IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
Integrating Multilayered Agility into Production Planning and Control: A Conceptual Model for Enhanced Manufacturing Efficiency
No full textPart 1: Lean Thinking Models for Operational Excellence and Sustainability in the Industry 4.0 EraInternational audienceThis article introduces a conceptual model that integrates multilayered agility into Production Planning and Control (PPC), addressing the urgent need for manufacturing firms to adapt to the unpredictable challenges of the 21st-century marketplace. Grounded in the dynamic capabilities theory, our model synergizes organizational, operational, and personal-workforce agility layers within the PPC framework to enhance manufacturing efficiency in a Volatile, Uncertain, Complex, and Ambiguous (VUCA) environment. Through a comprehensive literature review, we identify gaps in the operationalization of agility in PPC and propose a structured approach to its integration. Our model offers a novel perspective on embedding agility into PPC, aiming to provide a roadmap for manufacturing firms to achieve operational excellence and a sustainable competitive advantage. The anticipated benefits of this model include improved responsiveness to market changes, enhanced operational flexibility, and a stronger innovation capability, paving the way for future empirical research in agile manufacturing
Robust Novel Defect Detection with Neurosymbolic AI
No full textPart 6: Applications of Artificial Intelligence in ManufacturingInternational audienceDetecting novel product defects whose classes have not been seen at all during training time, is an important aspect of practical automated visual inspection in manufacturing. Without proper handling it is possible that these unknown defects will remain unnoticed causing production quality to deteriorate. Collecting more and more defect data is also not a solution as defects occur rarely in production and the ramp-up time of the AI-driven quality inspector becomes significantly slower. Since traditional machine algorithms are not always designed for handling these challenges, this paper applies an innovative approach based on Neurosymbolic AI. Specifically, we use a Logic Tensor Network that expresses the outputs of an unsupervised out-of-distribution detector as symbolic rules and uses them to drive the training of a neural network classifier. The resulting algorithm shows improved results in comparison to other related methods, especially in terms of defect recall, meaning that few defects remain undetected even if completely novel. More specifically, it achieves similar or better recall scores than semi-supervised and unsupervised methods when handling novel defects, but significantly outperforms them in defects that were seen during training. Similarly, when compared to supervised methods, it maintains high performance on known defects but significantly improves on novel ones. These best-of-both-worlds results are illustrated through higher F1-scores in the majority of the test datasets of manufacturing products
Exploring the Hydrogen Transition Within the Maritime Value Chains
No full textPart 3: Transforming Engineer-to-Order Projects, Supply Chains, and Systems in Turbulent TimesInternational audienceThe maritime industry is experiencing a transition towards the adoption of sustainable alternatives to traditional fossil fuels. The focus is on hydrogen due to its emergence as a feasible solution to the problem of pollution across the industry. The context comprises four value chains: those pertaining to the design, production, and management of fuel, ships and infrastructure (i.e., port and bunkering facilities); as well as the value chain associated with maritime operations (i.e., sea voyage between two ports). To successfully introduce hydrogen in maritime operations, the industry must be aligned in terms of the design and management of fuel, ships and infrastructure to accommodate the necessary changes. The study investigates the relationships within and between value chains, which can either support or hinder the transition. To this aim, a systematic literature review is performed. The analysis of 42 articles provides insights at the intersection of these value chains. The study identifies the decisional variables that will have implications on the industry at large, including the storage form of the fuel and related requirements, the design of the ship and its fuel consumption, and the availability of suitable supporting infrastructure. Finally, the aim is to examine current research trends, highlight any shortcomings, and identify future research directions
Exploring the Intersection of Artificial Intelligence and Machine Learning in Supply Chain Management: A Structured Literature Review
No full textPart 6: Applications of Artificial Intelligence in ManufacturingInternational audienceThe purpose of this paper was to identify the contributions and applications of Artificial intelligence (AI) and Machine Learning (ML) algorithms to Supply Chain Management (SCM) through a systematic review of the existing literature. The database used to identify the articles was Scopus, and among 454 articles that met the inclusion criteria and were published during 2010–2022, 81 articles were included in the final analysis. The review process was based on a clear framework that consisted of the definition, selection, analysis, and conclusion phases. By distributing the frequency numerically and using bibliometric analysis and network visualization, the trends of SCM research and the most prevalent AI and ML approaches were outlined. The results demonstrated increased research activity after 2019, with AI and ML utilized mostly in production, marketing, market trend analysis, procurement management, demand forecasting, logistics, supplier selection, and supply chain risk management. The study also established that there is a growing tendency to blend Artificial Neural Networks (ANNs) with other AI approaches to improve SCM results. The literature review also highlighted gaps in the existing literature, for example, the lack of real-world implementation studies and insufficient discussion of the impact of AI on the human side of SCM. Based on the findings of this study, it is recommended that future research focus on methodological and application concerns and incorporate AI-based techniques while integrating Supply Chain Risk Management (SCRM) and sustainability perspectives
A Flexible Job Shop Scheduling Problem Involving Reconfigurable Machine Tools Under Industry 5.0
No full textPart 4: OptimizationInternational audienceThe rise of Industry 5.0 has introduced new demands for manufacturing companies, requiring a shift in how production schedules are managed to address human-centered, environmental, and economic goals comprehensively. The flexible job shop scheduling problem (FJSSP), which involves processing operations on various capable machines, accurately reflects the complexities of modern manufacturing settings. This paper investigates the FJSSP involving reconfigurable machine tools with configuration-dependent setup times, while integrating human aspects like worker assignments, moving time, and rest periods, as well as minimizing total energy consumption. A mixed-integer programming (MIP) model is developed to simultaneously optimize these objectives. The model determines the assignment of operations to machines, workers, and configurations while sequencing operations, scheduling worker movements, and respecting rest periods, and minimizing overall energy consumption. Given the NP-hard nature of the FJSSP with worker assignments and reconfigurable tools, a memetic algorithm (MA) is proposed. This meta-heuristic evolutionary algorithm features a three-layer chromosome encoding method, specialized crossover and mutation strategies, and neighborhood search mechanisms to enhance solution quality and diversity. Comparisons of MA with MIP and genetic algorithms (GA) on benchmark instances demonstrate the MA’s efficiency and effectiveness, particularly for larger problem instances where MIP becomes impractical. This research paves the way for sustainable and resilient production schedules tailored for the factory of the future under the Industry 5.0 paradigm. The work bridges a crucial gap in current literature by integrating worker and environmental impact into the FJSSP with reconfigurable machine models
Forging Resilience Through Supply Chain Collaboration: Insights from the Chinese Automotive Industry
No full textPart 2: Resilience Management in Supply ChainsInternational audienceThis paper explores the pivotal role of supply chain collaboration in building resilience within the Chinese automotive industry amidst the new kind of disruption catalysed by the COVID-19 pandemic. Resource Dependency Theory (RDT) provides the theoretical foundation for this research, which used case studies and interviews to examine resilience strategies. The research found information sharing, resource sharing (supporting smaller suppliers), and adjustments (seeking external help) played key roles. This research highlights collaboration’s importance in building supply chain resilience and offers practical guidance for navigating disruptions
Energy Conscious Bi-objective Job Shop Scheduling: A New Formulation and Augmented ε-Constraint Method
No full textPart 4: OptimizationInternational audienceThis paper addresses a job shop scheduling problem in which machines can operate at varying speeds and different energy efficient strategies known as speed scaling and switching on/off are incorporated as well. When a machine runs at high-speed, the amount of time to complete the job shortens but energy consumed by the machine increases. Selection of different speed modes of machines for different jobs (i.e. speed scaling) generates compromise solutions. To save energy further, one must decide whether to shut down the machine during idle periods of consecutive jobs. One option is to turn off the machine whenever the idle period occurs regardless of its duration, which may result in machine breakdown due to excessive opening and closing. Alternatively, a threshold or time limit can be determined below which the machine is kept in standby mode by consuming very little energy. We aim to minimize two conflicting objectives, energy consumption resulting from usage while the machine runs at a particular speed or in standby state and total tardiness emanating from late completions. To this end, we developed a MILP formulation for the problem and Augmented ε-Constraint (Augmecon) method is implemented to find pareto optimal solutions. The experimental result reveals that energy consumption and total tardiness objectives are conflicting. Based on payoff table, while the total energy consumed is 25000, total tardiness is 270. When energy consumption increases to 32186, total tardiness reduces to 36. Between the two, Augmented ε-Constraint (Augmecon) method provides non-dominated optimal solutions based on 11 grid points
A Digital Twin Framework for Flexible Manufacturing System
No full textPart 3: Digital Twin Concepts in Production and ServicesInternational audienceIn the era of digitalization and automatization, several technologies and manufacturing paradigms emerged and became popular and attracted the interest of both researchers and industrials. Flexible Manufacturing System that is one of these paradigms is a production system able to switch between tasks easily making it adaptable to varying production needs. Furthermore, the concept of Digital Twin—a virtual counterpart of a tangible object—emerged and gained widespread popularity across diverse domains. Despite the focus of several research studies to develop these concepts, there is still a lack of work on an integrated and comprehensive framework that encompass a Digital Twin for Flexible Manufacturing System with a focus on smart manufacturing. Thus, this paper provides an attempt to propose a detailed Digital Twin framework for Flexible Manufacturing System with a focus on smart manufacturing and outlining various components and information. This is done by integrating different paradigms namely Machine learning; Simulation based Optimization, and Acquisition technologies. The proposed framework is composed of Physical part, Virtual part, and Stakeholders
Routing Heuristics for in-House Transportation in Assembly Systems
No full textPart 4: OptimizationInternational audienceThis research addresses routing challenges in high-variety mass customization environments, focusing on in-house transportation problems involving traversal between supermarket cells and consecutive assembly stations. Supermarket cells are organized separately from assembly stations in parallel rows, each cell having a designated pickup point and each assembly station a drop-off point. The routing problem aims to establish an efficient path, starting from the depot and visiting all pickup and drop-off points exactly once before returning. This study introduces a range of heuristic methodologies tailored to address the problem. These heuristics, such as nearest neighbor, S-shape, modified S-shape, and combined heuristics, are characterized by practical implementability, offering practitioners defined guided rules to navigate the shop floor. Artificial problem instances of varying sizes, characterized by the number of cells and pickup/drop-off points, are analyzed, comparing heuristic performance against an optimal benchmark solution. Results indicate that heuristics perform better in larger problem sizes, particularly with 100 rows, showing average gaps below 11.48% and 1.02% for 7-row and 100-row instances, respectively
