Please wait a minute...
, No 6

■ Cover page(PDF 3171 KB) ■  Table of Content, June 2023(PDF 89 KB)

  • Optimization Approaches for Cost Reduction in Preventive Maintenance Strategies: A Comparative Study
    Yassine Eddouh, Abdelmajid Daya, Rabie Elotmani, and Abdelhamid Touache
    2023, 19(6): 359-367.  doi:10.23940/ijpe.23.06.p1.359367
    Abstract    PDF (485KB)   
    References | Related Articles
    Production companies aim to achieve several strategic objectives to remain competitive in the market, including increased productivity, improved profitability, and reduced operational costs. One effective approach to reducing operational costs is implementing a preventive maintenance optimization. This study explores three optimization strategies for reducing maintenance costs: the age replacement model, optimal inspection model, and optimization via Design of Experiment (DOE). The first strategy involves simulating the equipment's aging process and scheduling replacement when it reaches a certain age or usage level. In the second strategy, the goal is to identify the optimal timing for maintenance inspections, which minimizes costs while ensuring equipment reliability. The third strategy involves developing a mathematical model to minimize the expected cost by using ANOVA analysis and response surface methodology to identify the optimal parameter values. The study assesses the efficiency of the three optimization strategies in reducing costs, using the Weibull distribution function as the basis for cost reduction evaluation. The results indicate that shape parameter and replacement time significantly impact the expected cost. Hence, production companies can utilize these findings to determine the most efficient and cost-effective preventive maintenance strategy for their equipment.
    Demographic and Clinical Factors Role Identification in Stroke Risk and Subtype Prediction
    Deepak Kumar, Chaman Verma, Purushottam Sharma, Deeksha Kumari, and Zoltán Illés
    2023, 19(6): 368-378.  doi:10.23940/ijpe.23.06.p2.368378
    Abstract    PDF (481KB)   
    References | Related Articles
    The purpose of this study was to analyze the factors associated with stroke risk in a patient population comprising 4798 individuals. Using k-means clustering analysis, we identified a significant relationship between subpopulations and the degree of paralysis in stroke patients. Furthermore, we developed a machine learning model that utilized demographic and clinical factors to predict stroke subtypes, achieving an impressive overall accuracy rate of 86%. The crucial determinants for classifying the stroke subtype were found to be the patient's neurological condition, consciousness and memory, body mass index (BMI), glucose levels, and risk score. To gain deeper insights into the interrelationships among different variables, we applied principal component analysis (PCA) to the target attribute of stroke (TOS). The PCA analysis revealed five key principal components that shed light on the underlying dynamics. Specifically, age, cholesterol, glucose, diastolic blood pressure, and modified Rankin Scale (MRS) strongly influenced PC2. Conversely, risk score, MRS, systolic blood pressure, not specified abbreviation (nhiss), and diastolic blood pressure had a strong impact on PC1. In summary, this study contributes to the understanding of stroke risk factors by highlighting the relationship between subpopulations and paralysis severity. Moreover, the developed machine learning model demonstrates promising accuracy in predicting stroke subtypes based on key demographic and clinical factors. The findings obtained through PCA provide valuable insights into the interplay among different variables, emphasizing the influence of specific factors on principal components PC1 and PC2.
    A Framework to Evaluate Maintainability of Service-oriented Architecture using Fuzzy
    Arvind Kumar Mishra, Renuka Nagpal, Kirti Seth, and Rajni Sehgal
    2023, 19(6): 379-387.  doi:10.23940/ijpe.23.06.p3.379387
    Abstract    PDF (471KB)   
    References | Related Articles
    An approach to developing software for business applications is known as service-oriented architecture (SOA). This paper discusses the essential aspects of developing a framework for evaluating SOA's maintainability. The SOA is maintained by the framework using a cyclomatic complexity of various methods and a fuzzy model. 15,714 Java files and 1.4 million lines are implemented in this framework. In order to determine whether or not the inputs and outputs are maintainable, the 10886 dataset is used. The maintainability of 276 projects is being evaluated. The majority of the files in the study projects are thought to be simple to maintain. The goal of this framework is to report on and improve the working relationship and strategies that exist between the development department and the client support department. This framework should be followed for each complaint or demand for change from customers. The functionalities of the proposed framework include the following: receiving and logging client calls; focusing on and classifying issues; assigning issues to client support, engineers, and coders; allowing for issue fixes and minor improvements; assessing, classifying, and completing issues; identifying normal repeating issues and closing them for all clients; measuring and further developing client assistance by giving; and handling requests. Lower cyclomatic complexity helps programs comprehend and is less risky to modify. The mean relative error of various codes is calculated and compared in this paper using cyclomatic complexity and a fuzzy model. It is concluded that SOA maintenance is simple if the MARE is less than 0.269406586.
    Ensemble Learning for Appraising English Text Readability using Gompertz Function
    Rakesh Kumar, Sunny Arora, Ashima Arya, Neha Kohli, Vaishali Arya, and Ekta Singh
    2023, 19(6): 388-396.  doi:10.23940/ijpe.23.06.p4.388396
    Abstract    PDF (659KB)   
    References | Related Articles
    To fulfill individuals' informational demands, text readability is crucial. The assessing necessity of text readability is rising as a result of the enormous increase of contemporary content. An ensemble approach to learning utilizing the Gompertz function is suggested to assess the readability of English writings in light of word, sentence, and text arrangement. The conventional approach of measuring the readability of English literature depends excessively on the capacity of artificial experts to identify characteristics, which restricts its applicability. It becomes increasingly challenging to manually identify deep features due to the diversity and volume of text being used, as well as the readability assessment characteristics that must be extracted, and it is simple to add redundant or unnecessary characteristics, which hurts the effectiveness of the framework. For this study, the authors experimented with 25,000 English sentences. Furthermore, they were classified by Flesch-Kincaid and annotated into seven distinct readability categories. The study proposes an ensemble based model that employs five machine learning models as its base classifiers. The outcomes produced by the suggested ensemble based model are outstanding and reliable. The suggested model had an accuracy, precision, recall and F-score of 90.58%, 0.9545, 0.9467 and 0.9506, respectively on the test set. The created model may be applied in educational settings for tasks like language acquisition and evaluating an individual's reading and writing skills.
    Envisaging Alzheimer’s Disease Stage through Fuzzy Rank-Based Ensemble of Transfer Learning Models
    Neha Kohli and Tapas Kumar
    2023, 19(6): 397-406.  doi:10.23940/ijpe.23.06.p5.397406
    Abstract    PDF (1443KB)   
    References | Related Articles
    Convolutional neural networks (CNN), which have been proven to be effective computational methods for identifying pictures, are of special interest in neuroscience research due to their significance in identifying Alzheimer's disease (AD). The most prevalent form of dementia in the elderly community is AD. The necessity of swift pathology of AD identification by magnetic resonance imaging (MRI) persists. Many pre-processing techniques have converted three alternative projections of T1-weighted volumetric MRI scans into 2D space. Four CNN models utilizing transfer learning, i.e., VGG-19, Wide ResNet 50-2, GoogleNet, and Inception v3, employ pre-processed MRI for generating the outcome values that the proposed ensemble model would later combine. The proposed prediction model employs an ensemble approach to produce fuzzy rankings of the fundamental classification approaches using the Gompertz function and automatically integrates the base models' decision results to arrive at ultimate forecasts on the test instances for the stage of AD, i.e., mild demented, moderate demented, non demented, and very mild demented. The system's reliability is demonstrated by the framework's outstanding results on publicly accessible MRI datasets. Achieving an accuracy of 99.22%, recall of 99.53, precision of 99.69, f1-score of 99.61, and AUC of 98, the suggested ensemble performs better than the other underlying base classifier. When combined with additional medical examinations, the proposed ensemble model will be a useful and effective diagnostic tool for MRI scans for AD.
    Deep Learning Approach based on Iris, Face, and Palmprint Fusion for Multimodal Biometric Recognition System
    Manvi Khatri and Ajay Sharma
    2023, 19(6): 407-416.  doi:10.23940/ijpe.23.06.p6.407416
    Abstract    PDF (1039KB)   
    References | Related Articles
    With rising concerns about data theft and stricter security rules in many countries, biometric technology is becoming increasingly integral to our everyday lives. Given the severe limits of current single-modal biometric systems, it is no surprise that multimodal biometric approaches are experiencing a surge in popularity. Based on these findings, this work provides a novel multimodal biometric person identification system that combines iris, face, and palmprint biometric modalities for human recognition via the use of deep learning algorithms. The network relies on a convolutional neural network (CNN) to extract features, and a SoftMax and Tanh classifier to label images. The Adam and Adadelta optimisation technique are utilised to construct the CNN model, and the categorical cross-entropy loss function was implemented. As a result, the functional and evaluation levels were fused together. Several tests were conducted on the PolyU-IITD, PolyU Cross-Spectral Iris Image, and Tufts Face datasets to empirically evaluate the performance of the proposed system. In a biometric identification system, employing three biometric characteristics was shown to be superior to using one or two biometric features. Furthermore, the findings demonstrate that the proposed method achieves 100% accuracy, much better than existing state-of-the-art approaches.
    A Secured and Privacy Preserved VANET Communication using Blockchain
    Kavita Pandey and Shikha Jain
    2023, 19(6): 417-424.  doi:10.23940/ijpe.23.06.p7.417424
    Abstract    PDF (623KB)   
    References | Related Articles
    Vehicular adhoc networks (VANETs) are an interesting area of exploration among the intelligent transportation research community. Communication among vehicles with infrastructure units is an essential component. Thus, trust and privacy are important concerns in addition to dynamic topology, which is the main characteristic of VANETs. Ensuring the vehicles do not broadcast false information as well as protecting the identity of vehicles against tracking attacks are the objectives of this article. Here, a blockchain-based solution has been proposed to establish an identity-preserving trust model for VANETs. It preserves the real identities of vehicles with the utilization of Ethereum blockchain technology. A trust evaluation algorithm has been implemented to stop the dissemination of fraudulent messages. Validation of the algorithm has been conducted by running the algorithm in different VANET scenarios.
ISSN 0973-1318