A variety of evaluation methods for user interfaces (UI) exist such as usability testing,

cognitive walkthrough, and heuristic evaluation. However, UIs such as guidance systems at transit

hubs must be evaluated in their intended application field to allow the effective and valid identi-

fication of usability flaws. However, what if evaluations are not feasible in real environments, or

laboratorial conditions cannot be ensured? Based on adapted heuristics, in the present study, the

method of heuristic evaluation is combined with immersive Virtual Reality (VR) for the identification

of usability flaws of dynamic guidance systems (DGS) at transit hubs. The study involved usability

evaluations of nine DGS concepts using the newly proposed method. The results show that compared

to computer-based heuristic evaluations, the use of immersive VR led to the identification of an

increased amount of “severe” usability flaws as well as overall usability flaws. Within a qualitative

assessment, immersive VR is validated as a suitable tool for conducting heuristic evaluations in-

volving significant advantages such as the creation of realistic experiences in laboratorial conditions.

Future work seeks to further prove the suitability of using immersive VR for heuristic evaluations

and compare the proposed method to other evaluative methods.

Deep Learning has enabled many advances in machine learning applications in the last few years. However, since current Deep Learning algorithms require much energy for computations, there are growing concerns about the associated environmental costs. Energy-efficient Deep Learning has received much attention from researchers and has already made much progress in the last couple of years. This paper aims to gather information about these advances from the literature and show how and at which points along the lifecycle of Deep Learning (IT-Infrastructure, Data, Modeling, Training, Deployment, Evaluation) it is possible to reduce energy consumption.

In the past decade, globalization and digitization have not only changed the way we work, but also the environment in which we work. More and more companies are introducing desk sharing office concepts in which employees must share a workstation. However, this poses challenges for ergonomic workplace design as constant and ergonomically correct workstation settings can hardly be guaranteed. Neglecting ergonomics at workplace, though, can cause musculoskeletal disorders. Therefore, a concept and prototype for a system are proposed which automatically adjusts the workstation to the individual's anthropometric characteristics. A setup of different mechanical and electronical components using microcontrollers, ultrasonic distance sensors and linear actuators assures an automatic adjustment where users only must sign in with their ID. An initial field study shows that the system can achieve high user acceptance. Simplicity, speed, and convenience are seen as added value of the system. The results have potential for future studies.

Der Mensch verfügt mit seiner Intuition und speziell mit seiner inspirierenden Intuition, über eine Quelle der Evolution. Intuition ist bislang eine spezielle Fähigkeit des Menschen (Müller 2021-2, S. 62). Die Erfindungskraft des Menschen, hat mit der künstlichen Intelligenz eine Spezies hervorgebracht, die eine Gefahr für die eigene nachhaltige Zukunft werden könnte. Segen oder Fluch lagen noch nie so nahe beieinander. Der Umgang mit der Künstlichen Intelligenz erfordert ein weitsichtiges Controlling-Denken, wollen wir nicht Gefahr laufen, unsere Vormachtstellung auf der Erde zu verlieren. Nachhaltigkeit im herkömmlichen Sinne sollte ergänzt und überdacht werden, wollen wir die KI-Entwicklung aktiv steuern und nutzen, statt sie passiv hinzunehmen oder schlimmstenfalls zu erleiden. Für das Controlling erwächst in Bezug auf die Nachhaltigkeit eine existenzielle-humanistische Dimension.

Artificial Intelligence (AI) is drastically transforming the world around us. Rather than replacing humans, hybrid intelligence combines human and machine intelligence to leverage each of their individual strengths. We summarize different requirements and approaches identified to achieve hybrid intelligence and focus on conversational AI to build a cognitive agent that supports knowledge management within an organization. The agent automatically extracts knowledge from artifacts provided or published by the us- ers. In addition, the knowledge base steadily grows while the agent talks to the users and the users provide feedback and the system is continuously learning to extract new types of entities and relations to answer more questions based on the knowledge graph and to access other sources of information. The first types of entities and relations extracted already support users in finding colleagues with relevant skills or inter- ests. Based on information provided by the agent, collaboration among employees and, thus, knowledge sharing and transfer is encouraged. The collaboration between the cognitive agent as an AI artifact and employees combined with a system that learns and adapts while in use stressing explainability and trust in its answers entails a step towards hybrid intelligence.