IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
A Study on Production Scheduling Methods for Ready-Made Meal Industries
No full textPart 2: Human-centred Manufacturing and Logistics Systems Design and Management for the Operator 5.0International audienceIn ready-made meal industries, which are the focus of this study, maintaining freshness is just as important as meeting delivery deadlines due to the handling of fresh ingredients. To prevent deterioration in freshness, it is necessary to reduce the time from the start of processing to shipping. In this paper, a production scheduling method that simultaneously considers maintaining freshness and due date adherence, is proposed and verified the effectiveness of the proposed method through computational experiments
Understanding the Drivers of Lean Learning in Industrial Environments
No full textPart 5: Experiential Learning in Engineering EducationInternational audienceOrganizations are increasingly recognizing the benefits of adopting lean tools. However, a successful lean transformation depends on the active involvement of all organizational members. Comprehensive employee training is, therefore, fundamental to the effective implementation of lean manufacturing practices. Although collaborative and dynamic practical sessions have been implemented in certain university settings for teaching lean principles, there is limited research exploring how learning outcomes are influenced by participants’ characteristics. Examining these relationships is even scarcer in the literature concerning industrial training activities. This study examines the impact of lean training on employee learning within an industrial organization, looking at 177 participants and analyzing the relationship between learning outcomes and variables such as self-efficacy beliefs, prior knowledge, motivation, and enjoyment. The results show that the training significantly improved the participants’ lean knowledge level. A significant relationship was found between self-efficacy beliefs and employees’ motivation, which in turn had a positive impact on their learning. Based on these findings, companies should consider lean training that engages employees’ curiosity using non-traditional teaching methods
Data-Driven Scheduling of Cellular Manufacturing Systems Using Process Mining with Petri Nets
No full textPart 1: Smart and Sustainable Supply Chain Management in the Society 5.0 EraInternational audiencePetri net is a mathematical model for representing parallel, asynchronous, and distributed systems. Petri nets can model parallel and synchronous activities in manufacturing systems at various levels of abstraction. In this study, we propose data-driven modeling and scheduling for cellular manufacturing systems using process mining with Petri nets. In the proposed method, the event log data is extracted from a virtual plant and then the Petri net model considering the movement of products and operators is developed by using the process mining technique with the Petri net model. We also derived an approximate solution for the derived Petri net model from the event log using a local search method using a Petri net simulator. The analysis and modification of the model are conducted in the proposed method. Near-optimal schedules are derived using Petri net simulations. The validity of the proposed model is evaluated
Surveillance and Mitigation of External Stimuli-Induced Sensory Overload in Autism with IoMT: Communicating Insights to Caregivers
No full textPart 1: SDG 3 Good Health and Well-BeingInternational audienceAutism Spectrum Disorder (ASD) poses a notable developmental obstacle linked to neural abnormalities, characterized by repetitive and non-functional behaviour. The sensory overloads lead to elevated aggression, social phobia, tantrums, hypervigilance, and heightened sensitivity to external stimuli, necessitating heightened awareness. Caregivers encounter challenges in ensuring constant monitoring and surveillance. In response to these difficulties, a project has been devised, integrating diverse sensors on a wearable device to identify the environmental disturbances through PGP for heart rate, MEMS for stereotypic sensing,GSR to analyse the sweat, also external stimuli as for over loudness and suspicious gas, an IoT module with Raspberry pi, and machine learning methodologies for classifying and categorisation exact situation by analysing data using Convolution Neural Network CNN with better 97.6% accuracy. This helps to give immediate alert to caregivers by GSM module also with self-assistance. The central control of the project is facilitated by a PIC microcontroller and driving relays. Through the monitoring of IoT-collected data, a detailed analysis illustrates the frequency of incidents, including aggressiveness detected by the accelerometer sensor, heightened gas levels, along with increased occurrences of loud noise noted along with the heartbeat and sweat sensor. This analytical method empowers caregivers and doctors with a comprehensive understanding of the patient’s medical condition
Explaining Sentiments in Indian Legal Judgments with LIME and SHAP
No full textPart 4: SDG 11 Sustainable Cities and CommunitiesInternational audienceThis paper presents a comparative study on sentiment analysis applied to legal texts, specifically legal opinions, and judgments. This paper uses embeddings such as T5, XLM Roberta and BERT, Roberta, and Legal BERT to convert the textual data into pre-trained language model word representation, to make it easier for ML algorithms to determine the sentiment (positive, negative, or neutral) of legal documents. The dataset was converted into embeddings, and various models, including KNN, ANN, SVM, random forest, XG boost, logistic regression, and decision tree, were employed for sentiment analysis. Each model was subjected to hyperparameter tuning using Grid Search CV, resulting in the highest accuracy of 66.66% with SVM on T5 embeddings. This research provides valuable insights into the emotional degrees within legal texts, demonstrating the effectiveness of our sentiment analysis approach
Precision Exercise Monitoring Through Advanced Body Language Detection Using Computer Vision
No full textPart 1: SDG 3 Good Health and Well-BeingInternational audienceThe paper aims to assist individuals in performing physical exercises with precision and safety. This model delivers real-time feedback on an individual’s workout style by utilising powerful computer vision and machine learning technology. It uses deep learning models for position estimation to recognise and track essential body features while exercising. It classifies workouts and their phases, such as “down” and “up,” based on a continual monitoring of the user’s body position. Advanced computer vision algorithms are used to analyse video frames from the camera feed in order to extract critical data about the user’s form. It can recognise and classify certain workouts such as the deadlift, squat, and push-up by continually monitoring the user’s body position. The system provides useful information such as the current stage of the exercise, the number of repetitions, and the chance of correct performance. The model can recognize and detect the exercises performed by the person and it also iterates the number of times the exercises are done and it also detects the position of the body of the person. Healthcare professionals who want to promote safe and successful exercise regimens
Harnessing the Right Talent for SETA Programs: Cybersecurity Roles and Competencies that Make a Difference
No full textPart 1: Awareness and EducationInternational audienceSecurity Education, Training, and Awareness (SETA) is considered among the prominent strategies to develop a cybersecurity culture. Even though many SETA programs have been developed, their effectiveness is questionable as evident by the ongoing struggle of organizations to create a sustainable cybersecurity culture. A key factor that often challenges the design of effective SETA programs is the lack of expertise to create engaging and tailored initiatives to influence employees changing their unsafe behavior and adopting best practices. To address this challenge, organizations can leverage the expertise from multiple cybersecurity career roles, formulating a strong SETA development team that can exhibit a diverse range of perspectives and skills which are essential to design impactful SETA programs. Enabling such a collective design and development approach might be a solution to the pursuit of achieving a sustainable cybersecurity culture. This research work identifies: 1) the core knowledge areas and transferable skills that professionals responsible to design effective SETA programs should demonstrate, 2) which career roles in the ENISA European Cybersecurity Skills Framework cover relevant knowledge areas and transferable skills, 3) the prominent career roles for demonstrating knowledge and skills across multiple essential areas for SETA program development, and 4) the significance of lifelong learning in cybersecurity for developing sustainable SETA programs
A Diary Study to Understand Young Saudi Adult Users’ Experiences of Online Security Threats
No full textPart 1: Management and RiskInternational audienceAn online diary study was conducted to investigate the experience of online security threats among Saudi young adults. Over a period of 30 days, 16 participants were asked to record up to three threats they received from online sources on any of their devices. 58 threats were received, and 98 cues were reported in detecting the threats. The Phish Scale proved useful to categorise the detection cues, but needed expansion, largely due to the proliferation of threat types, which can come through many online channels including SMS, WhatsApp and online voice channels. The majority of threats were phishing, with general email phishing and target email phishing (spear phishing) being the most common types. The cues most commonly used to detect threats were those related to language and content of the threat, technical indicators such as the lack of a sender name or email or a suspicious or hidden link to follow, and tactics such as posing as a business or making an offer “too good to be true”
Phish and Tips:
No full textPart 2: Social EngineeringInternational audienceOlder adults are particularly vulnerable to phishing attacks. Gamification has been shown to be less effective to develop confidence in distinguishing between genuine and phishing emails in this demographic. To overcome this, we present our novel, open source interactive training platform, Phish&Tips, based on a simulated inbox. Our multi-analysis approach provides comprehensive data that enables us to compare participant’s self-assessed competence with their performance on the training platform. We present results based on pre- and post-training surveys, focus groups and the analysis of the training platform data (N=37). Over half the participants demonstrated an improved understanding of various detection strategies and an increase in confidence in being able to interpret emails. However, these results were not evident in the analysis of the platform data. This disparity between participants’ perceived knowledge and their performance on the platform highlights the challenges of applying their knowledge effectively
Bloom’s Taxonomy Based Question Analysis for Personalized Learning
No full textPart 2: Data AnalyticsInternational audienceIn the 1950s, Benjamin Bloom and his associates introduced Bloom’s Taxonomy, a foundational framework for categorizing learning objectives and cognitive skills. While question classification typically operates at a fine level, such as sentences and phrases, and text classification focuses on the document level, past research has explored the intersection of question classification and Bloom’s Taxonomy to assess learners’ cognitive levels in higher education. However, existing feature types from previous studies may excel in datasets with narrowly focused queries, necessitating the development of multiple classifiers for diverse fields or areas.To address this, A new kind of feature called “taxonomy-based” is suggested to improve the accuracy of question classification in datasets from different fields. Utilizing datasets comprising questions from distinct topics, the study evaluates the effectiveness of taxonomy-based features. Support vector machines (SVMs) are chosen as the classifier for their reputed text classification accuracy. The research reveals that taxonomy-based features significantly improve classifier performance when applied to question sets from different domains.This effort is driven by the need to improve question categorization accuracy over a variety of datasets and the realization of Bloom’s Taxonomy’s continuing use in educational settings. Although Bloom’s Taxonomy offers an extensive framework for comprehending cognitive abilities and learning objectives, current question classification techniques frequently do not have the granularity necessary to fully utilize this taxonomy. Furthermore, prior work has mostly concentrated on fine-level categorization, ignoring the advantages of integrating Bloom’s Taxonomy into the classification procedure.The highest accuracy achieved using SVM with TF-IDF vectorizer is 74.17% for BCL’s Data set and 95.23% for BT Dataset. Employing the KNN algorithm with the TF-IDF vectorizer yields 59.17% accuracy for BCL’s Data set, with BT Dataset reaching a maximum accuracy of 89.11%. Naïve Bayes algorithm, coupled with the TF-IDF vectorizer, achieves a peak accuracy of 73.33% for BCL’s Data set and 90.99% for BT Dataset. Lastly, using the Random Forest algorithm with the count vectorizer results in a maximum accuracy of 74.17% for BCL’s Data set and 95.21% for BT Dataset