Fuzzy Expert System Research Articles

Formula omitted]: A method for synthetic opinions to yield a robust fuzzy expert system

In many domains, decision-making is challenging, as experts are often limited in availability. However, without a sufficient number of expert opinions, the associated solutions would not be robust. Motivated by this, MOSY, a Method for SYnthetic Opinions has been developed to produce a robust Fuzzy Expert System (FES) by specifying Nsr, the number of (synthetic) experts per rule. For every one of these “synthetic experts”, MOSY produces an opinion from a normal distribution characteristic of a human expert. Correspondingly, the FES is used to produce an opinion from an antecedent vector whose elements are sampled from a uniform distribution. Synthetic and human opinion vectors, resulting from all rules and number of experts per rule, are driven to agree through optimization of weights associated with the fuzzy rules. The weight-optimized MOSY was tested against sets of human expert opinions in two distinct domains, namely, an industrial development project (IDP) and passenger car performance (PCP). Results showed that the synthetic and human expert opinions correlated between 91.4% and 98.0% on an average over 5≤Nsr≤250, across five outcomes of the IDP. Likewise, for PCP, respective correlations varied between 85.6% and 90.8% for 10≤Nsr≤150 across the two performance measures. These strong correlations indicate that MOSY is capable of producing synthetic opinions to yield a robust FES where sufficient human experts are not available.•This method, known as MOSY, generates synthetic expert opinions to achieve robustness in an FES.•MOSY was validated against sets of human expert opinions in two distinct domains.•Strong correlations were observed between the synthetic and human expert opinions.

Fuzzy expert system and information systems auditing: an approach for risk assessment in audit pre-planning

Fuzzy expert system and information systems auditing: an approach for risk assessment in audit pre-planning

Fuzzy expert system and information systems auditing: an approach for risk assessment in audit pre-planning

Fuzzy expert system and information systems auditing: an approach for risk assessment in audit pre-planning

Prediction System for Diagnosis and Detection of Coronavirus Disease-2019 (COVID-19): A Fuzzy-Soft Expert System

In early December 2019, a new virus named “2019 novel coronavirus (2019-nCoV)” appeared in Wuhan, China. The disease quickly spread worldwide, resulting in the COVID-19 pandemic. In the current work, we will propose a novel fuzzy soft modal (i.e., fuzzy-soft expert system) for early detection of COVID-19. The main construction of the fuzzy-soft expert system consists of five portions. The exploratory study includes sixty patients (i.e., forty males and twenty females) with symptoms similar to COVID-19 in (Nanjing Chest Hospital, Department of Respiratory, China). The proposed fuzzy-soft expert system depended on five symptoms of COVID-19 (i.e., shortness of breath, sore throat, cough, fever, and age). We will use the algorithm proposed by Kong et al. to detect these patients who may suffer from COVID-19. In this way, the present system is beneficial to help the physician decide if there is any patient who has COVID-19 or not. Finally, we present the comparison between the present system and the fuzzy expert system.

A fuzzy set theory-based fast fault diagnosis approach for rotators of induction motors.

Induction motors have been widely used in industry, agriculture, transportation, national defense engineering, etc. Defects of the motors will not only cause the abnormal operation of production equipment but also cause the motor to run in a state of low energy efficiency before evolving into a fault shutdown. The former may lead to the suspension of the production process, while the latter may lead to additional energy loss. This paper studies a fuzzy rule-based expert system for this purpose and focuses on the analysis of many knowledge representation methods and reasoning techniques. The rotator fault of induction motors is analyzed and diagnosed by using this knowledge, and the diagnosis result is displayed. The simulation model can effectively simulate the broken rotator fault by changing the resistance value of the equivalent rotor winding. And the influence of the broken rotor bar fault on the motors is described, which provides a basis for the fault characteristics analysis. The simulation results show that the proposed method can realize fast fault diagnosis for rotators of induction motors.

Fuzzy Logic-Based System for Liver Fibrosis Disease

The diagnosis of liver fibrosis (LF) is crucial as it is a deadly and life-threatening disease. Artificial intelligence techniques aid doctors by using the previous data on health and making a diagnostic system, which helps to take decisions about patients’ health as experts can. The historical data of a patient’s health can have vagueness, inaccurate, and can also have missing values. The fuzzy logic theory can deal with these issues in the dataset. In this paper, a multilayer fuzzy expert system is developed to diagnose LF. The model is created by using multiple layers of the fuzzy logic approach. This system aids in classifying the health of patients into different classes. The proposed method has two layers, i.e., layer 1 and layer 2. The input variables used in layer 1 for diagnosing liver fibrosis are Appetite, Jaundice, Ascites, Age, and Fatigue. Similarly, in layer 2, the input variables are Platelet count, White blood cell count, spleen, SGPT ALT (Serum Glutamic Pyruvic Transaminase Alanine Aminotransferase), SGOT ALT (Serum Glutamic-oxalacetic Transaminase Alanine Aminotransferase), Serum bilirubin, and Serum albumin. The output variables for this developed system are no damage, minimal damage, significant damage, severe damage, and cirrhosis. This research work also presents the examination of results based on performance parameters. The proposed system achieves a classification accuracy of 95%. Moreover, other performance parameters such as sensitivity, specificity, and precision are calculated as 97.14%, 92%, and 94.44%, respectively.

Approach for a Comprehensive Evaluation of Manufacturing Innovation Projects in the Early Phase of the Innovation Process

Driving innovation has become necessary for securing manufacturing companies' competitive capabilities in the context of digitalization, human centricity, sustainability, and resilience (Industry 5.0). However, due to its costs and uncertainties, the innovation process carries a high economic risk. Decisions of this magnitude can only be legitimized by a well-founded and thus often costly project evaluation. This paper aims to present a comprehensive evaluation approach to support companies in evaluating and classifying manufacturing innovation projects in the early phase of the innovation process. Based on a comprehensive literature analysis and expert interviews, a concept for an evaluation technique combining four dimensions has been elaborated. The first dimension is the potential analysis. It is evaluated using a type-1 Mamdani Fuzzy expert system, analyzing potentials to compute the effect of innovation on competitive strategic goals in manufacturing. Second, the initial efforts are identified by providing a structured list of possible expenses. The third dimension is an innovation project's overall risk to a company. It is calculated using an adapted Failure Modes and Effects Analysis. The failure modes are identified with the help of a comprehensive overview of potential risks that need to be evaluated regarding their significance, probability of occurrence, and reaction possibility. The fourth dimension helps the user to classify the selected innovation projects as incremental or radical for the company. The results of the four evaluation modules are summarized and visualized for their use in the decision-making process. Concluding, the approach has been validated by an initial use-case application to an industrial decision problem and verified by evaluating the approach's performance regarding the fundamental requirements for decision support systems.

Design of a Fuzzy Expert System for Organizational Performance and Innovation Evaluation Based on the Role of Knowledge Management and Intellectual Capital

Objective: This study aimed to evaluate organizational performance and innovation based on the role of knowledge management and intellectual capital. Methodology: This study was conducted using a qualitative approach (interviews with experts). Through expert interviews and grounded theory, research criteria including causal conditions, intervening factors, strategies, and outcomes were derived. The criteria and sub-criteria were then ranked using the fuzzy Analytical Hierarchy Process (AHP) method. Causal factors and strategies were used as inputs to the model, while outcomes were treated as output variables. The expert system in this study had two inputs and one output. For the input sub-criteria, rules were defined based on expert opinions, and fuzzy inference was performed. The outputs of the expert system were the sub-criteria of the outcomes, which included five variables, and the fuzzy section utilized MATLAB software. Findings: Among the identified factors, causal factors were found to be more important than contextual and intervening factors. Within the causal factors, social factors were of greater significance. In the strategies category, human resources were identified as the most important, and ultimately, organizational outcomes were considered the most significant results. Conclusion: Overall, it can be stated that evaluating organizational performance and innovation based on the role of knowledge management and intellectual capital is of great importance. The utilization of a fuzzy expert system approach in this context can significantly assist managers and decision-makers in assessing organizational performance and innovation.

НЕЧІТКА ЕКСПЕРТНА СИСТЕМА ДЛЯ КЕРУВАННЯ ТЕМПЕРАТУРОЮ ПОВІТРЯ У ПРИМІЩЕННЯХ ТОРГОВЕЛЬНО-РОЗВАЖАЛЬНОГО ЦЕНТРУ

Abstract The article presents the stages of developing a fuzzy expert system for the operator of a heating point of a shopping and entertainment center. The principle of operation of the air heating system of the mall is shown, implemented using computer-integrated control with a SCADA system. For the effective operation of the air temperature control system in different areas of the mall, functions for belonging to comfort zones have been developed in terms of visitors and the requirements of sanitary standards. It is indicated that typical air temperature stabilization controllers cannot always provide the desired mode due to the influence of a large number of uncontrolled disturbances on the control system. In this case, in the absence of additional blocks for adapting control algorithms, it becomes expedient to remotely control the operator using a mnemonic diagram of the ventilation process. And to improve the efficiency of the process of remote control of thermal equipment, an expert system was proposed that implements the algorithm of I. Mamdani, to control the coolant flow valves. The controller receives information from the outdoor air and air temperature sensors in the control area of the mall. After processing the signals from the sensors, the controller transmits information to a computer with an expert system. Depending on the temperature and season of the year, the program calculates the optimal position of the valve (in percentage of stroke) for the flow rate of the coolant and recommends installing it to the operator. Also, it is shown that in the process of developing a fuzzy expert system, it is recommended to choose the Gaussian type of membership functions; minimal number of terms - sets - three functions of dependency; weight coefficient of each rule in the knowledge base is the same and equal to 100; the method of defasification is the method of the center of gravity. It is indicated that the introduction of expert advice to operators can also lead to a reduction in the response time to emergency situations up to 90% increase in the reliability of the equipment as a whole.

Improving the fuzzy expert system for diagnosing depressive disorders

This paper presents an improving knowledge base and inference engine of a medical expert system for diagnosing depressive disorders. This medical expert system calls PORUL.DEP. PORUL.DEP’s knowledge base includes more than 850 positive rules. PORUL.DEP has been tested on more than 260 medical records of depressed patients. PORUL.DEP gives a correct diagnosis of more than 95% with light depressive disorder and without depressive disorder, but the remaining depressive disorders are not accurate. Average percent of more than 24 %. A new expert system, called STRESSDIAG, was developed on combining positive rules (for confirmation of conclusion) and negative rules (for exclusion of conclusion) for diagnosing depressive disorders. STRESSDIAG’s knowledge base consists of more than 850 positive rules of PORUL.DEP and more than 120 negative rules. Abelian group operation of Mycin is used to improve the inference engine based on fuzzy relations. STRESSDIAG gives the correct diagnosis of more than 76% with 4 depressive disorders types and without depressive disorders. Average percent of more than 82 %, up nearly 60% compared to PORUL.DEP.

Mamdani Fuzzy Expert System for Online Learning to Diagnose Infectious Diseases

E-learning and expert systems can be implemented for learning in the health sector. Through the e-learning system, prospective health workers can analyze problems by exploring the material in the system. However, material learning alone is less effective, so case study-based learning using an expert system is needed to strengthen understanding. The research applies an expert system to online learning to diagnose several infectious diseases. The disease diagnosis process uses the backward chaining method and the Mamdani fuzzy inference system. The fuzzy Mamdani inference system determines the intensity of disease severity so that appropriate treatment recommendations can be made. The test findings on 15 test datasets yielded a backward chaining accuracy value of 100%. Three test scenarios were used to establish the test using the Mamdani fuzzy inference method. Scenario 1: Testing with the Center of Gravity defuzzification and Fuzzy Mamdani Min inference system Tests employing the Fuzzy Mamdani Min inference method and center average defuzzification are used in Scenario 2. Scenario 3 involves testing using the Fuzzy Mamdani Product Inference System with Center Average Defuzzification. The average outcome for the intensity of disease severity utilizing the Fuzzy Mamdani Min inference system with Center of Gravity defuzzification was greater than that of the two test scenarios that were suggested, which was 49.43%.

Time series forecasting using fuzzy cognitive maps: a survey

Among various soft computing approaches for time series forecasting, fuzzy cognitive maps (FCMs) have shown remarkable results as a tool to model and analyze the dynamics of complex systems. FCMs have similarities to recurrent neural networks and can be classified as a neuro-fuzzy method. In other words, FCMs are a mixture of fuzzy logic, neural network, and expert system aspects, which act as a powerful tool for simulating and studying the dynamic behavior of complex systems. The most interesting features are knowledge interpretability, dynamic characteristics and learning capability. The goal of this survey paper is mainly to present an overview on the most relevant and recent FCM-based time series forecasting models proposed in the literature. In addition, this article considers an introduction on the fundamentals of FCM model and learning methodologies. Also, this survey provides some ideas for future research to enhance the capabilities of FCM in order to cover some challenges in the real-world experiments such as handling non-stationary data and scalability issues. Moreover, equipping FCMs with fast learning algorithms is one of the major concerns in this area.

Optimizing an irrigation treatment using an evolutionary algorithm and a knowledge discovery framework based on Deep models

Nowadays, learning models have good accuracy and learn complex patterns which are hidden in the data. Occasionally, patterns which are discovered using these models cannot be discovered by a human. So representing the discovered patterns and knowledge as an explainable knowledge for human is important. In this paper, a framework is proposed for knowledge discovery from real agricultural datasets. In the proposed framework, Deep Learning (DL) models are used for learning the patterns and effective features from tabular datasets; The integrated gradient method is used as an indirect metric for representing the black-box models as an interpretable model; A fuzzy expert system is used for interpreting the discovered knowledge as an expert system; An especial evolutionary algorithm is used for extracting the knowledge as a fuzzy expert system. In order to evaluate the proposed framework, some experiments are performed on the real tabular dataset on the agricultural field for extracting knowledge about productivity. For validating the discovered knowledge, an optimization algorithm is proposed which uses the discovered expert system for designing optimal irrigation treatments on the selected dataset. DSSAT, which is a farm simulator, is used for validating discovered knowledge. The obtained results show that the discovered knowledge can improve the productivity by 30% and decrease the used water by 30%. In addition, yields of the designed treatment and discovered knowledge correlate more than 0.8.

Fuzzy expert system for troubleshooting of the deodorizer unit in a palm oil refining plant

AbstractIn this paper, a troubleshooting technique is developed to provide a sequence of necessary actions during deodorizer unit failure in a palm oil refinery. This paper introduces a fuzzy expert system that incorporates the knowledge of experienced plant workers in dealing with the complex problems of the palm oil deodorizer. The presented methodology uses both qualitative and quantitative data. Qualitative data are collected from several interview sessions with the technical staff in a palm oil refinery. Quantitative data are gathered from various sources, such as control and monitoring systems. The qualitative data are analyzed to discover the possible pattern of faults in the deodorizer failure. This tool can diagnose possible faults and provide a sequence of necessary actions when deodorizer failure occurs. The obtained results show the developed deodorizer troubleshooting tool is useful to guide plant workers/operators in diagnosing the possible faults and abnormalities during the troubleshooting process, as well as to provide field training for inexperienced staff.

Assessment of Land Use Land Cover Changes for Predicting Vulnerable Agricultural Lands in River Basins of Bangladesh Using Remote Sensing and a Fuzzy Expert System

The aim of this study was to assess the LULC changes over 26 years from 1995 to 2021 to find the most changed land use conditions within the 25 km territory of the main river systems of Bangladesh. In addition, the prediction of vulnerable areas for agricultural land use in terms of inundation by river water was also analyzed. The study area includes river networks distributed through eight administrative divisions (Rangpur, Rajshahi, Mymensingh, Sylhet, Dhaka, Khulna, Barishal and Chittagong) of Bangladesh, covering an area of 64,556 km2. The study was conducted by identifying permanent water bodies from NDWI indices and preparing LULC maps that include the five main land use classes (water body, bare land, vegetation, agricultural land, and urban area) in the Google Earth Engine platform using supervised classification. The LULC maps were then analyzed in the ArcGIS® environment. A vulnerability map for agricultural land use was also prepared using a fuzzy expert-based system applying multicriteria analysis. From the land use land cover map of the study area, it was found that among the five land use classes, water bodies, bare land, vegetation, and urban areas increased in size by 3.65%, 2.18%, 3.31% and 2.55%, respectively, whereas agricultural land use significantly decreased by 11.68%. This decrease in agricultural land use was common for the analyzed area of all administrative divisions. According to the vulnerable area map of the eight divisions, more than 50% of the analyzed area of the Khulna and Dhaka divisions and more than 40% of the analyzed area of the Rajshahi, Mymensingh, Sylhet, Barishal and Chittagong divisions were highly vulnerable to agricultural land use due to the possibility of inundation by water. However, approximately 44% of the analyzed area of the Rangpur division was not vulnerable for agricultural land use. The prepared LULC and vulnerability maps can be helpful for the future land use planning of Bangladesh to meet the increasing demand for food production and livelihoods for increasing populations.

Multisensor Data-Fusion-Based Gas Hazard Prediction Using DSET and 1DCNN for Underground Longwall Coal Mine

The combustible and noxious gases are among the prominent issues affecting the life and the mining operations of underground coal mines (UCMs). Commonly adopted hazard monitoring methodologies in UCM are fuzzy, rule-based systems, statistical methods, and other expert systems, but these models are not reliable for highly complex and nonlinear systems. The neural network’s ability to learn and create nonlinear relationships is beneficial to making hazard prediction models. Especially, convolutional neural networks (CNNs) auto feature extraction capabilities to make it more suitable for the task. But, UCM’s harsh and crucial environment may result in sensor malfunctioning and faults, giving rise to data uncertainty. Like other data-driven models, data uncertainty significantly affects CNN’s performance. This study involves designing an effective and reliable gas hazard monitoring system using a hybrid of Dempster–Shafer evidence theory (DSET)-based filter and one-dimensional CNN (1DCNN) classifier. The novelty of this study is the integration of DSET and 1DCNN to predict the UCM hazard more reliably, even in malfunctioning node scenarios. Inherent usage limitations on traditional communication techniques restrict the application of cloud-based machine learning (ML) methods and this study use novel edge implementation of the filter and the classifier using edge ML (EML) technology. The proposed model’s hazard classification accuracy is 99.6%, even in the faulty node scenarios, where the traditional approaches fail.

Computational intelligence applied to cybersecurity

Computational intelligence applied to cybersecurity

An Ecological Expert System Optimization for Assessing Environmental Water Requirements of Hypersaline Lakes

The present study proposes an applicable methodology to optimize environmental water requirement of hypersaline lakes with a focus on Urmia lake as the case study in which remote sensing analysis, machine learning model and fuzzy expert system are linked. A machine learning model was developed to simulate effective abiotic parameters in which bands of operational land imager (Landsat 8) were inputs and depth and total dissolved solids were the outputs of the model. Moreover, an ecological expert system using Mamadani fuzzy inference system was developed to generate the habitat suitability map for the selected target species. Then, a multivariate linear model was developed to assess unit habitat suitability in which water level and total inflow of the lake were the variables of the model. An optimization model was developed to assess environmental water requirement in which habitat suitability between natural and regulated flows and water supply loss was minimized. The multivariate linear model was applied to assess habitat suitability in the optimization model. Based on the results in the case study, the proposed combined model is able to balance the ecological requirements and water demand by allocating 60% and 40% of total inflow to environmental water requirement and water demand respectively. Average habitat loss proposed by the optimal environmental water requirement was less than 20% which implies the robustness of the model. Generating habitat suitability maps of the lake by a reliable method which is used in the environmental flow optimization might be the significance of the proposed method.

Comparison of Artificial Intelligence Control Strategies for a Peristaltically Pumped Low-Pressure Driven Membrane Process.

Peristaltic pumping is used in membrane applications where high and sterile sealing is required. However, control is difficult due to the pulsating pump characteristics and the time-varying properties of the system. In this work, three artificial intelligence control strategies (artificial neural networks (ANN), fuzzy logic expert systems, and fuzzy-integrated local models) were used to regulate transmembrane pressure and crossflow velocity in a microfiltration system under high fouling conditions. A pilot plant was used to obtain the necessary data to identify the AI models and to test the controllers. Humic acid was employed as a foulant, and cleaning-in-place with NaOH was used to restore the membrane state. Several starting operating points were studied and setpoint changes were performed to study the plant dynamics under different control strategies. The results showed that the control approaches were able to control the membrane system, but significant differences in the dynamics were observed. The ANN control was able to achieve the specifications but showed poor dynamics. Expert control was fast but showed problems in different working areas. Local models required less data than ANN, achieving high accuracy and robustness. Therefore, the technique to be used will depend on the available information and the application dynamics requirements.

Land Suitability Analysis for Potential Vineyards Extension in Afghanistan at Regional Scale Using Remote Sensing Datasets

Grapes are one of the world’s most widely distributed crops and are cultivated in more than 100 countries in the global scheme. Due to climate change and improper vine growth variable selection, production has significantly decreased across countries. Therefore, the primary purpose of this study was to develop a land suitability analysis method using a fuzzy expert system at a regional scale. The fuzzy membership function was used in the ArcGIS® environment to perform the spatial analysis, and the overlay function was used to generate the final suitability map for Afghanistan considering policy planning. The results indicated that 23% (15,760,144 ha) of the areas were potential and located in the highly suitable region for grape production; however, 11% (7,370,025 ha) of the regions were not suitable for vineyards throughout the country of Afghanistan. In the present study, it was observed that most of the vineyards were in highly suitable areas (90%, 80,466 ha), while 0.01% (5 ha) of the vineyards were in less suitable areas. The present analysis demonstrated that the significant extension of grape vines can be possible in highly suitable areas. The results of this research can support decision-makers, farm managers and land developers to find more prospective acreage for expanding vineyards in Afghanistan.