This paper is concerned with the use of fuzzy reasoning in predicting breath alcohol concentration(BrAC)after drinking. Fuzzy reasoning is adopted to resolve the difficulties in concentration curve, where weight and age affect the maximum value and the time history of BrAC after drinking. The prediction model applied uses the human weight, age and body mass index(BMI)as input variables, and the impulse response functions with time-delay as the output in the fuzzy control rules. The deffuzified value of the output from eight kinds of function proposed in this paper could estimate weighted averages of all the outputs.

View full abstract

Fuzzy cognitive maps(FCMs)are effective to represent causal reasoning by numeric processing. They graphically represent uncertain causal reasoning. However, an expert has to construct FCM manually:there is no method to construct FCM automatically. In this paper we propose a novel method to construct FCM automatically from numerical data. First the system approximates a causal relation between every concept pair iteratively. Each causal relation is expressed as a triangular membership function or a linear function. Then the accuracy of the approximation is calculated. Using the value, the number of connections in a FCM can be reduced and the confidence factor is calculated to evaluate each causal relation. The factor contributes to improve the performance of the FCM greatly. Computer simulation results indicate the validity and effectiveness of the proposed method.

View full abstract

In this paper, by considering the experts'fuzzy understanding of the nature of the parameters in the problemformulation process, we formulate multiobjective nonconvex nonlinear programming problems with fuzzy numbers and present an interactive fuzzy satisficing method through floating point genetic algorithms. Using the α-level sets of fuzzy numbers, the corresponding nonfuzzy α-programming problem is introduced. After determining the fuzzy goals of the decision maker, if the decision maker specifies the degree α and the reference membership values, the corresponding extended Pareto optimal solution can be obtained by solving the augmented minimax problems for which the floating point genetic algorithm, called GENOCOP III, is applicable. In order to overcome the drawbacks of GENOCOP III, we propose the revised GENOCOP III by introducing a method for generating an initial feasible point and a bisection method for generating a new search point efficiently. Then an interactive fuzzy satisficing method is presented together with an illustrative numerical example.

View full abstract

In this paper, we discuss solution methods for a real world manpower allocation problem. The problem includes more than 7, 000 0-1 variables rendering the branch and bound method infeasible when considering the huge amount of computation time. We consider an application of the genetic annealing(GAn)which is an extension of genetic algorithm(GA)and simulated annealing(SA). In order to confirm that an optimal solution or a sub-optimal solution can be obtained GAn, using small-sized problems, the GAn solutions are compared with the solution of the branch and bound method. At the same time, we compare the obtained solutions with GA's and SA's. As a result, it is shown that the GAn solutions are optimal or at least sub-optimal. Appying the algorithms to the real size problem, it is shown that GAn produces a better solution. To obtain a further better solution, a diversification technique is introduced and its advantage is confirmed. Moreover, applying GA with GAn's genetic operators and GAn with GA's genetic operators to the problem, we analyzed the reason behind the success of GAn where GA failed.

View full abstract

This paper deals with multiobjective integer programming problems by considering fuzzy goals of the decision maker for objective functions. After determining the fuzzy goals of the decision maker, if the decision maker specifies the reference membership values, the corresponding Pareto optimal solution can be obtained by solving the augmented minimax problem which becomes an integer programming problem. For solving the problem, decoding algorithms for 0-1 programming problems are revised and ringed double strings are also introduced. Then an interactive fuzzy satisficing method is presented together with an illustrative numerical example.

View full abstract

This paper shows a novel medical image segmentation suitable for the human brain magnetic resonance(MR)volumetric images. The method is based on fuzzy information granulation. The concept is introduced by Zadeh. He considers that the information including imprecision and uncertainty consists of some fuzzy granules. Fuzzy information granulation is defined as deriving the granules from the information. The human brain MR volumetric images consist of several parts; the gray matter, white matter, cerebrospinal fluid and so on. We treat their parts as the fuzzy granules. By developing the fuzzy matching technique to aid the fuzzy information granulation, we can segment the brain region from the MR volume. An experiment is done on 50 human brain MR volumes. The error ratio, on the average, is 2.3%, compared to the manually segmented volumes by a medical doctor.

View full abstract

Fuzzy control rules are described with if-then expressions, and the fuzzy controls can incrorporate experts' control rules. It is, however, difficult to guarantee the stability of fuzzy control system. Studies have been done to analyze the stability of fuzzy control system. These studies are effective to analyze the stability of the fuzzy control system. However, the distinguishing feature of fuzzy controls, i.e.easily understandable linguistic expressions, have not been utilized in these analyses. A new method for stability analysis of fuzzy control systems using petri nets has been proposed. By simplifying the fuzzy control system as a discrete system, the fuzzy control system can be expressed using the petri nets. The transitions in the petri nets have one-to-one correspondence with fuzzy rules, and it is easy for us to comprehend the behavior of fuzzy control systems from the petri nets. The difficulty of this method is that the stability of the control system is guaranteed by manually following the firing sequences in the petri nets. This paper presents a new representation of fuzzy control system using matrix based on a bipartite directed multi-graph, and derives a theory of asymptotic stability. This approximation sacrifices the rigorous description of the behavior of the fuzzy control system. This paper also studies a method to analyse transitions between fuzzy sub-regions using numerical expressions. Simulation is done to verify the proposed stability analysis method.

View full abstract

In this paper, the genetic algorithms(GAs)are applied to the optimal design of the continuum structures. The chromosomes for the structures are defined by taking as the genes the control points of the functions and the material code. The population is constructed by the individuals with such chromosomes. The genetic operators such as the selection, the crossover and the mutation are applied to the population in order to determine the optimal structure. The minimum weight design of a plate is considered as a numerical example. The present scheme, firstly, is applied to the determination problem of the boundary profile of the structure in order to confirm the validity of the shape optimization algorithm. Next, it is applied to the simultaneous determination problem of the profile and the material of the structure.

View full abstract

This paper proposes a new fuzzy clustering technique to detect a number of local linear varieties based on a given data set. The main proposal is the introduction of an objective function that aims at obtaining an appropriate dimensional linear variety corresponding to the data distribution within each cluster. Then, the paper proposes a new type of fuzzy models expressed by elliptic type membership functions. One can develop this kind of models based on the clustering results only, without considering the model structure. An interactive simulation using the developed model helps us understand the data distribution in the multi-dimensional space and the relationships between variables

View full abstract

For people who are speech impaired and use Japanese sign language, it is helpful to develop an interactive Japanese sign language interface. Yamaguchi and Yoshihara etc. have proposed and realized a method for it. It uses only spatial motion features of different sign languages and doesn't think about the shape of hands. However, both spatial motion and shape of hands are important in sign language recognition. In this paper, we propose a method which employs skills of image processing to process and analyse simple shape of hands, and combine it with the method mentioned above. In our system, we also use Fuzzy Associative Memory Organizing Unit System(FAMOUS), which has property of robustness, to perform associative inference. Our system is person independent with high recognition rate.

View full abstract

An image of a set that produces a multiset from an ordinary set is proposed and contrasted with the ordinary image. Let X={a, b, c, d}, A={a, b, c}, B={a, a, b, b, c}, and C={(a, 0.1), (a, 0.2), (b, 1), (c, 0.5)}.Then A is an ordinary set, B is a crisp multiset, and C is a fuzzy multiset. Assume Y={v, w}and define f:X →Y by f(a)=f(b)=v and f(c)=w. From an ordinary extension of the image f(A)=f(B)=Y and f(C)={(v, 1), (w, 0.5)}, i.e., we do not obtain a multiset. In this paper we introduce another image, denoted by f[A], as follows. For each input x⋴A, f(x)is added to the output regardless whether or not there already exists f(x) in the output. For the above example, f[A]={f(a), f(b), f(c)}={v, v, w}and f[C]={(v, 0.1), (v, 0.2), (v, 1), (w, 0.5)}. Theoretical properties such as commutativity zof f[・]with α-cut and multiset addition are proved. Two applications are studied: one is a rough approximation using a natural projection onto the equivalence classes; the other is a query language in fuzzy relational databases.

View full abstract

This paper proposes an efficient way of solving Sup ・ min composite relation equations. The proposed method can lead all independent solutions which do not include each other. The independent solutions are obtained by simple logical operations, utilizing both the criterion matrix and the auxiliary matrix. These matrices are derived from the relationship between fuzzy relation R and fuzzy set b. Furthermore, applying the rule of exclusion, it is possible to remove some awkward elements which result in redundant solution sets.

View full abstract