Multinationalen Unternehmen wird im Kontext ihres Wirkens in globalen Lieferketten eine bedeutende Rolle für die Erreichung globaler Nachhaltigkeitsziele zugeschrieben. Sie nehmen unter anderem durch Instrumente eines Sustainable Supply Chain Managements (SSCM) Einfluss auf die Umsetzung sozialökologischer Standards und die Einhaltung von Mindestanforderungen an ihren weltweiten Produktions- und Zulieferstandorten, etwa im Bereich von Menschenrechten oder des Umweltschutzes. In diesem Kontext bildet das Engagement in Brancheninitiativen als horizontale Multi-Akteurs-Netzwerke einen bedeutenden strategischen Baustein im Rahmen des SSCM vieler fokaler Unternehmen. Diese horizontalen Kooperationen gewinnen durch ihre Verankerung in internationalen Leitlinien und Zielsetzungen sowie in nationalen Regulierungen zur unternehmerischen Sorgfaltspflicht strukturell zunehmend an Bedeutung. Ihr potenzieller und tatsächlicher Beitrag zur Umsetzung der globalen Nachhaltigkeitsziele findet jedoch im wissenschaftlichen Kontext bisher noch wenig Berücksichtigung. Aus diesem Grund setzt sich der vorliegende Beitrag mit dem nachhaltigkeitsbezogenen Wirkungspotenzial SSCM-orientierter Brancheninitiativen auseinander und wirft einen Blick auf das in öffentlich zugänglichen Dokumenten formulierte Selbstverständnis verschiedener Initiativen. Dabei werden die dort zu findenden Erfolgskonzepte aus den Perspektiven des betrieblichen Nachhaltigkeitsmanagements („Performance“) und einer übergeordneten Nachhaltigkeitsgovernance („Wirkung“) eingeordnet, etwa in den Bereichen Menschenrechte und Umweltschutz.

Natural human locomotion contains variations, which are important for creating realistic animations. Most of all when simulating a group of avatars, the resulting motions will appear robotic and not natural anymore if all avatars are simulated with the same walk cycle. While there is a lot of research work focusing on high-quality, interactive motion synthesis the same work does not include rich variations in the generated motion. We propose a novel approach to high-quality, interactive and variational motion synthesis. We successfully integrated concepts of variational autoencoders in a fully-connected network. Our approach can learn the dataset intrinsic variation inside the hidden layers. Different hyperparameters are evaluated, including the number of variational layers and the frequency of random sampling during motion generation. We demonstrate that our approach can generate smooth animations including highly visible temporal and spatial variations and can be utilized for reactive online locomotion synthesis.

Hierarchical neural networks with large numbers of layers are the state of the art for most computer vision problems including image classification, multi-object detection and semantic segmentation. While the computational demands of training such deep networks can be addressed using specialized hardware, the availability of training data in sufficient quantity and quality remains a limiting factor. Main reasons are that measurement or manual labelling are prohibitively expensive, ethical considerations can limit generating data, or a phenomenon in questions has been predicted, but not yet observed. In this position paper, we present the Digital Reality concept are a structured approach to generate training data synthetically. The central idea is to simulate measurements based on scenes that are generated by parametric models of the real world. By investigating the parameter space defined of such models, training data can be generated in a controlled way compared to data that was captured from real world situations. We propose the Digital Reality concept and demonstrate its potential in different application domains, including industrial inspection, autonomous driving, smart grid, and microscopy research in material science and engineering.

The simulation of humanoid avatars is relevant for a multitude of applications, such as movies, games, simulations for autonomous vehicles, virtual avatars and many more. In order to achieve the simulation of realistic and believable characters, it is important to simulate motion with the natural motion style matching the character’s characteristic. A female avatar, for example, should move in a female style and different characters should vary in their expressiveness of this style. However, the manual definition, as well as the acting of a natural female or male style, is non-trivial. Previous work on style transfer is insufficient, as the style examples are not necessarily a natural depiction of female or male locomotion. We propose a novel data-driven method to infer the style information based on individual samples of male and female motion capture data. For this purpose, the data of 12 female and 12 male participants was captured in an experimental setting. A neural network based motion model is trained for each participant and the style dimension is learned in the latent representation of these models. Thus a linear style model is inferred on top of the motion models. It can be utilized to synthesize network models of different style expressiveness on a continuous scale while retaining the performance and content of the original network model. A user study supports the validity of our approach while highlighting issues with simpler approaches to infer the style.

Servitization and product-based services are used to support the integration of products and services with customers, enabling companies to maintain a competitive advantage in their markets. However, in order to achieve these capabilities is necessary to have flexible processes and services. The enterprise needs to become self and context aware to meet these new challenges, and with the Internet-of-Things development, resources can be shared across companies to reduce costs. Enterprise integration is an essential component of enterprise and service engineering but traditional modelling techniques need to evolve and become more dynamic, separating concerns but at the same time promoting knowledge reuse. This paper contributes to a more flexible environment for information systems and service development, proposing a model-driven framework for dynamic process development in the enterprise of the future. It applies the concept of the liquid-sensing enterprise following the Osmosis processes paradigm, supporting the enterprises to model and design their processes at business and technical level. With the support of a modelling toolbox the enterprises are able to parameterize their processes and accelerate the advancement from the design phase into services execution phase.

Recent market trends require extremely short product life cycles to cope with individual customer requirements. The key technology to deal with these requirements is plug-and-produce, which reduces engineering time to change the production lines rapidly. This challenge is solved by the concept of a modular factory which allows to reconfigure individual machine stations without the need of extensive engineering effort. Current solutions focus on hardware approaches such as modular frames, cables and sensors. However software is still a barrier where the change of production line needs the reprogramming of individual devices. To deal with this challenge, the authors utilize the BEinCPPS architecture, which is built from open-source software components.

Many of the existing data-driven human motion synthesis methods rely on statistical modeling of motion capture data. Motion capture data is a high dimensional time-series data, therefore, it is usually required to construct an expressive latent space through dimensionality reduction methods in order to reduce the computational costs of modeling such high-dimensional data and avoid the curse of dimensionality. However, different features of the motion data have intrinsically different scales and as a result we need to find a strategy to scale the features of motion data during dimensionality reduction. In this work, we propose a novel method called Scaled Functional Principal Component Analysis (SFPCA) that is able to scale the features of motion data for FPCA through a general optimization framework. Our approach can automatically adapt to different parameterizations of motion. The experimental results demonstrate that our approach performs better than standard linear and nonlinear dimensionality reduction approaches in keeping the most informative motion features according to human vision judgment.

Reorienting the cooperative structure in selected Eastern European countries: Case study on the former German Democratic Republic , Reorienting the cooperative structure in selected Eastern European countries: Case study on the form... , مرکز فناوری اطلاعات و اطلاع رسانی کشاورزی

For ergonomic assessment of manual assembly tasks, digital simulation has received increasing attention due to its efficiency compared to physical prototypes. One of the crucial parts of digital simulation is an accurate animation of the digital human model (DHM). Current digital simulation tools such as Delmia V5 require interactive manual editing to produce animations, which is time consuming and can look unnatural. On the other hand, data-driven motion synthesis that is based on motion capture data can produce natural motions with little user involvement. The practical difficulty lies in processing motion data into a parameterized motion model. A common approach is decomposing motions and categorizing them into finite short motion primitives. For each motion primitive, motion data is represented as a numerical vector, on which functional principal component analysis (FPCA) is applied to reduce dimensionality. In this work, different ways of representing joint angles from motion capture data are explored: Euler angle, quaternion and exponential map. The data representations are evaluated for their reconstruction error with FPCA. In the tests, quaternion representation shows best performance for motion data representation, which contradicts a preference in literature for exponential map representation. Therefore, quaternion representation is considered appealing for statistically modelling motion data.

The emerging Liquid-Sensing Enterprise (LSE) concept provides to manufacturing enterprises the required enablers to modernize traditional strategies for product design and validation. The proposed osmosis processes integrates innovative processes and paradigms comprising a MDA/MDI based approach with the focus to potentiate the generation of technological innovations. These osmosis processes address the existing real, digital and virtual related data of specific products design supported by sensing assets to facilitate ready-to-run business processes, to then perform efficient product development and validation. This paper integrates model-driven techniques and runtime environments as well as novel form of information access and visualization for systems design and process development. The developments are accompanied with a real example from a manufacturing environment of engine camshafts.

In this paper a new architecture named CPS4MRO, that goes a step further previous research development and industry-oriented projects led by the authors, is specified. The aim is to provide in the near future an optimized planning and control of MRO operations of a fleet of complex transportation systems using the paradigm of Cyber-Physical Systems.

With the spreading of sensors networks, the management of business reality has been progressively changed exploiting t he inherent features of the future IoT and smart objects. Today’s enterp rises need to become self-aware not only in terms of their networked eco system but also in face of their inner sub-systems and devices. Indeed , the integration of service and event technology, smart agents and “sma rt objects” enables enterprise innovation and competitiveness in a glob alized economy. Those enterprises would be capable of sensing and r eacting from physical to complex business incitements, maximizing their innovation potential and sustaining interoperability along the operational life cycle. This paper presents the OSMOSE paradigm for the liquid-sensing enterprise, a novel concept being developed in the context of the Future Internet to enable new business models for the sens ing enterprise. Initial developments and a conceptual architecture are pres ented and illustrated with the help of an on-going industrial scenario.

The Sensing-Liquid Enterprise paradigm enhances sensing capabilities of the Sensing Enterprise with fuzzy boundaries of the Liquid Enterprise by interconnecting real, virtual and digital worlds through semi-permeable membrane behavior. Shadow images of the different worlds need to be kept consistent. Osmotic data flows between the real, digital and virtual world allow events to break out of their inner world behavior and to advance to inter-world events. This paper combines semantic web technologies and complex event processing to enable osmotic event detection and processing for the Sensing-Liquid Enterprise. Events are enriched with semantic information and examined for inter-world relevance. The presented approach is accompanied with its application in the OSMOSE Project.

The ATOP (Agent-Based Technologies and Applications for Enterprise Interoperability) workshop series focuses on technologies that support interoperability in networked organizations, on successful app

As of today, the behavior of avatars in virtual worlds is usually realized by script sequences which provide the illusion of intelligent behavior to the user. In the research project ISReal, our research group developed the first platform for deploying virtual worlds based on Semantic Web technology, which enables agents to reason about and plan with semantically annotated 3D objects. Powerful tool support is required to design agents which exploit the functionality of the ISReal platform. We decided to reuse existing facilities provided by the model-driven Bochica framework for AOSE and extended it with a platform model for agents situated in semantically-enhanced simulated realities.