Tobias Meisen | Bergische University Wuppertal (Germany) (original) (raw)

Papers by Tobias Meisen

In the last years, enterprises increase their effort to collect large amounts of data from many h... more In the last years, enterprises increase their effort to collect large amounts of data from many heterogeneous data sources and store it in modern architectures like data lakes. However, this approach faces different drawbacks for finding and understanding data sources. Ontology-Based Data Access (OBDA) originating from the Semantic Web enables a homogeneous access to the data sources by using a mapping, called semantic model, between a data source and a target ontology. However, OBDA requires a detailed ontology, which is usually created by ontology engineers and domain experts resulting along with high effort for designing and maintaining. To overcome these limitations, we develop an approach consisting of a knowledge graph, which features an internal growing ontology and linked data-source specific semantic models. The ontology continuously evolves on-demand based on newly added data sources along with their corresponding semantic models, which are created by domain experts. To ensure the knowledge graph's stability, we develop an intuitive user-oriented assistant and combine it with a semi-supervised evolving strategy that assists the user with the help of external knowledge bases. We evaluate accuracy and usability of our approach by conducting a user study with a heterogeneous group of participants that define semantic models upon pre-defined data sets. The results show that semantic models become more objective and consistent with our provided user assistant and thus lead to a knowledge graph with higher interconnectivity and stability.

Lecture Notes in Computer Science, 2017

This work contributes to a technology stack that pursues the goal of integrating intelligent enti... more This work contributes to a technology stack that pursues the goal of integrating intelligent entities in a production environment by means of communication technologies based on scalable interfaces supporting semantic modeling. The proposed architecture is realized based on a model-driven interconnection of multi-agent systems with OPC Unified Architecture. The integration of these technologies enables a usage of intelligent mechanisms within modern production sites while ensuring semantic integrity during all communication processes and compliance to essential security standards. The goal of this work is to enable a autonomous, reactive production by means of intelligent communication in autonomous systems. By making use of a model-based representation of intelligent software agents, an integration of cyber-physical systems with products and production units in manufacturing systems can be realized. The integrability of these multi-agent systems with high-level applications in terms of generic vertical interoperability is shown by means of seamless information exchange with an ERP system.

Journal of Intelligent Manufacturing, Feb 5, 2022

The deep reinforcement learning (DRL) community has published remarkable results on complex strat... more The deep reinforcement learning (DRL) community has published remarkable results on complex strategic planning problems, most famously in virtual scenarios for board and video games. However, the application to real-world scenarios such as production scheduling (PS) problems remains a challenge for current research. This is because real-world application fields typically show specific requirement profiles that are often not considered by state-of-the-art DRL research. This survey addresses questions raised in the domain of industrial engineering regarding the reliability of production schedules obtained through DRL-based scheduling approaches. We review definitions and evaluation measures of reliability both, in the classical numerical optimization domain with focus on PS problems and more broadly in the DRL domain. Furthermore, we define common ground and terminology and present a collection of quantifiable reliability definitions for use in this interdisciplinary domain. Concludingly, we identify promising directions of current DRL research as a basis for tackling different aspects of reliability in PS applications in the future.

Proceedings of the 24th International Conference on Enterprise Information Systems

Der Wunsch nach individueller Mobilität und individuellem Waren-und Güterverkehr im Spannungsfeld... more Der Wunsch nach individueller Mobilität und individuellem Waren-und Güterverkehr im Spannungsfeld von Zeit, Kosten und Qualität erfordert dynamische Innovationsprozesse, welche mit Blickrichtung auf die Faktoren Sicherheit, Funktionalität, Umwelt und Globalisierung neben neuen Technologien auch zukunftsweisende Logistiksysteme, Logistikkonzepte und logistische Dienstleistungen umfassen. In dem Beitrag In drei Wochen von Duisburg nach Peking-Ist die Bahn eine Transportalternative für die deutsche Stahlindustrie? von Carola Obermeier-Hartmann und Eric Sucky wird die von China geplante "Neue Seidenstraße" betrachtet, d.h. Chinas "Belt and Road Initiative" (BRI), welche neue Handelsrouten zwischen Asien, Afrika und Europa schaffen sowie alte wiederbeleben soll. Der Beitrag beschäftigt sich mit der Frage, ob die Bahn eine Alternative zum Schiff auf der Route von Europa nach Asien sein kann. Während in bisherigen Veröffentlichungen insbesondere auf die Zeitersparnis auf der Route von China nach Deutschland abgestellt wurde und produktunspezifische Analysen erfolgten, fokussieren sich Carola Obermeier-Hartmann und Eric Sucky konkret auf das Produkt Stahl. Aspekte des Supply Chain Managements stehen im Fokus der weiteren Beiträge. Supply Chain Management beschreibt die an den Kundenbedürfnissen ausgerichtete, kooperative Planung, Steuerung und Kontrolle von produkt-oder produktgruppenbezogenen, unternehmensübergreifenden Wertschöpfungsnetzwerken mit dem Ziel, die Wettbewerbsfähigkeit sowohl der einzelnen Supply Chain-Akteure als auch der gesamten Supply Chain zu erhöhen. Es umfasst dabei sowohl die zielgerichtete Gestaltung der einzelnen Supply Chain-Ebenen als auch die zielgerichtete Koordination der Prozesse in der Supply Chain.. Um Supply Chains robust, resilient und agil zu gestalten, bedarf es daher Flexibilitätspotenziale. Da deren Aufbau mit Kosten verbunden ist, ist Flexibilität an der Stelle im Güterfluss einer Supply Chain zu positionieren, an dem sie einen hohen Nutzen erzeugt. Immanuel Zitzmann untersucht in seinem Beitrag Positionierung von Flexibilität in der Supply Chain, wo Flexibilitätspotenziale in einer Supply Chain den höchsten Beitrag zur Bewältigung von Unsicherheiten leisten können. Dies geschieht anhand der Ergebnisse aus zwei Simulationsstudien. Das aktive Management komplexer Wertschöpfungsnetzwerke, d.h. Supply Chains, erfordert die Kooperation der in einer Supply Chain agierenden, rechtlich und wirtschaftlich eigenständigen Unternehmen. Die Kooperationsbereitschaft ist insbeson

2019 International Conference on High Performance Computing & Simulation (HPCS), 2019

With the latest advances in digitalization and Industry 4.0, the manufacturing industry is collec... more With the latest advances in digitalization and Industry 4.0, the manufacturing industry is collecting more and more production data. However, with the increasing interconnection of machines, not only the volume but also the variety of data is being expanded. The data life cycles of collection, processing, combining, analyzing and feeding new findings back into sources are becoming increasingly challenging for data scientists to complete. Reference architectures such as the RAMI 4.0 provide conceptual guidelines to address these problems. In this paper, we focus on the implementation of an agent-based architecture that is in line with RAMI 4.0. This architecture implements the guidelines provided by RAMI 4.0 by applying modern approaches from the areas of data lake based data acquisition, semantic description, look up and processing as well as information utilization.

Benchmark Corpus for semantic labeling and modeling. This corpus contains 101 data sets from diff... more Benchmark Corpus for semantic labeling and modeling. This corpus contains 101 data sets from different open data portals.<br> Each data set consists of the following data: Raw csv data [rawdata_csv] Large json data sample [json_sample_large] Small json data sample [json_sample_small] Textual description / Metadata [descriptions] Semantic model as rdf/xml [semantic_models] Mappings describing mapping between raw data attributes and concepts from the ontology [mappings] List of attributes that have been ignored during modeling [ignored_attributes] Additionally the corpus contains a target ontology and an additional ontology for mappings to the PLASMA platform, both as rdf/xml [ontology]. The individual data sets are licensed by the licenses specified in the attached Excel sheet (DataSetOverview.xlsx) These data are provided "as is", without any warranties of any kind. The data are provided under the Creative Commons Attribution 4.0 International license.

2018 International Conference on Innovations in Information Technology (IIT), 2018

In times where large amounts of time-dependent data is generated, the importance of time interval... more In times where large amounts of time-dependent data is generated, the importance of time interval data sets in general and similarity analyses on these in particular continues to increase. In this context, various approaches regarding the comparability of two time interval data sets have been developed in recent years. The STIDes approach as a bottom up approach offers on the one hand the possibility to focus on individual properties of the intervals, on the other hand it allows time delays or scaling to be taken into consideration. In this paper, we take a closer look at the management of time delays and different scales and show that a similarity analysis using STIDes can be completed in polynomial time. Furthermore, we improve the handling of cardinality differences in the data sets to be compared.

Enterprise Information Systems, 2019

Data lakes offer enterprises an easy-to-use approach for centralizing the collection of their dat... more Data lakes offer enterprises an easy-to-use approach for centralizing the collection of their data sets. However, by just filling the data lake with raw data sets, the probability of creating a data swamp increases. To overcome this drawback, the annotation of data sets with additional meta information is crucial. One way to provide data with such information is to use semantic models that enable the automatic interpretation and processing of data values and their context. However, creating semantic models for data sets containing hundreds of data attributes requires a lot of effort. To support this modeling process, external knowledge bases provide the background knowledge required to create sophisticated semantic models.

2019 IEEE Symposium Series on Computational Intelligence (SSCI), 2019

With the trend of globalization and digitalization, many transnational companies are continuously... more With the trend of globalization and digitalization, many transnational companies are continuously collecting and storing unstructured text data in different languages. To exploit the business value of such high-volume multilingual text data, cross-lingual information extraction utilizes machine translation and other natural language processing (NLP) techniques to analyze this data. However, results of these analysis heavily depend on the order in which the tasks are performed as well as the used machine translation and NLP approaches or trained models. In this paper, we defined and evaluated a series of cross-lingual text processing pipelines for English and Chinese language. We therefore combine multiple commercial machine translation services with different automatic keyphrase extraction and named entity recognition techniques and evaluate their performance with regards to the order of execution. Hence, we evaluate the combination of machine translation systems and natural language processing techniques with two processing sequences in our experiment. One is to translate the document before extracting keyphrase and named entities. The other is to translate the processing results. The experiment outcomes indicate that translating documents is a better choice than the other way around in both tasks. However, there exists a substantial disparity between the performance of the cross-lingual text processing pipelines and the corresponding monolingual references.

ArXiv, 2020

The need for more transparency of the decision-making processes in artificial neural networks ste... more The need for more transparency of the decision-making processes in artificial neural networks steadily increases driven by their applications in safety critical and ethically challenging domains such as autonomous driving or medical diagnostics. We address today's lack of transparency of neural networks and shed light on the roles of single neurons and groups of neurons within the network fulfilling a learned task. Inspired by research in the field of neuroscience, we characterize the learned representations by activation patterns and network ablations, revealing functional neuron populations that a) act jointly in response to specific stimuli or b) have similar impact on the network's performance after being ablated. We find that neither a neuron's magnitude or selectivity of activation, nor its impact on network performance are sufficient stand-alone indicators for its importance for the overall task. We argue that such indicators are essential for future advances in t...

2019 IEEE 13th International Conference on Semantic Computing (ICSC), 2019

Procedia Manufacturing, 2019

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.

2016 IEEE 14th International Conference on Industrial Informatics (INDIN), 2016

In terms of current industrial manufacturing sites, a major challenge is to deal with growing com... more In terms of current industrial manufacturing sites, a major challenge is to deal with growing complexity by enabling intelligence on the shop floor of existing production processes. A possible solution to reach this goal consists in an integration of smart cyber-physical production systems into the automation systems of a production. One promising approach to do so is based on the agent paradigm. By deploying Multi-Agent Systems into the manufacturing components, each production step is able to gain a self-representation and to achieve intelligent behavior of the entire system. One problem though is the formalization of agent based systems and their communication among each other, which is currently rather hard-coded or application-specific. In this research paper, we propose an architectural approach for a Multi-Agent System that is based on OPC UA as modeling interface and as semantic approach for the integration of agent-based systems into existing manufacturing sites. For this purpose, we define a domain ontology for the representation of intelligent software agents and for the mapping of an agent-based communication by making use of the OPC UA meta model. Due to this conceptual approach, the integration of intelligent entities such as agents into grown manufacturing systems can be performed in a structured and well-defined way as well as by using existing interfaces and semantic standards. The according agent representation intends to upgrade all production resources that can be linked through OPC UA with intelligent behavioral skills. We evaluate the proposed concept by means of an Industry 4.0 demonstrator implementing agents for the representation actual manufacturing machinery based on Raspberry Pi devices.

Cyber-Physical Systems, 2017

Abstract Cyber-physical production systems (CPPSs) are the fundamental basis for the realization ... more Abstract Cyber-physical production systems (CPPSs) are the fundamental basis for the realization of the German initiative “Industrie 4.0,” which covers not only the usage of intelligent embedded devices and their interconnectedness, but also models for describing different processes according to the product’s life cycle. This article focuses on challenges regarding the integration of different views on urgent aspects, technologies, and paradigms to formulate one consistent modeling approach. Different use cases then describe the application of modeling and implementation concepts as well as benefits of new possibilities for process control. These are: model-based human-robot interaction for flexible assembly automation, a cloud-based approach for advanced condition monitoring, and product-centered control in the Laboratory for Machine Tools and Production Engineering (WZL)’s Smart Automation Lab.