Fuzzy reasoning and genetic algorithms for decision making problems in uncertain environment

Limitations on the representation of knowledge, and hence the inference methods, naturally limit the domains of the first-generation expert systems. In response to these limitations, fuzzy expert systems or the second-generation of expert systems provide two essential and unique advantages in the design, development, and implementation of expert systems: (1) fuzzy knowledge representation and (2) fuzzy inference methods. It is suggested that second-generation expert systems be designed and implemented with the expressive powers of representation and improved inference methods based on fuzzy logic. A number of research topics and application areas are identified. >

This is a summary of the Ph.D. thesis defended by the author in December 2009 at Ecole des Mines de Nantes and Universidad de los Andes in Bogota. The thesis was advised by Christelle Gueret and Andres L. Medaglia and co-advised by Bruno Castanier and Nubia Velasco. The manuscript is written in English and it is available from the author upon request. The focus of the dissertation is to study real-world vehicle routing problems (VRPs) in uncertain environments. First, the thesis proposes a set of new methods for the VRP faced by a public utility and reports how these methods were embedded into a decision support system. Second, the thesis introduces a stochastic VRP widely found in practice but never studied in the literature before: the multi-compartment VRP with stochastic demands (MC-VRPSD). To solve the problem the dissertation proposes a set of solution methods that offer different tradeoffs between accuracy, speed, simplicity and flexibility. Lastly, the thesis proposes two multiobjective approaches to address the risk behavior of decision makers towards the cost spread in stochastic routing problems and applies them to the MC-VRPSD.

In this paper, a fuzzy expert system is described for decision-making in geographical information systems (GIS), because fuzzy logic and expert system technology are available methods for approximating human reasoning and enhancing the level of intelligence in GIS. This system computes a potentiality map for problem solution with a problem specification provided by the user and expert knowledge (by means of production rules). It is a fuzzy expert system using a generalised modus ponens based on the fuzzy implication of Brouwer-Godel which is implemented by a fast algorithm. Several kind of rules are accepted and especially gradual rules. We have experimented our system on "Palni Hills", a mountainous region of meridional India to locate areas where the profitability of agriculture could be generally the best. The system is explicitly designed to provide the user with a decision-making environment for the cartographic problem solution to be carried out in a flexible manner.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

With the increasing need for flexibility and adaptivity in computerized systems, the application of fuzzy expert systems is becoming increasingly commonplace in today's industry. Fuzzy logic expert systems often improve performance by allowing knowledge to generalize without requiring the knowledge engineer to anticipate all possible situations. Thus, for many types of applications, soft such as Fuzzy logic can incur lower overhead in terms of representing and engineering task knowledge. My project investigates the application of fuzzy expert systems to motion tracking. Previous research showed that Fuzzy Logic can be used to track the motion of a brightly colored object against a dark background, with relatively low development and run-time costs. The system I am developing identifies, tracks, and predicts the motion of multiple objects using unique identification patterns against a dark background. My poster will describe the fuzzy inference systems for tracking and motion prediction of such objects. An essential step to obtaining the fuzzy inputs for the motion tracking fuzzy inference system is to use convolution correlation data to obtain the centers of mass of the objects. Image processing information from the region around the center of each object provides good fuzzy inputs for recognizing object patterns and determining orientation. I am investigating two fuzzy inference systems for motion tracking and prediction in order to identify their strengths and weaknesses. In the long run, I hope to exploit the strengths of both by dividing the task between the two approaches. The Adaptive Neuro Fuzzy Inference System (ANFIS) is effective for pattern recognition. Fuzzy Proportional Derivative (FPD) is useful for computing errors and tuning inferences. Cascading the two fuzzy systems allows the system first to acquire generalized patterns and then fine-tune them for error tolerance. To perform this work, I am using the Matlab fuzzy logic toolkit, which is a simple and flexible tool for designing fuzzy inference systems.

This chapter describes a Genetic Algorithm (GA) based Fuzzy Goal Programming (FGP) model to solve a Multiobjective Bilevel Programming Problem (MOBLPP) with a set of chance constraints within a structure of decentralized decision problems. To formulate the model, the chance constraints are converted first to their crisp equivalents to employ FGP methodology. Then, the tolerance membership functions associated with fuzzily described goals of the objective functions are defined to measure the degree of satisfaction of Decision Makers (DMs) with achievement of objective function values and also to obtain the degree of optimality of vector of decision variables controlled by upper-level DM in the decision system. In decision-making process, a GA scheme is adopted to solve the problem and thereby to obtain a proper solution for balancing execution powers of DMs in uncertain environment. A numerical example is provided to illustrate the method.

A fuzzy logic expert system to estimate intrinsic extinction vulnerabilities of marine fishes to fishing

In construction, many owners mitigate the risk of unforeseen contractor default by accepting only bonded contractors who must endure a rigorous evaluation process by surety brokers and surety underwriters. This evaluation process includes a financial analysis and a review of work on hand and past performance, all of which have reliable structured methods for their evaluation. Additionally, a number of subjective criteria are considered that are more difficult to capture and assess objectively but which can be modeled effectively using fuzzy logic. The purpose of this paper is to illustrate how fuzzy logic and expert systems can be combined to provide a structured approach to evaluating contractors for surety underwriting purposes. Fuzzy logic is used to model both the objective and subjective factors considered in contractor evaluation using linguistic terms, and expert rules are used to capture the surety experts’ reasoning process. A fuzzy expert system, SuretyAssist, is presented that can be used to provide an initial evaluation of general contractors as well as periodic reviews to determine whether or not to accept them as clients for bonding. SuretyAssist was validated using 31 actual cases of contractor evaluation and found to be accurate in 81% of the cases.

Although fuzzy logic controllers and expert systems have been successfully applied in many complex industrial processes, they experience a deficiency in knowledge acquisition and rely to a great extent on empirical and heuristic knowledge which, in many cases, cannot be objectively elicited. Among the problems to be resolved in fuzzy controller design are the determination of the linguistic state space, definition of the membership functions of each linguistic term and the derivation of the control rules. Some of these problems can be solved by application of machine learning. First, it is desirable to simplify and automate the specification of linguistic rules. Secondly, it is also desirable that modification of control rules is possible in order to cope with previously unknown or changes in process dynamics. Machine learning methods have, in recent years, emerged from the use of learning algorithms modelled on natural and biological systems. These methods attempt to abstract the advanced mechanisms of learning exhibited by such systems, which can consequently be applied to intelligent control. One of these new algorithms is the genetic algorithm which is modelled on the processes of natural evolution. The paper develops the application of genetic algorithm techniques for fuzzy controller design. Genetic algorithms are used to automate and introduce objective criteria in defining fuzzy controller parameters.

Multi-period uncertain portfolio optimization model with minimum transaction lots and dynamic risk preference

The fuzzy-associative metaheuristic approach addresses the urgent task of developing a marketing decision support system based on a fuzzy trained associative rules expert system, aimed at improving the accuracy and efficiency of consumer preference analysis. The proposed system combines the interpretability of fuzzy logic with data-driven learning via associative rules and parameter identification using an adaptive multi-agent optimization method. To achieve this goal, associative rule learning techniques (Apriori and FP-Growth) were used to extract frequent consumer behavior patterns. A fuzzy expert system was developed, in which the parameters of membership functions are optimized by the Adaptive Vibrating Particle System (AVPS) metaheuristic. Unlike traditional vibrating particle systems, AVPS integrates iteration-dependent control of particle positions, enabling global search in early iterations and local refinement at later stages, thus improving convergence speed and solution precision. The architecture was implemented using Python-based tools (TensorFlow, Keras, Pandas, mlxtend, Scikit-Fuzzy), and validated on the “Consumer Behavior and Shopping Habits” dataset. The fuzzy expert system achieved an accuracy of 0.98, outperforming human experts (0.80), traditional VPS optimization (0.93), and backpropagation-based training (0.90). The system also reduces reliance on manually tuned parameters and increases robustness to data incompleteness and noise. Scientific novelty lies in combining a fuzzy associative rule-learning framework with AVPS-based optimization, offering a scalable and interpretable decision-making mechanism. The developed system contributes to the advancement of intelligent recommendation engines, personalized marketing tools, and decision support systems in consumer-oriented analytics.

An assessment of neonatal outcome may be obtained from analysis of blood in the umbilical cord of an infant immediately after delivery. This can provide information on the health of the newborn infant and guide requirements for neonatal care, but there are problems with the technique. Samples frequently contain errors in one or more of the important parameters, preventing accurate interpretation and many clinical staff lack the expert knowledge required to interpret results. The development and validation of an expert system to overcome these difficulties is described. The initial development utilised conventional 'crisp' logic within the rule base and this system was evaluated to commercial release. This expert system validates the raw data, provides an interpretation of the results for clinicians and archives all the results, including the quality control and calibration data, for permanent storage. Subsequent development went on to incorporate fuzzy logic into part of the expert system knowledge base, but tests of this preliminary fuzzy system showed that it performed worse than the original crisp expert system. A tuning algorithm was then employed to modify the fuzzy model and this process resulted in improved performance to a level comparable to clinicians and superior to the crisp system. Finally, the entire knowledge base was converted to utilise fuzzy logic and this 'integrated' fuzzy expert system was validated against international expert opinion.

This paper describes a fuzzy expert system for design project performance evaluation and prediction. It presents a comprehensive framework of factors that impact design performance and factors used to measure performance. A new approach to generating membership functions based on objective data is presented. This approach provides for membership functions that are widely applicable in a given context and can be calibrated to suit different contexts. A method of generating expert rules to relate factors impacting design performance is presented. A survey was conducted to collect data to develop and test the proposed methods. These methods were used in developing the fuzzy expert system. Based on validation of the system, the fuzzy expert system provides accurate linguistic predictions of design performance parameters. The methods presented in this paper are useful and realistic in modeling design performance and in capturing the inherent subjectivity involved.Key words: construction, design, evaluation, expert systems, fuzzy logic, performance, prediction, productivity.

Great emphasis has been put on applying expert systems for transmission system fault diagnosis. However, very few papers deal with the unavoidable uncertainties that occur during operation involving the fault location and other available information. The paper proposes a method using fuzzy sets to cope with such uncertainties. A fuzzy expert system is developed, which requires much less memory to store an active database than that used by conventional rule-based expert systems. The fuzzy expert system identifies two basic network section sets, Shealthy for the healthy subnetwork and Sisland for the fault islands, using the postfault status of circuit breakers and relays. It then conducts inference to determine the most likely fault sections and the associated fault sequences. Two case studies are provided to demonstrate salient features of the proposed approach, and to compare results both with a conventional rule-based expert system and a refined genetic algorithm. The proposed fuzzy expert system can be operated either in a stand-alone manner, or as part of the overall architecture of a diagnosis system for parallel decision making.

Objectives: Operational effectiveness is an ultimate objective of any manufacturing organization. On the shop floor the main focus is on elimination of waste and delays. The main target is designing a product for which manufacturing is a low cost process. Once organizations select appropriate tools and methods, the decision on performance measures can be taken through which progress can be monitored. The present work offers fuzzy expert system for manufacturing method selection in a dynamic environment where the organization’s objectives are subject to constant amendments. Methods/ Statistical Analysis: In order to incorporate ambiguity into the manufacturing environment, the authorshave designed and developed a fuzzy expert system using Sugeno model based on the dynamic ranges of triangular membership functions for input objectives which is extremely efficient as compared to the Mamdani method in classifying vague data. In this paper, the authors proposearchitecture for fuzzy expert system based on multi objective criteria for selection of manufacturing method employing Sugeno model with constant output. Fuzzy expert system is created outside MATLAB and MATLAB is used only for creating user interface for querying methods based on objectives and for the evaluation of rules. Findings: As more than one manufacturing method may serve a single objective, each manufacturing method is encoded using a binary digit and the output is decimal representation of this binary encoding. Also, it is observed that, the number of fuzzy rules increase exponentially as a function of number of objectives. In order to cater to this problem, instead of generating so many rules the required modifications are performed at the code level to incorporate multi objective criteria. Finally a simulink model is developed for selective methods and objectives. The results obtained using Sugeno Fuzzy Expert System is compared with that obtained using Mamdani method and crisp expert system. Results: It is found that the crisp expert system and Sugeno type FIS yield similar results for classification while Mamdani type FIS offers more flexibility in method selection due to the nature of output members functions which are overlapping fuzzy sets. This yields the manager a greater freedom in method selection based on infrastructure and other resource availability. Keywords: Expert System, Fuzzy Logic, Manufacturing, MATLAB, Membership Functions, Simulink.

Power transformers are very vital components in electrical power network. Deterioration of transformer’s insulation has significant impacts on its health condition. Hence, detailed analysis of insulation testing attributes is very important in diagnosing the transformer. It is possible with an effective decision-making system. In the present paper, a new grey relational-based fuzzy logic expert system is proposed for continuous health monitoring of power transformers. In this approach, ranking-based qualitative relational analysis is initially carried out in order to reduce large number of input attributes into three input health grades. These input grades are further integrated with a technically designed expert fuzzy rule base. Outputs of the proposed model are validated with the outputs of multi-criterion based fuzzy logic model presented in [Mharakurwa E T and Goboza R 2019 Multiparameter-based fuzzy logic health index assessment for oil-immersed power transformers Adv. Fuzzy Syst. 2019]. Testing data of 200 in-service transformers are utilised in the validation of both the models. It is observed from the comparison that similarity index of the proposed grey-fuzzy system is 92 percent. Several issues exist in regular transformer health assessment models such as the large number of rules, uncertainties, requirement of large training data, complexity issues are overcome by integrating the grey relational analysis with the fuzzy logic system. The proposed model can be easily implemented by all utilities and diagnostic experts of transformers.