IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
Assessing Additive Manufacturing and Digital Inventory Ecosystem in the Oil & Gas Context
No full textPart 1: Digital Transformation Approaches in Production and ManagementInternational audienceThe integration of additive manufacturing and digital inventories presents a paradigm shift in supply chain management, which promises reduced lead times, decreased complexity, lowered inventory storage costs, and improved sustainability. Despite the substantial benefits, the oil and gas (O&G) industry has been slow in adopting this technology. Drawing insights from companies already leveraging AM, this study investigates the critical functions in a DI ecosystem, as well as the challenges and actions required for the effective utilisation of DI within the O&G sector. Through qualitative analysis, the research identifies key factors influencing the adoption and implementation of DI in on-demand manufacturing and inventory management strategies. By understanding the challenges and opportunities associated with AM and DI integration, this study contributes to the discourse on innovative supply chain strategies and provides practical insights for DI ecosystem stakeholders
Instance Segmentation and Digital Twin Use Case for WIP Tracking in Heavy Industry
No full textPart 3: Computer Vision-based Digital Twin and Digital Services for Dynamic Production and Logistics EnvironmentInternational audienceIndustries such as railways, aerospace, shipbuilding, and construction are known for their heavy manufacturing processes, which involve the production of large and complex products. The railway industry is known for its tight delivery schedules, which make timely production and logistics critical. To address these challenges, manufacturers are increasingly turning to advanced technologies such as artificial intelligence vision (AI-vision) and digital twin (DT) technology. These technologies allow for the automatic creation and consolidation of key production and logistics information, resulting in improved manufacturing productivity. This study focuses on the implementation of a DT in a railway train manufacturing plant. Specifically, this study discusses the use of object detection through instance segmentation to track train body parts, particularly underframes. Using this method, manufacturers can identify status changes in each production process, enabling them to facilitate work-in-process logistics within factory workstations. This approach contributes to gain real-time process visibility and corresponding work efficiency. By embracing these advanced technologies, industrial facilities can build point-of-production systems in a cost-effective way. The use of AI-vision and DT technology is transforming the manufacturing industry, allowing manufacturers to boost productivity, reduce costs, and improve the overall quality of their products
A Methodology for Identification of Reconfigurability Enablers and Application in a Manufacturing System
No full textPart 2: New Horizons for Intelligent Manufacturing Systems with IoT, AI, and Digital TwinsInternational audienceManufacturing systems are continually challenged by evolving market demands, environmental constraints, and regulatory uncertainties. Manufacturers therefore need to develop and implement the reconfigurability capability to rapidly and cost-effectively adapt to new requirements throughout their life cycles. To this end, reconfigurability enablers —physical or logical factors that facilitate such adjustments— are needed. To support the identification and implementation of necessary reconfigurability enablers in existing factories and manufacturing systems, this study proposes a structured methodology to identify and select reconfigurability enablers and applies it within a case study in a brownfield factory. In the case study, an existing factory was analyzed. The current configuration was investigated, including the characterization of strategic requirements and anticipated trends. The analysis allowed to identify and select reconfigurability enablers for the manufacturer. The preliminary findings from the case study underscore the need to further implement and validate the methodology in industrial cases. The proposed methodology provides a foundation for future research on the systematic design of reconfigurability enablers. This study emphasizes the necessity for systematically designing reconfigurability enablers at various levels of detail, highlighting how enablers at workstation and system levels influence each other
The Incoherency Risk in the EU’s New Cyber Security Policies
No full textPart 2: Digital Transformation and Organizational InnovationInternational audienceThe European Union (EU) has been pursuing new cyber security policies in recent years. This paper presents a short reflection of four such policies. The focus is on potential incoherency, meaning a lack of integration, divergence between the member states, institutional dysfunction, and other related problems that should be at least partially avoidable by sound policy-making. According to the results, the four policies have substantially increased the complexity of the EU’s cyber security framework. In addition, there are potential problems with trust, divergence between industry sectors and different technologies, bureaucratic conflicts, and technical issues, among other things. With these insights, the paper not only contributes to the study of EU policies but also advances the understanding of cyber security policies in general
Understanding the Challenges of Fake News in the Tourism and Travel Industry: A Systematic Literature Review
No full textPart 3: Healthcare, Social Well-Being, and EthicsInternational audienceFake news in the tourism industry can influence consumer behavior, expectations, and opinions about the destinations or entertainment options they want to visit. For these reasons, it is important to understand the impact of fake news on the industry and the stakeholders who are impacted by it. The study employed a systematic review methodology and identified thirty-one papers from four academic databases. Thematic analyses were used to analyse the papers. The findings showed that consumers being impacted by fake news resulted in a ripple effect on businesses and countries. It was further reported that fake news was mostly distributed on online platforms such as social media, online reviews, and websites, and offline on editorials, opinion pieces, and entertainment shows. The creators of fake news mainly did it to benefit themselves. Both automated and manual methods were used to identify fake news. Tourism departments can mitigate these challenges by increasing awareness by hosting workshops and online campaigns for businesses and consumers to learn how to look out for fake news. Further investigation can be done into the willingness of the stakeholders to adopt the mentioned guidelines
Distributed Backdoor Attacks in Federated Learning Generated by DynamicTriggers
No full textInternational audienceThe emergence of federated learning has alleviated the dual challenges of data silos and data privacy and security in machine learning. However, this distributed learning approach makes it more susceptible to backdoor attacks, where malicious participants can conduct adversarial attacks by injecting backdoor triggers into their local training datasets, aiming to manipulate model predictions, for example, make the classifier recognize poisoned samples (injected with specific triggers) as specific images. In order to effectively detect backdoor attacks and protect federated learning systems, we need to know how backdoor attacks are generated and developed. Currently, most backdoor attacks to federated learning use centralized attacks with static triggers, which are easily detectable by current defense methods. In this work, we propose a distributed backdoor attack method that fully leverages the distributed nature of federated learning. It starts by generating unique and independent global dynamic triggers for infected benign samples and then decomposes the global trigger into multiple sub-triggers, embedding them into the training sets of multiple participants. During the training phase, data poisoning is introduced. Through extensive experiments, we demonstrate that this attack method exhibits higher persistence and stealthiness, achieving a significantly higher success rate than standard centralized backdoor attacks. Compared to classical distributed backdoor attack (DBA) methods, it shows noticeable improvements in attack performance
Advances in Production Management Systems. Production Management Systems for Volatile, Uncertain, Complex, and Ambiguous Environments: 43rd IFIP WG 5.7 International Conference, APMS 2024, Chemnitz, Germany, September 8-12, 2024, Proceedings, Part III
No full textInternational audienceBook Front Matter of AICT 73
Information Extraction to Identify Novel Technologies and Trends in Renewable Energy
No full textPart 3: Interdisciplinary and Cognitive Approaches in TRIZAchieving carbon neutrality by 2050 requires unprecedented technological,economic, and sociological changes. With time as a scarce resource, it iscrucial to base decisions on relevant facts and information to avoid misdirection.This study aims to help decision makers quickly find relevant information relatedto companies and organizations in the renewable energy sector. Over the courseof this PhD program, we will propose several text-mining methods applied tothe renewable energy sector in order to detect technological breakthroughs andnew, innovative companies. These techniques include specialized Named EntityRecognition (NER) models, news summarization, and trend analysis of scientificarticles. Further steps in this project will contain a TRIZ-based analysis of scientificarticles in order to attribute a multi-factor score on the innovative potential ofnovel technologies
TRIZ in Customer Experience Management - the Study of Current Research Problems
No full textPart 4: Customer Experience and Service Innovation with TRIZInternational audienceThe issue addressed in this paper refers to the improvement of methods and tools for systematic innovation in the field of marketing management. The problem of Customer Experience (CE) and Customer Understanding in the area of new technologies in marketing of Big Data, Machine Learning, and Artificial Intelligence in considered. The main objective of this paper is to reflect the actual application of TRIZ in the problem-solving process related to new categories of marketing problems. The systematic review of the literature is conducted. The database of the publications, using the search terms related to Customer Experience and TRIZ and including sources of data such as Web of Science, is formulated. The bibliometric and content analysis of the selected subsets of most relevant publications is conducted. As the results, a set of problems raised in publications in which TRIZ is used in the area of the so-called customer driven quality management was identified. In the analyzed set of publications, TRIZ is integrated with tools supporting quality management, i.e., Failure Mode and Effect Analysis (FMEA), Quality Function Deployment (OFD), SERVQUAL, Kono Model, as well as Kansei engineering (KE) or Six Sigma and Lean Management concepts. The results of the research indicate that by integrating marketing elements included in quality management, it may be possible to solve current AI-based marketing management problems using specific TRIZ methods and tools
Inventive Design for Transition to Industry 5.0 Based on Risk Management: Statement and First Proposition
No full textPart 2: Digital Transformation, Industry 4.0, and Predictive AnalyticsInternational audienceIndustry 5.0 is not just a technological evolution; it's not just more intelligence in components, products or equipment. There's a whole philosophy behind it that's a real threat to Small and Medium-sized Enterprises integrating into this approach. It therefore becomes important to identify and analyze the risks attributable to the transition to I5.0. The aim of the paper is to conduct a literature review about the research base on risk management related to the digital transition of SMEs into I5.0 and try to find all the aspects of risk management implementation involved in it. Hence, this literature review aim to adopt innovative approaches and methods to boost the robustness and capability of the managerial systems, technologies, knowledge, etc. available to Small and Medium Enterprises, so that they are ready to counter the various obstacles that could impact the transition process. Through this study, we aim to help deciders to integrate digital transformation by identifying and analyzing the risks and contradictions involved in moving towards I5.0