This study investigates the feasibility of automated setup of profile wrapping processes on new products using machine learning on past setup examples. The task is characterized by high complexity of the considered production system in combination with highly varying products and a very small available database. This database also reveals ambiguous ground truth due to human, unsystematic preferences. A simple geometric-physical motivated preprocessing is proposed. On the resulting data, a Deep Convolutional Neural Network in the form of an autoencoder is shown to be very suitable for predicting wrapping actions for new products. The good but improvable results are discussed extensively with respect to the technological background and possible solutions are proposed.

In this Paper a methodical approach to hybrid modelling for contextual anomaly detection on time-series data is presented. It enhances widely used proximity-or distribution-based anomaly detection approaches for industrial processes by a hybrid model. This hybrid model consists of a physical model using a priori process knowledge and a data driven model constructed through a machine learning method. The main advantage of the novel approach is that it is capable of detecting contextual anomalies that remain otherwise undiscovered.

Hot forming is a widely used manufacturing process of crash-relevant structural components with complex geometries. In this paper a previously presented method of anomaly detection is further optimized allowing more data sources to be used in the outlier evaluation process. The method is based on a hybrid model consisting of a physical first-principles model of the hot forming press and a neural network in series. It allows a wide range of sensor data to be considered while keeping the anomaly detection process physically explainable.

311

Hot forming is a widely used manufacturing process of crash-relevant structural components with complex geometries. In this paper a method is presented to detect anomalies during the hot forming process giving indications of possible quality defects. The method is based on a physical model of the thermal energy transfer between hot blank, pressing tool and cooling water. The model is built using lumped heat capacities and virtual heat resistances between the components. Within this model the temperature profiles of these components are simulated using input data from real process sensors. The physical model is then combined with a neural network to form a hybrid model. The neural network is trained using the input data and a parameter optimization algorithm and determines the currently optimal parameters of the physical model. During the production, a comparison of the simulated and the real temperature profile reveals anomalies in the hardening process. This way indications for potential quality defects are gained.

BACKGROUND: Biomedical research projects deal with data management requirements from multiple sources like funding agencies' guidelines, publisher policies, discipline best practices, and their own users' needs. We describe functional and quality requirements based on many years of experience implementing data management for the CRC 1002 and CRC 1190. A fully equipped data management software should improve documentation of experiments and materials, enable data storage and sharing according to the FAIR Guiding Principles while maximizing usability, information security, as well as software sustainability and reusability. RESULTS: We introduce the modular web portal software menoci for data collection, experiment documentation, data publication, sharing, and preservation in biomedical research projects. Menoci modules are based on the Drupal content management system which enables lightweight deployment and setup, and creates the possibility to combine research data management with a customisable project home page or collaboration platform. CONCLUSIONS: Management of research data and digital research artefacts is transforming from individual researcher or groups best practices towards project- or organisation-wide service infrastructures. To enable and support this structural transformation process, a vital ecosystem of open source software tools is needed. Menoci is a contribution to this ecosystem of research data management tools that is specifically designed to support biomedical research projects.

In this paper, a distance vector routing algorithm for IP computer networks is presented which is able to aggregate routes automatically and detect and prevent routing loops even on the aggregated routes. Route aggregation is the method of summarizing two or more routes to the corresponding destination IP subnetwork addresses into one common IP prefix address. Thus, it enables non-hierarchical network scalability by keeping the volume of routing and forwarding information within a reasonable size. Nevertheless, under certain circumstances route aggregation can cause network anomalies, e.g., routing and forwarding loops, which can seriously impact the performance and failure safety of the whole computer network. We describe an approach based on well-known Routing Information Protocol (RIPv2) that allows for the automation of route aggregation without manual intervention and still ensures loop-free forwarding paths and, therefore, short convergence times.

The role of an existing systems architecture (SA) in requirements engineering (RE) is recognised as important, but under-researched. A recent exploratory study of ours investigated this issue in a laboratory setting involving student participants. While the initial findings are promising, much work still remains to solidify the results. Therefore, we conducted a replication of the study, and its significant extension, on a large-scale prototypical rail project. Specifically, we identify (i) the effects of SA on RE decisions, (ii) the characteristics of the RE decisions and (iii), the impact of such decisions on development activities and the rail system. The findings of this study have implications on tighter RE-SA integration across subsystems, impact analysis of requirements on SA, and planning and risk management. We also propose three emergent hypotheses from this case study as a driver for future empirical work in RE. This case study involved examining the 10-year history of requirements and architecting decisions in several major components of the rail project. The data collected was from numerous project documents and extensive interviews with the developers and planners.

The Internet has become a complex system with increasing numbers of end-systems, applications, protocols and types of networks. Although we have a good understanding of how data is transferred over the network we cannot observe what happens with our data after sending and before receiving it - how packets traverse through the network and with which QoS characteristics remains unknown. Towards this objective we have developed a multi-hop packet tracking system intended to be used in experimental facilities, such as PlanetLab, where we have made our first tests. This paper describes our packet tracking realization and the results from our prototype implementation.

Before sharing or publishing network traffic data, anonymisation is regarded as a necessary step to protect the privacy of end users. This is especially important for Internet protocol (IP) addresses that could be resolved to a single end user. The most frequently used IP address anonymisation algorithms replace each IP address with a randomly or deterministically computed pseudonym. This static mapping however can present an anonymisation vulnerability, since pattern analysis or spoofing may allow to revert the mapping for selected addresses. In this paper, we propose a new algorithm for anonymising connection data, with the emphasis on IP packet-based network data captured on computer networks. It is worth noting however that except for IP packet-based network data, it is possible to use the proposed algorithm to anonymise any kind of connection data, such as aggregated packet data, t'packet flow data, telephone connection data as well as data associated with the usage of Web services or the accesses to Web servers. We first present the new algorithm and then show that it provides better security against reversing the IP-to-pseudonym mapping at the cost of slightly reduced usefulness of the anonymised data. Specifically, we evaluate the advantages of the proposed algorithm over the most frequently used IP address anonymisation algorithms in terms of the usefulness of the anonymised data with respect to network attack detection methods.