Refine
Year of publication
Document Type
- Conference Proceeding (68) (remove)
Institute
- Forschungszentrum Business Informatics (68) (remove)
Is part of the Bibliography
- yes (68)
Keywords
- evolution strategies (3)
- Constrained Optimization (2)
- Data science (2)
- Evolution Strategies (2)
- Optimization (2)
- SME (2)
- Supply Chain Management (2)
- Value co-creation (2)
- constrained optimization (2)
- meta-es (2)
Stress testing is part of today’s bank risk management and often required by the governing regulatory authority. Performing such a stress test with stress scenarios derived from a distribution, instead of pre-defined expert scenarios, results in a systematic approach in which new severe scenarios can be discovered. The required scenario distribution is obtained from historical time series via a Vector-Autoregressive time series model. The worst-case search, i.e. finding the scenario yielding the most severe situation for the bank, can be stated as an optimization problem. The problem itself is a constrained optimization problem in a high-dimensional search space. The constraints are the box constraints on the scenario variables and the plausibility of a scenario.
The latter is expressed by an elliptic constraint. As the evaluation of the stress scenarios is performed with a simulation tool, the optimization problem can be seen as black-box optimization problem. Evolution Strategy, a well-known optimizer for black-box problems, is applied here. The necessary adaptations to the algorithm are explained and a set of different algorithm design choices are investigated. It is shown that a simple box constraint handling method, i.e. setting variables which violate a box constraint to the respective boundary of the feasible domain, in combination with a repair of implausible scenarios provides good results.
Towards a high productivity automatic analysis framework for classification. An initial study
(2013)
Through mandatory ESG (environmental, social, governance) reporting large companies must disclose their ESG activities showing how sustainability risks are incorporated in their decision-making and production processes. This disclosure obligation, however, does not apply to small and medium-sized enterprises (SME), creating a gap in the ESG dataset. Banks are therefore required to collect sustainability data of their SME customers independently to ensure complete ESG integration in the risk analysis process for loans. In this paper, we examine ESG risk analysis through a smart science approach laying the focus on possible value outcomes of sustainable smart services for banks as well as for their (SME) customers. The paper describes ESG factors, how services can be derived from them, targeted metrics of ESG and an ESG Service Creation Framework (business ecosystem building, process model, and value creation). The description of an exemplary use case highlighting the necessary ecosystem for service creation as well as the created value concludes the paper.
With the emergence of the recent Industry 4.0 movement, data integration is now also being driven along the production line, made possible primarily by the use of established concepts of intelligent supply chains, such as the digital avatars. Digital avatars – sometimes also called Digital Twins or more broadly Cyber-Physical Systems (CPS) – are already successfully used in holistic systems for intelligent transport ecosystems, similar to the use of Big Data and artificial intelligence technologies interwoven with modern production and supply chains. The goal of this paper is to describe how data from interwoven, autonomous and intelligent supply chains can be integrated into the diverse data ecosystems of the Industry 4.0, influenced by a multitude of data exchange formats and varied data schemas. In this paper, we describe how a framework for supporting SMEs was established in the Lake Constance region and describe a demonstrator sprung from the framework. The demonstrator project’s goal is to exhibit and compare two different approaches towards optimisation of manufacturing lines. The first approach is based upon static optimisation of production demand, i.e. exact or heuristic algorithms are used to plan and optimise the assignment of orders to individual machines. In the second scenario, we use real-time situational awareness – implemented as digital avatar – to assign local intelligence to jobs and raw materials in order to compare the results to the traditional planning methods of scenario one. The results are generated using event-discrete simulation and are compared to common (heuristic) job scheduling algorithms.