Papers by Christos Emmanouilidis
Utilising Learning Analytics in Knowledge Management Adoption for Effective Manufacturing Education
Springer eBooks, Aug 21, 2013
We propose a mathematical model to solve an extension to the to the mixed-model sequencing proble... more We propose a mathematical model to solve an extension to the to the mixed-model sequencing problem with work overload minimization (MMSP-W) for production lines with serial workstations and parallel homogeneous processors and regularizing the required workload. We performed a computational experience with a case study of the Nissan engine plant in Barcelona.
International Journal of Computer Integrated Manufacturing, Aug 6, 2021
In this paper, a decision-making framework for the design of redistributed manufacturing (RdM) ne... more In this paper, a decision-making framework for the design of redistributed manufacturing (RdM) networks is developed. Redistributed manufacturing, a manufacturing paradigm greatly empowered by the Industry 4.0 toolset, is the shift in production towards geographically dispersed interconnected facilities. The framework is context independent, accounts for the collective impact of all decision-making levels on one another in an iterative manner, and incorporates uncertainty. The framework has been applied to a case study in the aerospace spare parts production sector. Results indicated that the RdM paradigm demonstrated considerable improvements in service level when compared with a traditional centralized counterpart, while it was not as competitive with regards to total cost. This paper contributes to the literature on model-based distributed manufacturing systems design under uncertainty, and enables informed decision-making regarding the redistribution of resources and decentraliz...
Embedded event detection for self-aware and safer assets
IFAC-PapersOnLine, 2015
Abstract The increasingly internetworked nature of physical entities and industrial assets is the... more Abstract The increasingly internetworked nature of physical entities and industrial assets is the result of deeper penetration of Internet of Things technologies (IoT) in production environments. Such technologies empower physical product and asset entities to become not only production but also information actors. The physical IoT entities may thus exhibit various levels of intelligence, supported by internet-working, web-based programming and advanced data processing capabilities. At the lower information processing level of physical assets, IoT-enabled entities are now highly integrated with the physical products and assets, forming what is often termed as Cyber-Physical Systems (CPS). They contribute to the upgrade of conventional assets to intelligent ones. For an asset to exhibit intelligent behaviour mere data exchange does not suffice and some form of self-awareness is needed. This paper presents the development of a methodology for embedding event detection to support self-awareness, leading to safer asset operation. Based on wireless sensors, as well as on embedded intelligence, an awareness-enabled asset is a dual entity, consisting of its physical, as well as well its cyber counterpart. This CPS entity is not only capable of performing its intended physical function as part of a production process, but is also capable of handling sensorial monitoring information, process it and perform detection in sensor nodes. The methodology is demonstrated in a lifts monitoring piloting case.
9. Human in the Loop of AI Systems in Manufacturing
A multi-objective genetic algorithm approach to feature selection in neural and fuzzy modeling
Visual Quality Control via eXplainable AI and the Case of Human in the AI Loop
Springer eBooks, 2023
Procedia Manufacturing, 2021
In this paper, a model-based decision-making framework for the design of localized networked prod... more In this paper, a model-based decision-making framework for the design of localized networked production systems under largescale disruptions is developed. The framework consists of optimization and agent-based simulation models that run successively in an iterative manner, gradually improving the performance of the perceived system. The framework integrates uncertainty, provides decisions at different decision-making levels and embeds an algorithm that allows for communication between demand nodes and production sites once inventory shortages occur. The framework has been applied on a case study for the design of localized production and distribution networks, powered by additive manufacturing (AM), in South East England during the early stages of the COVID-19 pandemic outbreak. Results revealed that implementing the framework indeed results in performance improvements to AM-powered production networks, particularly with regards to inventory shortages and lead time.
Development and application of a human-centric co-creation design method for AI-enabled systems in manufacturing
IFAC-PapersOnLine, 2022
2019 IEEE International Conference on Prognostics and Health Management (ICPHM)
Time-series change detection has been studied in several fields. From sensor data, engineering sy... more Time-series change detection has been studied in several fields. From sensor data, engineering systems, medical diagnosis, and financial markets to user actions on a network, huge amounts of temporal data are generated. There is a need for a clear separation between normal and abnormal behaviour of the system in order to investigate causes or forecast change. Characteristics include irregularities, deviations, anomalies, outliers, novelties or surprising patterns. The efficient detection of such patterns is challenging, especially when constraints need to be taken into account, such as the data velocity, volume, limited time for reacting to events, and the details of the temporal sequence. This paper reviews the main techniques for time series change point detection, focusing on online methods. Performance criteria including complexity, time granularity, and robustness is used to compare techniques, followed by a discussion about current challenges and open issues.
Generalised network architectures for environmental sensing: Case studies for a digitally enabled environment
Array
Remote monitoring services are required to meet the very high demands on availability and efficie... more Remote monitoring services are required to meet the very high demands on availability and efficiency of industrial systems. The fast evolution of technologies associated with the deeper penetration of Internet of Things in industry creates considerable challenges for such services. These are related to the whole data lifecycle, encompassing data acquisition, real-time data processing, transmission, storage, analysis, and higher added value service provision to users, with adequate data management and governance needed to be in place. The sheer complexity of such activities the need to ground such processing on sound domain knowledge emphasises the need for context information management. The aim of this paper is to survey and analyse recent literature that addresses internet of things context information management, mapping how context-aware computing addresses key challenges and supports delivering appropriate monitoring solutions.
2018 IEEE 5th International Congress on Information Science and Technology (CiSt), 2018
This paper argues based on evidence from the literature that learning analytics, when undertaken ... more This paper argues based on evidence from the literature that learning analytics, when undertaken by higher education institutions, is not considered within a holistic knowledge management strategy, which could provide significant improvement to the outcomes of learning analytics. Particularly, a synthesis of knowledge extraction via learning analytics and appropriate handling of such knowledge via knowledge management is not typically implemented in higher education practices, but it constitutes a promising path to improving it, and eventually contributes to improving learning services. Essentially, knowledge management can support improvements and innovation in analytics tools, translate an organisation's strategic vision into action, and enable sharing of information among different actors. These are all necessary requirements for effective learning analytics.
International Journal of Computer Integrated Manufacturing, 2021
In this paper, a decision-making framework for the design of redistributed manufacturing (RdM) ne... more In this paper, a decision-making framework for the design of redistributed manufacturing (RdM) networks is developed. Redistributed manufacturing, a manufacturing paradigm greatly empowered by the Industry 4.0 toolset, is the shift in production towards geographically dispersed interconnected facilities. The framework is context independent, accounts for the collective impact of all decision-making levels on one another in an iterative manner, and incorporates uncertainty. The framework has been applied to a case study in the aerospace spare parts production sector. Results indicated that the RdM paradigm demonstrated considerable improvements in service level when compared with a traditional centralized counterpart, while it was not as competitive with regards to total cost. This paper contributes to the literature on model-based distributed manufacturing systems design under uncertainty, and enables informed decision-making regarding the redistribution of resources and decentralization of decision-making. The novelty of this paper is the approach employed to handle complexity, nonlinear interrelationships and uncertainty, within the domain of RdM network design. These computationally demanding attributes are handled through simulation, and only their impact is passed back to an analytical model that generates the RdM network.
IFAC-PapersOnLine, 2020
Internet of things (IoT)-generated data from industrial systems are often collected in nonactiona... more Internet of things (IoT)-generated data from industrial systems are often collected in nonactionable form, thus not directly aiding maintenance actions. Context information management is often seen as an enabler for interoperability and context-based service adaptation, acting as a mechanism for linking data with knowledge to adaptive data and services. Ontology-based approaches for semantic maintenance have been proposed in the past as a data and service mediation mechanism and are adopted here as the starting point employed to develop a context resolution service for industrial diagnostics. The underlying ontology of the context resolution mechanism is relevant to failure analysis of mechanical components. The terminology and relationship between concepts are structured on the basis of relevant standards with a reliability-oriented knowledge grounding. A reasoning mechanism is employed to deliver context resolution and the derived context can add a metadata layer on data or events generated by automated and human-driven means. The approach is applied on a gearbox test rig appropriate for emulating complex misalignment cases met in many manufacturing and aerospace applications.
Procedia Manufacturing, 2018
Under the concept of "Industry 4.0", production processes will be pushed to be increasingly inter... more Under the concept of "Industry 4.0", production processes will be pushed to be increasingly interconnected, information based on a real time basis and, necessarily, much more efficient. In this context, capacity optimization goes beyond the traditional aim of capacity maximization, contributing also for organization's profitability and value. Indeed, lean management and continuous improvement approaches suggest capacity optimization instead of maximization. The study of capacity optimization and costing models is an important research topic that deserves contributions from both the practical and theoretical perspectives. This paper presents and discusses a mathematical model for capacity management based on different costing models (ABC and TDABC). A generic model has been developed and it was used to analyze idle capacity and to design strategies towards the maximization of organization's value. The trade-off capacity maximization vs operational efficiency is highlighted and it is shown that capacity optimization might hide operational inefficiency.
Development of a Context-aware framework for the Integration of Internet of Things and Cloud Computing for Remote Monitoring Services
Procedia Manufacturing, 2018
Abstract With global competition and technological progress, there have been growing demands by i... more Abstract With global competition and technological progress, there have been growing demands by industry for more efficiency in monitoring the health status of the manufacturing equipment in real time. Remote monitoring services in the era of Industry 4.0 are nonetheless faced some challenges such as big data’s 4V challenges (volume, velocity, variety, veracity), scalability, data heterogeneity, as well as relevant to integrating data with domain knowledge. While all these pose problems in conventional monitoring, they become even more challenges when integrating IoT and cloud computing to deliver advanced services to offer infrastructure availability and ubiquitous accessibility. Although it offers many benefits and solution enablers, substantial effort is required to manage and exploit the data generated by "things". Among the key instruments to tackle such complexity is the concept of context information management. This paper proposes a conceptual context-aware framework for the integration of Internet of Things and cloud computing for remote monitoring services.
Lecture Notes in Mechanical Engineering, 2013
The landscape of maintenance and asset management has been reshaped as key technology enablers ar... more The landscape of maintenance and asset management has been reshaped as key technology enablers are making a significant impact on every day applications. The growing maturing of web-based and semantic maintenance, the ubiquity of mobile and situated computing and the lowered costs and increased capabilities of wireless sensing and identification technologies are among the enabling technologies having the most significant impact. They are recognised as the key constituents of e-Maintenance, the technological framework that empowers organisations to streamline their asset management services and data delivery across the maintenance operations chain. This paper takes a look at these key contributing technologies, alongside their adoption prospects and current hurdles preventing the wider penetration of e-Maintenance in industry.
Next Generation Maintenance through the Adaption of E-Maintenance
Conference Proceedings of EUROMAINTENANCE 2010, XX International Maintenance Conference, Verona I... more Conference Proceedings of EUROMAINTENANCE 2010, XX International Maintenance Conference, Verona Italy, 12-14.5.2010, 250-253 A remarkable step forward in maintenance has been taken through the adoption of e-maintenance techniques. The new methodology enables the efficient use of maintenance related information and services wherever and whenever they are needed. New monitoring techniques together with advanced use of mobile devices and RFID technology help the maintenance engineer to assimilate many different information sources, enhancing the quality and cost effectiveness of maintenance decisions. With the help of personal digital assistants (PDA) the maintenance engineer is offered not only highly efficient and practical monitoring and diagnosis tools, but also direct assistance on how to carry out the maintenance tasks effectively and reliably so that the possibility of introducing faults during the execution of maintenance tasks is also minimized. The introduction of maintenance...
Engineering Asset Lifecycle Management, 2010
Today the number of industries putting emphasis on maximising Overall Equipment Effectiveness (OE... more Today the number of industries putting emphasis on maximising Overall Equipment Effectiveness (OEE) is rapidly increasing. Instead of looking only at one aspect of production, modern enterprises seek to plant a series of joint activities, aimed at minimising losses, by improving on Performance, Quality and Availability at the same time. Emphasising these three factors simultaneously leads to the introduction of efficient maintenance, including sound strategies, such as Condition Based Maintenance (CBM). The incorporation of key enabling technologies within a Condition Based Maintenance strategy, such as wireless networking, internet & mobile computing, minature sensing devices and location awareness, has paved the way to the introduction of e-Maintenance. The paper looks at the key features pertaining to the successful implementation of e-Maintenance into modern industry. It then highlights steps taken towards the design and implementation of an e-Maintenance architecture within the EU Integrated Project 'Dynamite'. The paper concludes with a discussion on current challenges and future prospects of e-Maintenance.
Uploads
Papers by Christos Emmanouilidis