1,721,015 research outputs found
Ontology-based Process Reengineering To Support Digitalization Of MRO Operations: Application To An Aviation Industry Case
No full textISSN:2212-827
Method for Data-Driven NC-Code Optimization based on Dexel Material Removal Simulation and Tool Holder Vibration Measurements
Full text linkThis paper presents a novel approach for data-driven NC-Code optimization, based on the integration of dexel-based material removal simulation and an instrumented tool holder, capable to measure vibrations during milling close to the cutting zone. Considering measured cutting vibrations, machine tool axis and NC-line data, a model has been developed optimizing cutting parameters to generate a NC-Code right after a first machining trial with mitigated vibration effects. Different modules for human-assisting vibration visualization and automated optimization of cutting parameters are presented, using milling use-cases implemented on a CNC machining center
Deep Reinforcement Learning as an Optimization Method for the Configuration of Adaptable, Cell-Oriented Assembly Systems
No full textThis paper investigates the feasibility and performance of Deep Reinforcement Learning (RL) as a method for optimizing assembly cell configurations in adaptable cell-oriented assembly systems (ACAS). ACAS can be as productive as conventional assembly lines, while offering greater flexibility and resilience. However, optimizing their layout configuration and resource assignment poses a complex challenge for conventional optimization methods. A RL and simulation-based method is evaluated in an ACAS use-case setting, including a benchmark with metaheuristics. The findings show the limitations of RL for static aspects of the optimization problem, but also indicate RL's considerable benefits for dynamic optimization tasks in ACAS
Multi-sourced modelling for strip breakage using knowledge graph embeddings
Full text linkStrip breakage is an undesired production failure in cold rolling. Typically, conventional studies focused on cause analyses, and existing data-driven approaches only rely on a single data source, resulting in a limited amount of information. Hence, we propose an approach for modelling breakage using multiple data sources. Many breakage-relevant features from multiple sources are identified and used, and these features are integrated using a breakage-centric ontology which is then used to create knowledge graphs. Through ontology construction and knowledge
embedding, a real-world study using data from a cold-rolled strip manufacturer was conducted using the proposed approach
Enhancing resilience on the shopfloor through data and service ecosystems - an industrial pre-pilot
Full text linkIn volatile markets, resilient supply chains and shopfloors are essential to mitigate the impact of disruptions, such as crises, machine failures or quality issues, which result in significant costs. To address these challenges, information about products, processes and resources is required to design, test, and deploy resilience assessment and reconfiguration tools. Nowadays, this information is intended to be made available through data spaces and ecosystems, which necessitates preserving the data sovereignty of the respective companies involved. The architecture proposed by the Flex4res project accommodates these requirements. The implemented pre-pilot use case allows testing and eases the transition for companies
Predictive Maintenance Key Control Parameters for Achieving Efficient Zero Defect Manufacturing
No full textPredictive maintenance is a subbranch of Zero Defect Manufacturing concept. The goal is to achieve higher quality at the final product with the most optimum and efficient way. Predictive maintenance may be applied with various alternative ways for achieving the same goal. The current paper investigates and identifies the key control parameters for an effective predictive maintenance, such as prediction horizon. The identified parameters were implemented in a dynamic scheduling tool and simulations were performed for different manufacturing system layouts and the effect of the identified parameters to each layout were identified. (c) 2021 The Authors. Published by Elsevier B.V.LIC
A comparison of and critical review on cycle time estimation methods for human-robot work systems
No full textAlthough human-robot work systems seem promising in terms of addressing flexibility due to dynamic changes in market demands, SMEs often do not see their potential when it comes to cycle time requirements. The reason for this is the lack of planning methods capable of mapping dynamic interactions between humans and robots and thus supporting the design phase. This paper discusses different methods for cycle time estimation and explains their drawbacks in applying them. Cycle time values defined by existing estimation approaches vary up to 50% compared to cycle time values identified in real applications due to a number of shortcomings. The authors give examples of implications such as safety, task allocation, applied hardware or operator diversity and highlight requirements for a holistic human-robot cycle time estimation method to overcome the aforementioned challenges
A Methodological Approach for Monitoring Assembly Processes
No full textTo enable transparency of assembly systems feedback on physical progress is crucial. However, this currently requires manual operator feedback, which risks reducing the productivity of the assembly system. Technologies such as IoT sensors have been developed to increase the transparency of process states by generating suitable information for further data-driven use. The identification of the correct measurement parameters of the sensors themselves makes industrial application more difficult. This article presents a methodological approach to identify the correct measurement points in the assembly process and shows an efficient way to signal analysis of corresponding IoT sensors. In a test scenario, the assembly progress of a toy truck was continuously monitored based on the methodical procedure and thus the transparency in the assembly process was increased
Towards the 5th industrial revolution: a literature review and a framework for process optimization based on big data analytics and semantics
No full textThe digitalization of modern manufacturing systems has resulted to increasing data generation, also known as Big Data. Although there are several technologies and techniques under the term Data Analytics for gathering such data, their interpretation to information, and ultimately to knowledge remains in its infancy. Consequently, albeit engineers currently can monitor the factory level, optimization is cut off of the data acquisition, and is based on data related methodologies. The focus should be pivoted on designing and developing suitable frameworks for integrating Big Data to process optimization based on the context of information gathered from the shopfloor. This paper aims to investigate the opportunities and the gaps as well as the challenges arising in the current industrial landscape, towards the efficient utilization of Big Data, for process optimization based on the integration of semantics. To that end, a literature review is performed, and a data-based framework is presented
Towards a robust digital production and logistics network by implementing flexibility measures
No full textDigital transformation potentially improves business processes, leading to increased flexibility in manufacturing and logistics networks. Therefore, decision support systems for responsive manufacturing are gaining importance. In this paper, a discrete event simulation model is applied. In order to assess the potential of digital transformation in the context of Industry 4.0 towards production and logistics network robustness, different simulation scenarios are observed. The results contribute to identify potential flexibility measures and to quantify their impact in order to mitigate control uncertainties within production and logistics networks
- …
