on; the personal preference plays an essential role in choice of products. We have proposed a subjective evaluation model. Particularly, a preliminary research is conducted to select bipolar words for each sensory attribute by means of semantic differential method and then a population of subjects is called to assess evaluated objects. These subjective data are then used to generate sensory profiles by making use of the voting statistics. Depending on the customer-specified target-oriented preference, we can obtain the probabilities of meeting different targets for each selected sensory attribute. To consider the priorities among targets the appropriate PRI-AVE operator is used to aggregate the partial values.

An agent holds a great number of beliefs. By the notable AGM model, nine tenths of the previous beliefs of the agent will be revised if a new belief is inconsistent with them. In reality, more often than not, people hope that only the relevant parts of a belief need to be adjusted. In a finite language, Parikh proposed the concept of splitting of belief set and the existence of the finest splitting of any set of formulae in 1999. While Kourousias and Makinson proved that the existence of the splitting of any set of formulae in an infinite language in 2007. Nevertheless, neither of them have presented us with a constructive method to gain the finest splitting. In the paper, a constructive method was proposed to find the finest splitting of any set of formulae. And we analyse the complexity of the method.

We develop a new approach for linguistic decision making with Dempster-Shafer (D-S) belief structure by using the 2-tuple linguistic representation model. Then, we are able to represent the D-S problem with linguistic information and without loss of information in the computing process. For doing this, we suggest the use of different types of linguistic aggregation such as the 2-tuple linguistic ordered weighted averaging (2-TOWA) operator. Then, we will obtain the belief structure - 2-tuple linguistic ordered weighted averaging (BS-2-TOWA) operator.

A version of the conditional probability of an IF-event is formulated in this contribution. Max and min operations with IF-sets (KRACHOUNOV¹) are considered instead of Łukasiewicz operations. Also some properties of conditional probability are proved.

In this paper we deal with the study of triangle functions defined on the class , of distance distribution functions with range in , where n is a given positive integer. This study can be formulated in terms of triangular conorms on the set Σ_n of non-decreasing n-lists (a₁,…, a_n) ∊ [0, +∞]ⁿ equipped with the natural (product) order. Using triangular conorms on [0, +∞] and triangular norms on {0, 1, …, n} we describe different classes of appropriate triangular conorms on [0, +∞]ⁿ.

Process capability indices are summary statistics to point out the process performance. In the literature, various capability indices are used for this aim. Process accuracy index (C_a) measures the degree of the process centering and gives alerts when the process mean departures from the target value is analyzed when the specification limits and process mean are fuzzy. A theoretical application is illustrated to show the usage of .

The basic model of attribute computing network based upon qualitative mapping does not consider that the input includes both qualitative attribute and quantitative attribute. To solve this problem, the thesis defines a mixed input attribute computing network model. Meanwhile, since the current attribute computing network study algorithm will encounter the problem of excessive time-consuming training in case of too many training samples, the thesis accordingly initiates boundary study algorithm.

Process perfonnance can be analyzed by using process capability indices (PCIs) successfully. They are summary statistics to depict the process situation. PCIs have not reliable results if the process observations have correlation. In this paper, PCIs called Robust Process Capability Indices (RPCIs) are analyzed unlike traditional PCIs. Also specification limits and standard deviation are defined as fuzzy number. Fuzzy RPCIs are created for piston manufacturing process.

A parallel implementation of certain neural newtork algorithms using Java concurrent programming is described. Although the techniques used are by no means new it does not seem to be widely known that thus a significant improvement of performance can be achieved when using one of the new multiprocessor machines that begin to dominate the PC market. In order to demonstrate the potential of these methods preliminary test results concerning a ranking algorithm are reported.

In this paper, an Infinite Impulse Response-Locally Recurrent Neural Network (IIR-LRNN) is employed for modeling the dynamics of the Lead Bismuth Eutectic eXperimental Accelerator Driven System (LBE-XADS). The network is trained by recursive back-propagation (RBP) and its ability in estimating transients is tested under various conditions. The results demonstrate the robustness of the locally recurrent scheme in the reconstruction of complex nonlinear dynamic relationships.

This paper analyses the application of artificial neural networks (ANN) for routing in proactive (using protocols such as DSCV, OLSR, etc) mobile ad-hoc networks (MANET). In this context, neural networks are capable of recognizing a behavioural pattern in a network and predict the next step. For instance, when a route will be no longer active, the neural network searches an alternative route before it fails routing traffic. Simulation results show how this neural approach improves the performance of the routing as the downtime in the routing is significantly reduced when a route is lost.

Query evaluation improvement and association rules are two interesting research topics in data query and management. In this paper, the relations are decomposed with respect to the mined association rules, and several basic data query expressions are rewrote which have relatively less time cost. This is desirable for processing queries in an efficient manner.

In this study, we bring forward a concept of collaboration and knowledge sharing of information granules represented in the language of the formalism of fuzzy sets. The growing interest in agent systems in which collaboration plays a pivotal role; we discuss ways of communication of information granules-fuzzy sets and show mechanisms of reconciliation of findings obtained at the level of individual data sites. In particular, it is shown how the mechanisms of collaboration are used to augment processing of fuzzy clustering being a cornerstone of fuzzy modeling.

The aim of this paper is to build a user profile according to his own perception. We focus on the dynamic construction of the profile, which will increase the satisfaction of the user by being more personalized and accommodated to his particular needs. We suggest two methods to define the perception and transform it into a profile; the first method is achieved by querying the user and getting answers, which will guide through the process of implementation of the profile; the second method is achieved by comparing different subjects and ending up by an appropriate choice. As a case study, we took the example of color perception. We rely on fuzzy logic to represent the different aspects of color and to formulate the questions and answers.

We suggest the use of parametric LU-fuzzy numbers, obtained by families of monotonic spline functions, to model fuzzy partitions and to use as basic functions in the direct and inverse fuzzy transforms; we point out the advantages of the parametrizations in flexible approximation of continuous and differentiable functions and in curve fitting.

Structural pattern recognition techniques are an efficient way to apply a pattern oriented data retrieval paradigm. Some techniques have already been implemented in the JET Analysis Cluster (JAC) by means of a general purpose tool (software application) to allow the identification of similar patterns (structural shapes) inside temporal evolution signals. Data retrieval methods are based on three essential aspects: feature extraction (to reduce signal dimensionality), the classification system (to index objects according to some criteria) and similarity measure (to compare how similar two objects are), but there is not a single solution or unique criterion to handle these key elements. This paper provides a new solution to the localization and extraction of similar patterns in time-series data. Alternative searches are proposed to objectively increase the recognition of similar patterns so as to achieve better results on the data retrieval. In the proposed approach, patterns are represented by string of characters. Looking for patterns means looking for characters. The recognition problem is translated into a character-matching problem. Thinner search strategies have been studied with excellent results in the detection of long subpatterns. Long subpatterns are not so easy to identify since even a single mismatch in one character can compromise similarity between two patterns. Identifying long patterns in a fast, fault tolerant and intelligent way is the aim of the analyzed strategies, formally based on statistical criteria and some aspects of probability theory.

A large number of accuracy measures for image classification are actually available in the literature for cris classification. Overall accuracy, producer accuracy, user accuracy, kappa index and tau value are some examples. But in contrast to this effort in measuring the accuracy in a crisp framework, few proposals can be found in order to determine accuracy for soft classifiers. In this paper we define some accuracy measures for soft classification that extend some classical accuracy measures for crisp classifiers. This class of measures takes into account the preferences of the decision maker in order to differentiate some errors that in practice may not be have same relevance.

In nuclear reprocessing plants, it is required to verify the coherence of material flow and the conformity of the process to the prescribed procedures. To aid the verification, the Data Analysis and Interpretation (DAI) software has been developed by some of the authors with the aim of detecting and recognizing different functional behaviors in the liquid transfer from one tank to another during the reprocessing. Monitored pressure and temperature signals are converted using tank calibration curves into volume signals and used for the detection of the start and end times of the functional behaviors and the identification of the transfer type, (i.e., by pump, air-lift or siphon). The present paper illustrates a fuzzy clustering algorithm for performing the identification of various types of non-linear liquid transfers. The performance of the fuzzy clustering classification algorithm is tested on real plant data.

Searching for similar behavior in previous data plays a key role in fusion research, but can be quite challenging to implement from a practical point of view. This paper describes the design of an intelligent measurement instrument that uses similar waveform recognition systems (SWRS) to extract knowledge from the signals it acquires. The system is perceived as an Ethernet measurement instrument that permits to acquire several waveforms simultaneously and to identity similar behaviors by searching in previous data using distributed SWRS. The implementation is another example of the advantages that local processing capabilities can provide in data acquisition applications.

Thermonuclear fusion devices of the Tokamak type can operate in distinct confinement regimes, which present different properties in terms of performance and plasma parameters. Discriminating among them in real time would represent a useful advantage for an efficient control of the experiments. A comparison between two automatic identifiers, one based on fuzzy logic and another based on classification and regression trees, is presented. A robustness assessment, adding Gaussian white noise to the input signals, was performed to determine the properties of the two approaches in realistic experimental conditions.

In this paper we analyze the structural properties of the inconsistencies detected by the crude algorithm for segmentation of digital images introduced by some of the authors in a previous work. Such analysis will suggest an alternative algorithm for image segmentation.

The Orthogonal Variant Moments (OVM) are proposed in this paper as a way characterizing any function or signal in general. Our approach to the theory of visual perception is based on the study of the low level vision system by Orthogonal Variant Moments while most of the works on this field use invariant-moments. An application of this method to computer-vision proves the efficiency of this approach.

In medical image analysis, one important step is the image segmentation. Existing segmentation procedures are generally based on features calculated from gray levels and other information directly extracted from images. They can not take into account the physical significance of each class obtained during the classification procedure, which is strongly related to human medical knowledge. For solving this problem, a number of tissue classification systems integrating specific medical knowledge have been developed. However, there systems are too oriented to specific applications and less flexible for being adapted to other applications. In this paper, we present a new tissue classification system which combines both the classical fuzzy c-means classification algorithm and the human medical knowledge on geometric properties of different tissues and organs. In this system, a general geometric model has been developed in order to generalize all kinds of medical images and mathematically formalize non structured and non normalized medical knowledge. A user friendly interface has been designed so that medical knowledge can be easily transformed into this data structure in an interactive way. This system has been successfully applied to MRI images for classification of tissues of thigh.

A novel non-invasive methodology is proposed for the classification of BWR two-phase flow regimes, Neutron radiography images (frames) of coolant flow recordings (videos) are used. Following image pre-processing via directly computable statistical operators, the extracted statistical characteristics are input into an ensemble of self-organizing maps. The individual decisions are averaged, resulting in the accurate on-line classification of each image into the corresponding flow regime.

This work deals with image processing based upon non-linear diffusion PDEs (Partial Differential Equations). Some analytic formulation will be introduced to obtain the 3D diffusion tensor, replacing Jacobi's numerical methods by expressions based on invariants of the symmetric matrix. Later, CED (Coherence Enhancing Filtering) anisotropic filtering properties will be observed and will be combined with isotropic diffusion, providing a type of filtering that allows combining noise removal and local structure preservation. Last, some applications 3D grey-level will be presented.

Knowledge-based multi-attribute classification problems are, at least, ill-structured or even unstructured ones, since a human being judgments are the primary source of information for their solving. Thus, not only the classification rules eliciting, but the application domain (AD) structuring as well is a complex problem itself, particularly, in the context of complete (up to the expert knowledge) and consistent knowledge base construction for a diagnostic decision support system (DDSS). Two techniques are proposed as aids for an expert in such problem structuring. It is argued that AD structuring and classification rules eliciting have to be arranged as interconnected procedures.

We show that on the basis of fuzzy transform, the problem of reconstruction of corrupted images can be solved. The proposed technique is called image fusion. An algorithm of image fusion, based on fuzzy transform, is proposed and justified. A measure of fuzziness of an image is proposed as well.

In this paper we provide a general description of how to combine a collection (not necessarily finite) of asymmetric distances in order to obtain a single one as output. Moreover, we apply the results obtained in our framework to the complexity analysis of programs and algorithms in Computer Science.

We develop a new decision making method by using distance measures and induced aggregation operators. We introduce a new aggregation operator called the induced ordered weighted averaging distance (IOWAD) operator. We study its definition and some of its main properties. We apply this aggregation operator in a business decision making problem. We focus on an investment selection problem.

In this paper we consider mixture operators to aggregate individual preferences and we characterize those that allow us to extend some majorities rules, such as simple, Pareto and absolute special majorities, to the field of gradual preferences.

The Reference Point Method (RPM) is a very convenient technique for interactive analysis of the multiple criteria optimization problems. The interactive analysis is navigated with the commonly accepted control parameters expressing reference levels for the individual objective functions. The final scalarizing achievement function is built as the augmented max-min aggregation of partial achievements with respect to the given reference levels. In order to avoid inconsistencies caused by the regularization, the max-min solution may be regularized by the Ordered Weighted Averages (OWA) with monotonic weights which combines all the partial achievements allocating the largest weight to the worst achievement, the second largest weight to the second worst achievement, and so on. Further following the concept of the Weighted OWA (WOWA) the importance weighting of several achievements may be incorporated into the RPM. Such a WOWA RPM approach uses importance weights to affect achievement importance by rescaling accordingly its measure within the distribution of achievements rather than by straightforward rescaling of achievement values. The recent progress in optimization methods for ordered averages allows one to implement the WOWA RPM quite effectively as extension of the original constraints and criteria with simple linear inequalities.

This paper will introduce a new method to obtain the order weights of the Ordered Weighted Averaging (OWA) operator for multi criteria decision making. We will first show the relation between using fuzzy quantifiers and neat OWA operators and then offer a new combination of them. Fuzzy quantifiers are applied for soft computing in modeling the optimism degree of the decision maker. In addition by using neat operators, the ordering of the inputs is not needed resulting in better computation efficiency.

Type-1 OWA operator provides us with a new technique for directly aggregating linguistic variables expressing human experts' opinions or preferences by fuzzy sets via OWA mechanism in soft decision making. However, the existing Direct Approach to performing type-1 OWA operation involves high computational overhead. In this paper, a fast approach, called α-level Approach, is suggested to implement the type-1 OWA operator. Experimental results have shown that the α-level Approach can achieve much higher computing efficiency in performing type-1 OWA operation than the Direct Approach.

We introduce a new generalization of the OWA operator called the induced linguistic generalized OWA (ILGOWA) operator. It is an extension that uses the main characteristics of three well-known aggregation operators: the IOWA, the LOWA and the GOWA operator. Therefore, in the same formulation, this operator uses linguistic information, generalized means and order inducing variables to reorder the arguments. Note that this generalization can be seen as a first step in the process of generalizing the LOWA operator with generalized means because other linguistic models can be considered. One of its main results is that it includes a wide range of linguistic aggregation operators such as the induced linguistic OWA (ILOWA), the induced linguistic OWG (ILOWG) and the linguistic generalized OWA (LGOWA). We further generalize the ILGOWA operator by using quasi-arithmetic means.

In this paper we introduce a generalization of Atanassov's operators relating to each interval-valued fuzzy set a family of fuzzy sets. We construct a class of operators and we show that, under suitable conditions, dimension two OWA operators can be obtained from Atanassov's operators.

A decision making technique is described for the selection among n alternatives based on the evaluation of n (distinct) group of persons according to the same m criteria. The evaluation of each person for each criterion is represented by a proportional ordinal 2-tuple and the overall opinion is aggregated by a pair of quantifier-guided OWA and P-OWA operators which can be accomplished alternatively by a Choquet integral of Fubini type.

This paper presents a new decision making procedure, the majority judgement, recently introduced by Balinski and Laraki, and contrasts it with the well-known 2-tuple representation model.

In order to deal with multiple attribute group decision making with incomparable linguistic preference information, some new kinds of linguistic-valued aggregation operators, namely, linguistic-valued ordered weighted averaging (LVOWA) operator and linguistic-valued hybrid aggregation (LVHA) operator are proposed. Based on the LVOWA and LVHA operators, an approach to multiple attribute group decision making is given.

Word clustering is useful for text information retrieval, text document classification, grammatical parsing and so on. This paper introduces a method to form self-organized map (SOM) of Chinese keyword by encoding them by the similarities of their contextual word sets, which are retrieved from within the centralized phrases rather than sentences to reduce the noise data. The experimental result shows that words can be clustered on the map according to both of their syntactic and semantic features.

An early warning system (EWS) is a timely surveillance tool to identifies potential crises and generate warning signals at a relatively early stage. This study aims to improve the learning functions of an EWS through training it using support vector machine (SVM) techniques. An adaptive pruning algorithm of SVM classification is developed which can improve prediction ability of EWS. This algorithm also can handle multi-data sources, multi-sensitive values, multi-indicators, and multi-crises issues in EWSs.

Software becomes a major part of today's enterprise business. Better managing risks on software development projects is vital to produce successful software systems. This study proposes a multi-criteria decision-making approach based on the Choquet integral for identifying, determining, and analyzing the potential and the most important software development risk factors from developers' perspective with an empirical case study from a Turkish industry.

We use conjunctions and fuzzy implications to define fuzzy dilations and fuzzy erosions in fuzzy morphology according to Nachtegael et al. 2003.¹ Further we observe that these fuzzy dilations and fuzzy erosions constitute fuzzy adjunctions that are also defined by a fuzzy implication. Adjointness between a conjunction and a fuzzy implication is analyzed. We show a conjunction that is adjoint to a fuzzy implication can be not only generated by an R-implication, but also by other fuzzy implications.

The paper gives an overview of the current nuclear activities in Belgium, emphasizing the research and service activities of the Belgian Nuclear Research Centre (SCK·CEN).

In his paper, Prof. Ye^[1] pointed out that the reduction approach introduced by Hu et al.^[2] will give wrong result in some situation. In this paper we come to a conclusion by analyzing that Ye's reduction approach is positive region reduction, and Hu's approach is to keep boundary region partition unchangeable virtually. Then, we analyze the logic characteristic of boundary region partition reduction standard.

Decision trees are considered to be one of the most popular data-mining techniques for knowledge discovery. Many approaches based on rough sets theory have been proposed for efficiently constructing decision tree. This paper presents a new approach for inducing decision trees by employing entropy and rough sets theory under characteristic relation. Examples show that the decision trees generated by the proposed method tend to have simpler structure than RC4.5.

In this paper, we construct a 6-element linguistic truth-valued intuitionistic fuzzy propositional logic (6LTV-IFP) based on lattice implication algebra. The implication operation of 6LTV-IFP can be deduced from four times implication of their truth values. Some logic reasoning properties of 6LTV-IFP are obtained then. Finally we proposed an approach for decision making using 6LTV-IFP.

We expand the classical model of a two-player game by inserting of fuzzy sets as payoff values in the game matrix. Players can thus formulate their payoff expectations with words instead of deciding on numerical entries of the matrix. In this way we count on the better verbal communication between players when designing the preliminaries of the game. As a final result we expect to obtain samples of the players' optimal strategies, which will preserve the profit of the game on the neutral level.

In this paper we demonstrate that artificial socially inspired agents play strategically a two-stage game, with asymmetric information, and replicate results obtained from experimental sessions with humans. The game is inspired in a negotiation supplier-client in two stages where there is not a priori bargaining power. Both sides can play strategically to get bargaining power and so get extra rewards from the expected payoff when trading on a good of low/high quality. Artificial agents are endowed with cognitive inspired mechanisms that evaluate the opponent's decisions to guess the opponent's social behavior: normative, altruist, cooperative or perverse. Each artificial player can not modify their assigned behavior in the game, but emotions lead their motivations to choose the fast and frugal heuristics that humans used in the experimental sessions, according to their own descriptions.

New methods for group ordering and classifying multi-attribute objects by inconsistent and contradictory preferences of many decision makers are suggested. These methods are based on the theory of multiset metric spaces. The proposed techniques are applied to real-life case studies: ranking companies and a competitive selection of projects, which are estimated by several experts upon multiple qualitative criteria.

In this paper we consider that a group of decision makers rank a set of alternatives by means of weak orders for making a collective decision. Since decision makers could have very different opinions and it should be important to reach a consensuated decision, we have introduced indices of contribution to consensus for each decision maker for prioritizing them in order of their contributions to consensus. These indices are defined by means of a consensus measure which assigns a number between 0 and 1 to each subset of decision makers. For putting in practice this idea, we have introduced a class of consensus measures based on distances on weak orders and we have analyzed some of their properties. We have illustrated the weighted decision procedure with an example.

In decision making problems dealing with linguistic information and multiple sources of information it may happen that the sources have different degree of knowledge about the problem then they provide their information in different linguistic term sets defining a multigranular linguistic context. Different approaches have dealt with this type of information that present different limitations. In this contribution we extend the structure of Linguistic Hierarchies in order to improve and make more flexible the management of multigranular linguistic information in Decision Making problems.

Fuzzy numbers do not always show a completely orderly group as can be done with real numbers. In applications of problems of multi-attribute decision making, when the final valuations are fuzzy, it is very difficult to distinguish the best possible alternative and the ranking of them. This paper develops an evaluation approach based on the Technique for Order Performance by Similarity to Ideal Solution (TOPSIS). In this approach, one step that is necessary to take into consideration is that related with the normalization.

The paper presents a simple and robust Multicriteria Decision Aiding Technique based on the use of statistical instruments. The objective is twofold: firstly, to provide a statistical tool to interpret enquiries on large samples of customers or clients facing multicriteria choices of some products or brands. Secondly, to help decision makers in making multicriteria choices in ranking alternatives corresponding to different preference profiles.

In this work we are interested in characterizing objective functions which do not change the set of efficient solutions (weakly efficient solutions, properly efficient solutions). Necessary and sufficient conditions for an objective function to be nonessential (weakly nonessential, properly nonessential) are presented. We establish relations between: weakly nonessential, properly nonessential and nonessential functions.

Real options valuation is a financial technique for evaluating investments under conditions of uncertainty, particularly uncertainty associated with market variables such as future product demand or the future value of an asset. The real option value of the investment opportunity is what a value-maximizing firm would pay for the right to undertake the investment project with its inherent decision points. This paper proposes a fuzzy multicriteria R&D project selection methodology based on the hierarchical TOPSIS, which includes a fuzzy real options valuation model.

In this paper, we develop a linear programming technique for multidimensional analysis of preference (LINMAP) method for solving multiattribute group decision making (MAGDM) problems with preference information on alternatives in fuzzy environment. Our aim is to develop a fuzzy LINMAP model to evaluate and to select of a waste treatment strategy for electrical and electronic equipments (EEE). EEE have already begun to accumulate at the garbage dumps. This garbage accumulation brings big danger to the environment and to the human health. That's why we should look for exploring the ways to dispose of these wastes. Waste treatment strategies contribute also to either local or global economies by creating a new sector and employment, and by reducing use of scarce resources.

When disasters occur completely at random, it means that the disaster entropy obtains the maximum value. Thus, we can apply the principle of maximum entropy to determine a probability distribution of the disaster loss series in a given period for a given area. It is easy to demonstrate that when the disaster entropy reaches its maximum value, the disaster loss series should follow the P-III distribution. In this paper, we introduce an application of the theory of disaster entropy in the risk analysis of disaster loss based on field disaster data.

The aim of this paper was to evaluation the relationship between annual severe grassland fire disaster and annual burned area. Using the historical data of grassland fire disasters in Northern China from 1991-2006, the grassland fire occurrence probability was obtained by applying information diffusion technology, and then the fuzzy relationship between the grassland fire occurrence and burned area was got in the provinces of Northern China by employing information matrix. It can provide an advantageous basis for fire fighting department to make a strategic decision to reduce the severe grassland fire disasters and establish a policy for disaster salvation.

It is important to predict financial failure risk in the changing Chinese economic environment. This paper proposes a novel genetic algorithm-based attribute reduction approach to improve performance of financial failure prediction model. The proposed approach is proved to outperform the exiting statistical methods in that it can achieve smaller attribute subsets with respect to support vector machine and higher accuracy.

Recent studies show that more than three quarters of enterprise resource planning (ERP) projects were judged as unsuccessful by the implementing firms. ERP implementation projects are fraught with risk. Because of its nature, risk includes imprecise and vague data. The aim of this paper is to identify the risk factors and evaluate the failure risk of a SAP/R3 implementation project. Fuzzy extended analytic hierarchy process (FEAHP) which is an efficient multi-criteria decision making tool to handle the imprecise and vague data is proposed. Triangular fuzzy numbers are used by the experts to construct fuzzy pair wise comparison matrices.

Joint Commission International (JCI) has been dedicated to improving the quality and safety of health care services. Today, the largest certification body of health care organizations in the United States, the Joint Commission surveys nearly 20,000 health care programs through a voluntary accreditation process. The Joint Commission is a none-profit organization. The JCI continuously improves the safety and quality of care in the international community through the provision of education and international accreditation. The JCI standard contains different subjects such as Prevention of Infection, Patient and Family Rights, Leadership, Facility Management, Safety.

Clients do not always appreciate differences among JCI consultants. This paper aims to provide an analytical tool to select the best JCI consultant providing the most customer satisfaction. The clients of three Turkish JCI consultancy firms were interviewed and the most important criteria taken into account by the clients while they were selecting their consultancy firms were determined by a designed questionnaire. The fuzzy analytic hierarchy process was applied to compare these consultancy firms. The means of the triangular fuzzy numbers produced by the customers and experts for each comparison were successfully used in the pair wise comparison matrices.

An early warning system (EWS) is critical to saving lives and mitigating loss from disasters. Literature addresses specific technical issues of EWSs in different hazard domains, however, only a few discussions on framework and standards. The paper proposes a set of practical designing standards and a comprehensive EWS integration framework. The framework takes considerations of human factors, lead-time and feedback issues, therefore more suitable for a wide range of applications in practice.

While we try our utmost to improve the performance of GMDSS (Global Maritime Distress and Safety System) equipments to reduce the increasing false alerts at sea, the ratio of false alert is stilI high. It has been making bad effect to the safety at sea and wasting the resource of SAR (Search and Rescue). Therefore it is necessary to make an alerting processing system to identify the false alert rapidly before taking SAR action. This paper designs a Marine Alert Processing System based on intelligent fusion and attribute theory. MAPS can identify alert rapidly and exactly and sends the result to SAR decision support system at RCC.

Road safety performance indicators have recently been proposed as a useful instrument in comparing countries on their performance of road safety risk factors. New insights can be gained in case one road safety index is composed of all risk indicators. The safety performance can be evaluated, countries can be ranked, trends identified and the impact of measures assessed. However, the aggregation process is still unclear in this context. In this paper, the use of ordered weighted averaging (OWA) operators will be experimented for an evaluation of road safety performance indicators. More specifically, several basic and more advanced aggregation operators will be applied to our indicator data set and the final index scores are then compared to the number of road fatalities per million inhabitants. It is demonstrated that compensation should not be allowed too much in the road safety context. All indicators should be incorporated in the final index to some extent and weaker performances should be stressed more.

The Double Traveling Salesman Problem with Multiple Stacks (DTSPMS) consists on finding the minimum total length tours in two separated networks, one for pickups and one for deliveries. One item is required to be sent from each location in the first network to a location in the second network. Collected items can be stored in several LIFO stacks, but repacking is not allowed. In this paper we present four new neighborhood structures for the DTSPMS, and they are embedded, together with other two existing ones, into a Variable Neighborhood Search heuristic that is used to solve the problem.

This work focuses on the multi-objective scheduling of reentrant lines. Three algorithms are tested : our Lorenz - Non dominated Sorting Algorithm (L-NSGA), the Strength Pareto Evolution Algorithm 2 (SPEA2) and the Multi-Objective Ant Colony System (MOACS). Numerous experiments on various configurations of the system are made and finally their efficiency are ordered following two measures.

According to Schema Theorem, the larger the fitness value of a schema is, the higher the chance of the sub-space corresponding to the schema being chosen for searching is. Therefore, the coding of a genetic algorithm should be designed to produces short building blocks at as more fixed positions of the strings as possible. In this paper, an upper and lower bounds of the fitness values of order-n building blocks are given in terms of the maximal fitness values of order-1 building blocks. It is shown that, if a linear-weighted coding supplies order-1 building blocks at most of the loci of the strings, then there is sure to be remarkable fitness differences among high order schemata. In such case, the trajectory of the fitness function contains peaks, and diversity loss is likely to occur in the local search within these peaks.

The aim of this article is to analyze and explain the results obtained when using a genetic algorithm (GA) to optimize the ascent trajectory of a conventional two-stages launcher. The purpose of this optimization is to obtain a final stable circular Low Earth Orbit (LEO) with maximum height. A subset of parameterized variables are optimized by the GA in order to achieve the mission purpose while satisfying the constraints and maximizing the final height of the orbit, given the technical characteristics and limitations of the launcher, and a fixed payload weight. Besides, an adaptive mutation rate has been used that broadens the search space whenever a population becomes uniform, avoiding local maximums. The results have been proved to be successful and useful.

In this paper different genetic algorithms (GAs) designs are evaluated to estimate internal parameters of broadband piezoelectric transducers. Several operational aspects and parameters of the GA algorithm are considered and assesed by computer simulations in order to seek the best performance in the estimation procedure. This procedure was applied to a practical 1 MHz transducer, and five of their constructive and design parameters were estimated. The use of GAs to find multiple optimal solutions for modelling noisy data is demonstrated.

In this paper, the Facility Layout Problem (FLP) is solved by using a hybrid method that combines a genetic algorithm (GA) and an Ant Colony Optimization (ACO) mixed with a Guided Local Search (ACOGLS). The GA is used to solve the Group Technology Problem (GTP) which groups simultaneously machines and products in cells. The particularity of our method is to use a production volume as input data. ACOGLS solves a QAP using GTP solution.

This paper introduces first the concept of distance density, and then proposes a new hybrid genetic algorithm based on distance density and quasi-simplex technique (HGABDDQT). HGABDDQT produces the offspring using the genetic operations and the quasi-simplex technique in parallel. In genetic operations, the crossover probability is determined adaptively by distance density, the mutation probability is determined adaptively by distance density and fitness. No binary encoding/decoding in mutation and crossover operations. HGABDDQT algorithm has been implemented and tested on typical benchmark functions. The experimental study has shown that HGABDDQT is more effective than the competitive algorithm in finding the near global optimal solutions.

Churn management and prediction systems' main goal is to determine churners who wish to switch another GSM (Global Services of Mobile Communications) operator for getting more optimum benefits and services. Churn management systems determine patterns to promote the subscribers and prevent them to be lost to another operator. In this study, ANFIS (adaptive neuro fuzzy inference system) and genetic approach based systems are used to determine churners. First classification step starts with parallel Neuro fuzzy classifiers. After then, FIS takes neuro fuzzy classifiers' outputs as input to make a decision about churner activity. Optimization process can be provided by using genetic algorithms to make fine and tuning in fuzzy process.

The area of assets Prognostics and Health Management (PHM) represents a promising application field for Soft Computing (SC), an area that provides a broad repertoire of techniques for solving requirements of typical PHM problems. To better understand these requirements, we leverage a decision-making framework, composed by the cross product of the decision's time horizon and domain knowledge used by SC models. We use this framework to analyze how SC is used to perform anomaly detection, anomaly identification, failure mode analysis (diagnostics), estimation of remaining useful life (prognostics), on-board control, and off board logistics actions. We illustrate this concept with a case study in anomaly detection, based on different SC technologies.

In this paper, a decision support system (DSS) based on fuzzy information axiom (FIA) is developed in order to make decision procedure easy. The system consists of knowledge base module including facts and rules, inference engine module including FIA and aggregation method, and user interface module including entrance windows. The main aim of this paper is to present a DSS tool to help the decision makers to solve their decision problems by modifying data-base of the program.

Bilevel decision techniques are developed for decentralized decision problems, which may be defined by fuzzy coefficients. Based on a fuzzy linear bilevel (FLBL) model and two FLBL algorithms, this research develops a FLBL decision support system (FLBLDSS). It first introduces a satisfactory-degree-adjustable FLBL model. Then, the system structure and function modules of this FLBLDSS are presented. Finally the key algorithms are illustrated.

This article continues the research line of the authors on knowledge extraction and verification of Rule Based Expert Systems (RBES) using algebraic inference engines based on Gröbner bases theory. Now a shell, that includes a graphic user interface and inference engines for different logics (both classic and modal many-valued), is presented. The shell distinguishes three levels: in the lower level, we provide the computer algebra system code of the algebraic inference engines; in the intermediate level, the RBES developer has to detail the rules and integrity constraints of a certain RBES; and, finally, in the upper level, the end user deals with a simple GUI, where he can perform knowledge extraction or verify the RBES, after choosing the logic and inputing a consistent set of facts. We believe that this shell can be really useful for teaching and for quick RBES designing.

In this contribution we present some working implementations of a mobile group decision making support system that allows to deal with experts' preferences expressed by means of fuzzy preference relations. The system supports incomplete information situations, that is, the situations where the experts do not give all the preference values that they are usually requested. Several implementations have been created in order to compare the advantages and disadvantages of different existing mobile technologies in both the client and server sides of the system.

A critical issue in clinical decision support system (CDSS) research area is how to represent and reason with both medical domain knowledge and clinical symptoms to arrive at reliable conclusions even when under uncertainty. This paper describes how to apply a recently developed generic rule-base inference methodology using the evidential reasoning (RIMER) approach to model clinical domain knowledge and clinical inference process in a CDSS. A simple case study is employed to illustrate the new belief rule-based CDSS, and the result shows that the proposed CDSS is capable of modeling and reasoning with both clinical domain knowledge and clinical symptoms under various types of uncertainties. Moreover, the diagnosis results generated by the CDSS can be used to rank the severity of patient cases.

Experiment design often involves multiple objectives and uncertain data in its optimizing process. Fuzzy multi-objective linear programming (FMOLP) is an appropriate method to handle this problem. For the case of modeling nonwoven-based resilient product experiment design, we first obtain the parameter values of objectives functions from a set of experiments. As these parameter values are in the form of distributions, we transfer them into membership functions and then formulate a FMOLP model. We finally develop a fuzzy multi-objective decision support system (FMODSS) to present this model and get a solution of this problem.

Different intervention strategies, including nutrient abatement and the construction of facilities to increase the water exchange between the lagoon and the outside sea, are evaluated by means of an additive multi-attribute utility model to achieve good water quality and good conditions for waterfowl in the Ringkøbing Fjord lagoon (Denmark). A decision support system, called the GMAA, is used. This system is intended to allay many of the operational difficulties involved in the decision analysis cycle is used. The system is an efficient tool for taking into account the uncertainty of the various components of the analysis and enables an analysis of the difference between an anthropocentric and an ecocentric view of the problem. It implements what is known as decision making with partial information, which takes advantage of the imprecise inputs.

This paper proposes a linguistic performance appraisal from a competency management perspective, where there are different sets of reviewers taking part in the evaluation process that have a different knowledge about the evaluated employees. The reviewers can express their assessments in different linguistic domains according to their knowledge. The proposed method will conduct each linguistic label provided by reviewers as a fuzzy set in the common domain to compute collective assessments that will allow to the management team to make their decisions about employees.

In this paper we present a decision support system for primary action of international organizations devoted to natural disaster relief. In particular, we pretend to build up an expert system that taking into account past experiences will help decision makers, mainly non-governmental organizations, to start or not an operation, depending on the place and the very first information about a possible natural disaster. The relevance of this issue is extreme, since such a decision must be taken as soon as possible.

In recent years, natural and man-made disasters have been affecting increasing numbers of people throughout the world. Organisations for emergency and humanitarian aid have experienced an important growth, and efficiency in management becomes crucial. There is a lack of specific tools devoted to logistics of this special kind of interventions in developing countries, demanded by the organisations. A goal programming model that sustains a decision support system currently in development is presented, focusing on the transport problem to distribute humanitarian aid to the affected population of a disaster in a developing country.

The paper addresses a hybrid technology upon that a decision support system (DSS) for disaster response is built. The hybrid technology combines intelligent technologies of profiling, ontology engineering and management, context management, constraint satisfaction, and Web Services. The application of the hybrid technology is illustrated by an example of real-time resource coordination for logistics management during fire response operations.

Unhealthy nuclear intentions of "rogue states" have challenged policy decision-makers for many decades. This paper suggests a new computationally intelligent approach based on the DiNI (Discerning Nuclear Intentions) Model to support policy decision-making in a complex, non-linear manner. DiNI is inspired from the well-validated theme of socio-biological evolution of altruistic behavior and works on a NetLogo platform. By changing the model parameter values in DiNI, policy makers can objectively assess i) the nuclear intentions of states based on the current policy situation ii) the effectiveness of new policy instruments e.g., incentive packages on the nuclear intentions of suspect states. Simulation conducted on DiNI using hypothetical data demonstrates that DiNI can indeed be used to proactively inform policy decision-makers on unhealthy nuclear intentions of suspect nation states.

Trust modeling is a challenging issue due to the dynamic nature of distributed systems and the unreliability of self-interested agents. In this context, the Agent Reputation and Trust (ART) Testbed has been used to compare trust models in annual Spanish and International competitions from 2006. In this paper we describe the agent we have presented to those competitions. This agent is an extension of a previously-published trust model AFRAS that used fuzzy sets to represent reputation. In addition this model we propose a cognitive model to implement adaptive behaviors. An implementation of this extension of AFRAS trust model has participated in the (Spanish and International) 2006 ART competitions.

The goal of this contribution is to present a computer-based application of an Adaptive Consensus Support System that deals with heterogeneous information. This application may be used to carry out consensus processes in Group Decision Making Problems defined in heterogeneous contexts. It allows experts to express their opinions using multiple expression domains in order to bring decision situation closer to real-word problems. In addition, the implemented consensus process is adaptive, i.e, it can adjust its behavior depending on the level of agreement reached in each consensus round, suggesting a greater number of changes when the agreement is far, and decreasing it when the consensus becomes nearly.

Collaborative Recommender Systems (CRS) are very useful tools that help people to select items in a huge search space, based on the idea that people with similar taste of preferences in an area make similar decisions concerning to that area. There are many commercial applications that show the utility of these systems. However, there exist areas in which these systems have not applied yet. In this contribution we want to introduce OrieB, a CRS working in the Academic Orientation domain in order to support advisors helping students of secondary school to make decisions about their academic future. OrieB will use students' marks as input data in order to suggest their academic possibilities by providing several recommendations using the fuzzy linguistic approach.

Recommender systems are tools whose objective is to evaluate and filter the great amount of information available to assist the users in their information access processes. In this paper, we present a recommender system for research resources based on fuzzy linguistic modeling, promoting the collaboration between several research groups.

Electronic shops provide an excellent choice to buy without leaving home. Nevertheless, people frequently have problems to find what they look for because of the wide range of items that e-shops offer. Recommender Systems are applications that help people in their searches in e-shops. They deal with information that people provide which is usually related to opinions, tastes and perceptions, and therefore, it is difficult to express them by means of precise numeric scales. Fuzzy linguistic approach provides a better way to express this kind of information. In this contribution we propose a Knowledge Based Recommender System that uses the fuzzy linguistic approach to handle the uncertainty of the human opinions and provides a multigranular context that allows users to utilize the term set that better fits with their degree of knowledge.

An idea of a new model of a Web-based system to support a group of decision makers in reaching consensus is proposed. The core of the system is a flexible human consistent representation of preferences. However many means, notably Web-based, are conceived to help a group member to form and express his or her preferences. Moreover the model integrates the mechanisms for the discussion guidance so as to smoothly and effectively run the session.

In this paper we propose a Subjective Logic - based framework which allows composition and rating of semantically described web services. The framework is intended to be applied in so-called SOKU (Service-Oriented Knowledge Utility) environments and involves data fusion from two different sources (the descriptions of services and user ratings assigned to them).

A Content-Based Recommender System lets to suggest new products to the user taking into account the likeness with the content (description) of other products that the user rated before. In this paper we present a recommender system that is capable of incorporating knowledge about the structure of the content of the products and help it to improve the recommendations. We use Bayesian Networks for modeling the structure of the products and the relations between them and the system users.

Generating personalized recommendations for new users is particularly challenging, because in this case, the recommender system has little or no user record of previously rated items. Connecting the newcomer to an underlying trust network among the users of the recommender system alleviates this so-called cold start problem. In this paper, we study the effect of guiding the new user through the connection process, and in particular the influence this has on the amount of generated recommendations. Experiments on a dataset from Epinions.com support the claim that it is more beneficial for a newcomer to connect to an identified key figure instead of to a random user.

This work achieves a system for supporting the user during the web services discovery. The approach allows end-users to model an ad-hoc service request by filtering semantic specifications rather than the exploitation of strict syntax formats. Adaptive hypermedia techniques assist the user in the formulation of a web service request, exploiting the semantic annotation of the browsed web resources. This annotation reflects concepts or ontological terms that are relevant for the user services request formulation. The system reply is a list of semantic web services that match the input request, by specifying the input and output concepts.

This contribution addresses the problem of expressing preferences among nonfunctional properties in a Web Service architecture. In such a context, semantic annotations are needed and added on service declaration and business process in order to select the best available service. These conditional and unconditional preferences are managed using Conditional Preference-Networks (CP-Nets). But in several cases, uncertainty related to the preferences has to be taken into account to achieve a better satisfaction rate. We propose the use of fuzzy linguistic information inside the whole process when it will be necessary.

In this paper an introduction to the citizen participation in electronic government is presented. A new participation model is described through mobile technologies and linguistic information. This model can produce new environments where to increase the citizen contribution in the government of cities and countries.

In this paper we briefly characterize the application of ontologies in multilevel information fusion. Then a scenario of multi-user environment with private ontologies is considered. We propose a social network-based ontology alignment scheme which aims to maintain knowledge consistency between communicating users thus ensuring effective collaboration and obtaining consistent results during information fusion tasks.

It has been observed that there is a substitution effect from website information flow to related human decision making and then the movement (human flow) in their social activities. This paper aims to reveal the general characters of how website information flow acts on human decision and human flow in geo-space. It uses collected statistic data in CNNIC, proposes and calculated substitution function and enforcement function, and then conducts a comparison between the two functions. The result shows that website information flow plays a determined role to guide related human decisions and their realistic movement from one dimension to multiple dimensions.

In several service businesses, web sites are almost required to establish a connection between the customer and the service provider. Web site quality also represents the business quality in domains such as e-commerce, finance, etc. Thus, an evaluation of the web sites is necessary to measure the performance of the business. In this paper, an axiomatic design based on fuzzy group decision making is adopted to evaluate e-learning web sites' quality.

Thermonuclear plasmas are complex and highly non-linear physical objects and therefore, in the most advanced present day devices for the study of magnetic confinement fusion, thousands of signals have to be acquired for each experiment, in order to progress with the understanding indispensable for the final reactor. On the other hand, the resulting massive databases, more than 40 Tbytes in the case of the JET joint Undertaking, pose significant problems. In this paper, solutions to reduce the shear amount of data by different compression techniques and adaptive sampling frequency architectures are presented. As an example of methods capable of providing significant help in the data analysis and real time control, a Classification and Regression Tree software is applied to the problem of regime identification, to discriminate in an automatic way whether the plasma is in the L or H confinement mode.

A validation tool for component monitoring and support of condition-based maintenance for technological systems has been developed and applied to heat exchanger validation and trend evaluation going beyond classical data reconciliation methods.

On-line sensor monitoring allows detecting anomalies in sensor operation and reconstructing the correct signals of failed sensors. Since in field applications the number of signals to be monitored is often too large to be handled effectively by a single reconstruction model, a more viable approach is that of decomposing the problem by developing a number of reconstruction models, each one handling an individual group of signals. In this paper, Multi-Objective Genetic Algorithms (MOGAs) are devised for finding the optimal groups of signals used for building reconstruction models based on Principal Component Analysis (PCA). A weighted scheme is adopted to combine appropriately the signal predictions of the individual models. The proposed approach is applied to a real case study.

This paper extends a method previously introduced by the authors for building a transparent fault classification algorithm by combining the fuzzy clustering and fuzzy logic techniques. The extension allows the treatment of splitted clusters. A numerical application is presented with regards to the fault classification in the Steam Generator of a Nuclear Pressurized Water Reactor.

The proper and timely fault diagnosis is of premier importance to guarantee the safe and reliable operation of nuclear power plants (NPPs). In this paper, fuzzy inference system is adopted for the diagnosis of abrupt faults in a nonlinear model of a typical pressurized water reactor (PWR). The fuzzy system is tested with different shape of membership functions (MFs). The if–then rules, representing the underlying processes are inferred from the available fault-symptom relations. The symptoms are generated using plant model measurements.

Fuzzy logic, neural networks and genetic algorithms are three popular artificial intelligence techniques that are widely used in many applications. Due to their distinct properties and advantages, they are currently being investigated and integrated to form new models or strategies in the areas of system control. This paper presents an adjustment strategy for a dual-fuzzy-neuro controller (DFNC) in the gas-fired water heater control. In this method, strategies to adjust the DFNC in accordance with the environment dynamics are automatically generated in off-line manner using genetic algorithms (GA). The generated strategies are stored in a neural network and used for adjusting the DFNC on-line. Therefore, the DFNC is automatically adjusted in accordance with the unknown dynamics of an environment using the generated strategies which are stored in the neural network. Fuzzy fitness evaluation method is proposed for the effective evolution of the neural network in the GA process.

Control charts are very useful tool for monitoring and analyzing the processes. When the quality characteristics cannot be represented in numerical form, like softness, appearance, chapped, damaged, etc., control charts for attributes are used. Especially, the binary classification into conforming and non-conforming of product/parts for implementing the p-control chart includes ambiguity or vague, or lack of available information due to process or human subjectivity. Traditional p-control chart is not a sufficient tool to consider these uncertainties and vagueness of data. In this case, the fuzzy set theory is a very useful methodology for dealing with sources of uncertainty or imprecise conditions. In this study, the methodology for constructing fuzzy p-control chart based on fuzzy median transformation method for both constant and variable sample size (n) is proposed by using α-cuts. Here, α-cuts approach to the fuzzy -control chart provides the ability of determining the tightness of inspection. Thus, the flexibility of control limits is achieved by incorporating fuzzy set theory and control limits calculations.

The fuzzy set theory is a powerful method to analyze the statistical data which includes ambiguity or vague comes from the structure of the process, measurement systems or environmental conditions. Crisp value collected process can transform the fuzzy numbers (a,b,c) by using the membership functions and calculate fuzzy control limits by using the traditional control limits equations. Thus, the flexibility on control limits can be achieved by analyzing the process like "in-control" or "out of control". The regression control chart is used especially to evaluate the tool wearing problem in industry. In the traditional regression control chart, all data assume crisp value. With fuzzy set theory, the fuzzy regression control chart can be handled based on α-cuts approach by using the fuzzy midrange transformation techniques. In this study, the theoretical structure of α-level fuzzy midrange for α-cuts for fuzzy -regression control charts and fuzzy control chart are proposed.

This article presents new stabilization conditions for discrete Takagi-Sugeno (T-S) systems by using the sum of squares (SOS) decomposition. These conditions are stated in terms of linear matrices inequalities (LMIs) and they are more relaxed than previous works. They have been applied to an example which gives proof of their less conservative behavior.

The paper briefly formulates the error loop as a tool for designing robust stability control systems in front of structured and unstructured uncertainties. In turn, the error loop formulation shows the main tool for accommodating such uncertainties is the noise estimator, which is the unique feedback channel from plant to control.

The design, construction and real time performance of a mobile monitoring system based on a Khepera robot are presented. For the robot navigation, a cascade fuzzy control algorithm is implemented. Fuzzy sensor fusion related to the perception of the environment has been used to reduce the complexity of the navigation function. The identification of test points is carried out by means of a Kohonen neural network. Finally, one-dimensional image processing is used for the recognition of landmarks located at each test point.

This paper presents an improvement for the software implementation (MOFS) of a user adaptive fuzzy control system for autonomous navigation of mobile robots in unknown environments. This improvement consists of a priority areas definition where the environment is measured by a PLS laser sensor, in order to get a reduction in the number of fuzzy rules and also in the computational cost, and hence obtaining improvements in the trajectory. This system has been tested in a pioneer mobile robot and on a robotic wheelchair, odometry sensors are used to localize the robots and the goal positions. The system is able to drive the robots to their goal position avoiding static and dynamic obstacles, without using any pre-built map. This approach improves the way to measure the danger of the obstacles, the way to follow the walls of corridors and the detection of doors. These improvements reduce the zigzag effect of the previous system by making the trajectories significantly straighter and hence reducing the time to reach the goal position.

This paper presents a path planner for Unmanned Air Vehicles (UAVs) based on Genetic Algorithms (GA) that obtains a feasible and optimal 3-D path for the UAV. It uses 9 different objective values which are calculated with a realistic model of the UAV and the environment and which are structured with 3 levels of priorities. Our planner works globally offline as well as locally online, which means that the algorithm can recalculate parts of the generated path in order to avoid unexpected risks. Finally, the effectiveness of the solutions given by this planner has been successfully tested against a simulator that contains the complete model of the UAV and the environment.

Choosing an adequate Human Reliability Analysis method for the safety assessment of a particular system is a difficult problem. Comparison of the available methods is best done by crew simulation in virtual control rooms. However, the quantitative evaluation of the outcomes of these simulations in terms of crew performance and human error probabilities is not trivial. In this paper, a fuzzy logic method is proposed for systematically assessing crew performance. The feasibility of the method is proved on a case study concerning a scenario of incomplete scram in a Boiling Water Reactor.

Although internal rate of return (IRR) has been studied extensively, addressing uncertainty in cash flows for the calculation of IRR has been relatively neglected. In this study, we propose an optimistic and a pessimistic fuzzy relation based approach to IRR, and define the investment decision procedure. Finally, we apply the proposed procedure to a simple project example, and investigate the PV-consistency for the decision.

A model for text analysis based on a real bag is developed in which a novel concept of bag neighborhood is used which takes nearness of words in a text sequence is taken into account. A condition such that a bag neighborhood defines a kernel that has been used in support vector machines is given. As a result kernel-based methods of data analysis can be applied to this model. Application to a set of medical incident report is studied using an algorithm of medoid clustering and principal component analysis.

Real estate valuation is a complex problem where experts make their job subject to an input information that is mostly fuzzy in nature. In this paper we present a real estate valuation model that allows the management of such a fuzzy information. In particular, we realize that our reference set of experts considers at a first stage a division of all potential clients into four main clusters, according to their preferences. An inference engine is then designed in order to obtain the degree as valuation of a real state described by the client and their own characteristics. A relevant feature of our proposal is that we allow fuzzy specifications.

Energy investments, that have a large portion in the world economy, have immense uncertainty. Traditional valuation methods are less viable in that situation and other methods which can minimize uncertainty become more important. In this study, firstly discounted cash flows (DCF) analysis, which is one of the traditional valuation methods, is compared to real options valuation (ROV) which decreases the uncertainty. Next, fuzzy discounted cash flows (FDCF) analysis and fuzzy real options valuation (FROV) are applied to the same oilfield data. In conclusion the results of these four valuation methods are compared.

The vagueness of multi criteria decision making (MCDM) is commonly handled through fuzzy sets theory, by assigning degree of membership. However, the spatial MCDM (SMCDM) problem encounters ambiguity in assigning the membership function to fuzzy pairwise comparisons, which is referred to as non-specificity. This paper attempts to reach a new method in making the comparison matrices in analytical network process (ANP) approach to consider some aspects of uncertainties and vagueness in the process of SMCDM based on the rule and logic of intuitionistic fuzzy (IF). The case study undertaken in this research includes investigation of criteria used in selecting the public parking lots in a part of Tehran metropolitan area. This paper successfully demonstrates that the proposed method has the ability to include a number of geospatial data and constraints to improve reliability of decision making using the Intuitionistic Fuzzy Analytical Network Process (IF-ANP) in a GIS environment.

In this paper we present a method to rate and rank regions for landing site selection, assuming that Spacecrafts are equipped with cameras that take pictures during the descent phase on planets. In the final stages of landing on a planet, we have to consider that the spacecraft size covers more than one pixel on a target image; hence instead of selecting pixels we have to select regions for safe landing. The rating method includes defining the set of regions to be classified (alternatives) and then uses the uninorm operator to determine the rating of each region, thus ensuring a full-reinforcement behavior for the regions rating process.

Recent research on highly industrial distributed control methods for complex systems has produced a series of new philosophies based on negotiation, which bring together process engineering with computer science. Among these control philosophies the ones based on multi-agent systems (MAS) have become especially relevant. However these MAS models have the drawback of an excessive dependence on up-to-date information about the products, elements or information systems. Radio Frequency Identification (RFID) can help solve these problems at a physical level. RFID enhanced MAS have been proven effective at plant level. The information obtained can also help in superior managerial levels. In this paper a new strategy is presented to help define the interface between physical a managerial levels in MAS platforms and societies. This new approach is analysed for the general case of manufacturing plants, but is made to the handling processes in a new airport.

Radio frequency identification (RFID) technology introduces the opportunity for increased visibility by facilitating easy tracking and identifying of goods, assets and even living things. However, high investment cost and inadequate technical capability remain as challenges for RFID system implementations. That being the case, fair evaluation of savings associated with increasing performance and investment costs has a great role in the success of RFID projects. In this study, a systematic framework for the economic analysis of RFID investment is proposed. The increment of order is determined in terms of delivery accuracy and delivery time via a fuzzy rule based system. A case study is constructed on the basis of expert conception to illustrate the proposed method.

While organizing cultural events, local authorities are able to make the most of the opportunity and promote the whole tourist destination, thus attracting new visitors. In order to find out the impact of each cultural event on tourist activities near the main destination, a RFID based action-tracking framework has been defined. The RFID based action tracking system works as a data retrieval system from key points within the destination area. To address the decision related to the technical issues and to define the logical architecture of the RFID based framework we propose to apply the Fuzzy Analytical Hierarchy Process Scoring (FAHPS) Methodology.

The objective of the present study is to select the most suitable location for wind power station through fuzzy AHP method. The problem of selecting location for wind power station was considered as a multi-criteria decision making problem. Essential and sub-criteria were specified and location selection was expressed as a hierarchic structure. Then, the problem was solved by using fuzzy AHP method and the best location for the power station was determined.

In this paper, we proposed a model for selection of the most suitable city for a new nuclear power plant in Turkey by using fuzzy analytical hierarchy process and fuzzy analytical network process due to linguistic terms. The implementation of the system is demonstrated by a problem having four stages of hierarchy which contains four criteria and twenty attributes. The study compares the fuzzy analytic hierarchy process and fuzzy analytic network process results.

Particle Swarm Optimization (PSO) is a metaheuristic technique based on socials aspects of intelligence. Some PSO models have been developed for combinatorial optimization, although none of them presented satisfactory results to optimize the combinatorial problem of the Nuclear Reactor Reloading Problem (NRRP). In previous research we developed the Particle Swarm Optimization with Random Keys (PSORK) model, to be applied to combinatorial problems such as the NRRP. In this paper, we survey the confinement of particles in the PSORK. A confinement analysis is interesting in continuous functions optimization since it may influence it, resulting in biases, favoring particular regions of the search space, although there are no similar studies for combinatorial optimization. We have submitted PSORK to a confinement analysis when applied to the Traveling Salesman Problem Rykel48 (ry48p), a benchmark for combinatorial optimization. We also present results for the optimization of Angra 1 Nuclear Power Plant reload. We have tested non-confined PSORK as well as Standard, Hyperbolic and Random Back confinements.

A novel algorithm is developed for solving the two-level stochastic programming in electricity market based on complementary optimization and linear equation sets. A simulated electricity market with six participating generation companies and IEEE 30-bus test system is served for demonstrating the feasibility and efficiency of the developed algorithm.

In this paper a new artificial immune system (AIS) algorithm is proposed to solve multi objective fuzzy flow shop scheduling problem. The objectives are considered to be minimizing the average tardiness and the number of tardy jobs. The developed new AIS algorithm is applied in an engine cylinder liner manufacturing process. The feasibility and effectiveness of the proposed new AIS is demonstrated by comparing with the Genetic algorithm. Computational results demonstrate that the proposed new AIS algorithm is a more effective meta-heuristic for the multi objective, flow shop scheduling problem with fuzzy processing time and due date.

The recent innovations in marine technology ensure providing great number of opportunities for shipbuilders especially in equipment selection. However, a critical decision-making problem has arisen to manage the optimum choices and integrity in marine system design. This paper proposed Fuzzy Information Axiom (FIA) based decision-making systematic which aids to primarily design layouts for carrying out shipbuilding projects. As a case application, the principal components of the compressed air system are comparatively assessed in order to identify optimum configuration. The proposed FIA-based approach provides invaluable supports for enhancing the communication network between the shipyards and marine equipment manufacturers.

In this paper we use an interval valued fuzzy similarity in the stereo matching problem. An algorithm of fuzzy thresholding is used to find the best membership function to fuzzify the images and we use an interval generator to construct IVFS from the fuzzy sets that represent the images.

In this paper, we address a method for terrain aided navigation of aerial vehicles based on vision based sensors which estimate the state of the vehicle (position and attitude) using sequential aerial images. The task is to localize an airborne platform by real-time comparing and matching of an aerial image with a database consists of satellite imageries and digital terrain models (DTM) precisely aligned to geo-coordinates. Such robust matching algorithm is formulated as a rule-based fuzzy image matching algorithm using salient features and descriptors. The corresponding features are then used in a mapping transformation to determine the position and orientation of the aerial vehicle in standard navigation coordinate system. Several experiments were run on aerial image sequences, testifying the robustness and good performance of the implemented matching and pose estimation methods.

Managing the verification of primarily design & installation projects for ship machinery systems is one of the crucial stages in shipbuilding process. Especially, the design of operator-system interfaces such as remote controls, displays, alarms, workstations, and labels makes required a high level of technical expertise and systematic control. This paper focuses on structuring Ship Design Project Approval Mechanism (SDPAM) based on Fuzzy Axiomatic Design (FAD) methodology towards system design of relevant interfaces on engine room control consol. General characteristics, accessibility, operational requirements, and ambient environment are guided as hard constraints for this complex problem. Hence, the proposed systematic requires performance assessment using fuzzy values. The outcomes of this study originally contribute the existing approval procedures of classification societies for shipbuilding projects.

In this paper a fuzzy c-means (FCM) based cell formation (CF) algorithm is used to design cellular manufacturing in a tractor manufacturing firm. In cell formation problems as the size of the part-machine matrix increases the problem gets complicated. Fuzzy models are suitable for CF problems by allowing the representation of uncertain information. FCM based algorithms seem to be efficient for cell formation problems at cellular manufacturing design.

This paper addresses the problem of car headlight lens inspection. First, the currently quality control of lenses with the defect characterization is presented. Second, a vision-sensor-planning system is developed. This system utilizes the CAD information of the headlight lens and the camera model to plan camera viewpoints. The desirable sensor poses are achieved by a genetic algorithm. To improve the performance of the system the customer requirements and the skill of the human inspector are included through a fuzzy system.

Fabric-hand evaluation is one of the key features and measures in textile material selection for fashion design. Fabric-hand evaluation requires considering multiple criteria with in a group of evaluators. The evaluation process often involves fuzziness in the weights of criteria and the judgments of evaluators. This study first develops a multi-level textile material fabric-hand evaluation model. It then proposes a fuzzy multi-criteria group decision-making (FMCGDM) method for the evaluation. A fuzzy multi-criteria group decision support system (FMCGDSS) is developed to implement the proposed method and applied in textile material fabric-hand evaluation.

The application of ant colony optimization for assembly lines design problem consisting of selecting machines for stations and determining the capacities of intermediate buffers is the main subject of this paper. The objective is to find the best configuration which maximizes the throughput rate of the line while having a maximum allowed budget. A hybrid ant colony optimization approach coupled with a guided local search is then applied to enhance the performances of the application. Numerical results show that the proposed algorithm performs optimally.

In this paper, we propose an intelligent system for extracting fashion design elements and modeling complex relations between concrete technical design elements and abstract fashion design elements. This system can effectively support decisions of fashion and garment designers in two aspects: 1) decompose abstract fashion design elements used by fashion designers as their design objectives into concrete design elements for garment design and production; 2) deduce or predict fashion messages implied in garments from concrete technical design elements. In this intelligent system, decision trees and fuzzy logic have been applied for finding inferring rules from a learning base. Learning data related to fashion design elements and technical design elements are collected from various designers by evaluating a set of representative fashion images. The effectiveness of the proposed system has been effectively validated by several specific design examples.

Training systems based on virtual reality are used in several areas. In these systems the user is immersed into a virtual world to have realistic training through realistic interactions. In such training is important to know the quality of user's training. An online assessment system allows the user to improve his/her learning because it can identify, immediately after the training, where he/she committed mistakes. In this paper, we present a new approach to online training assessment based on Gaussian Naive Bayes, a generalization of Naive Bayes Networks, for modeling and classification of simulation in M pre-defined classes.

The authors suggest improvement of validity, efficiency and usefulness of voice stress analysis in several aspects namely: a sufficient sample of people should participate in the VSA research; historic experience of a specific area should be used; intelligent systems should be used to make a thorough analysis; and intelligent systems should be integrated with contemporary VSA measurement and analysis methods and tools. The aforementioned aspects were implemented in practice when developing the Web-Based VSA Decision Support System for e-Examination (VSA-DSS-E). In order to demonstrate the validity, efficiency and usefulness of the developed VSA-DSS-E, the article also presents a case study.

Statistical mechanics helps to estimate corrections to the entropy and energy of the fluid with heat flux in terms of the nonequilibrium distribution function, f. This leads to the coefficients of wave model of heat: relaxation time, propagation speed and thermal inertia. With these data a quadratic Lagrangian and a variational principle of Hamilton's type follows for the fluid in the field representation of fluid's motion. We analyze canonical conservation laws and show the satisfaction of the second law.

The following sections are included:

• AUTHOR INDEX

• ACKNOWLEDGEMENT TO REVIEWERS