Kühne Logistics University

Extreme weather is often perceived as a major threat to the reliability of rail transportation. This study investigates regional rail operations in central Germany and examines whether severe weather conditions causally affect train arrival delays. We combine several years of operational data (2017–2022) with weather observations from the German Weather Service and define four component-specific treatments for adverse conditions: extreme temperature, strong wind, heavy rainfall, and snow presence. To estimate causal effects, we employ a Double Machine Learning framework based on partial linear regression (PLR–DML) and Causal Forest–DML. By reframing the weather–delay relationship as a causal question rather than a purely predictive one, the study provides evidence on whether extreme weather constitutes a materially relevant source of arrival delays. Across all four weather components, estimated average treatment effects are small in magnitude; analyses of individualized and group-level effects reveal no consistent or robust patterns of heterogeneity across lines, seasons, weekdays, or hours of the day. Robustness checks indicate that these operationally negligible estimates are not sensitive to outliers, rare-event imbalance, or to reasonable perturbations consistent with plausible unobserved confounding. Exploratory mediation analyses are consistent with the interpretation that passenger-flow variables do not materially amplify the already negligible estimated effects of severe weather on delays. Overall, the results suggest that, in this regional network, severe weather does not materially increase arrival delays. The findings underscore the importance of rigorous causal diagnostics for distinguishing perceived from materially relevant sources of delay in transportation reliability studies. • Causal machine learning is used to estimate weather effects on train arrival delays. • Severe weather shows no economically meaningful effect on arrival delays. • Robustness checks and diagnostics confirm the reliability of the causal estimates.

Roll-on roll-off vessels are a popular mode of transport in short-sea shipping. In this domain, appropriate stevedoring procedures are crucial to enhance efficiency. This includes dual cycling, where tugs simultaneously load and unload the vessel. Dual cycling reduces turnaround time, thereby giving the vessel more time to travel and allowing for slow steaming and reduced emissions. We extend the roll-on roll-off dual cycling problem by incorporating a continuous time horizon and differentiating cargo units that are handled by their own drivers and units that must be handled by a tug. We propose a mixed integer linear programming model for generating an efficient schedule that minimizes overall makespan by optimizing the sequence of cargo units and the assignment of cargo units to tugs. To solve instances of real-world size with acceptable computational effort, we provide a range of heuristics, including a biased random-key genetic algorithm. Compared to the linear programming model and on instances of real-world size, the genetic algorithm finds good solutions quickly. We derive managerial insights from a sensitivity analysis and show that dual cycling and strategic positioning of driver-handled units can reduce turnaround time by 14.5%, reducing emissions of the considered vessel by more than 8%. We demonstrate the robustness of these insights in uncertain environments through a simulation study. • Speeding up a vessel’s turnaround enables slower steaming and reduces emissions. • In roll-on roll-off terminals, stevedoring is the main driver of turnaround time • Dual cycling significantly improves a terminal’s stevedoring performance. • A mixed integer linear programming model finds optimal schedules for small problem instances. • A biased-random key genetic algorithm efficiently finds good schedules for large instances.

ABSTRACT In transport and logistics, effective route scheduling stands as a pivotal component to ensure timely deliveries and optimise resource utilisation. Our research addresses the need for improved scheduling solutions across logistics, transportation, and delivery services. It compares three approaches within the framework of discrete event simulation: fixed scheduling, distribution mining, and machine learning. The study evaluates these methods using the mean absolute error (MAE) metric to predict travel times across various routes. The findings reveal the superior performance of machine learning algorithms, which achieve lower MAE scores compared to those derived from distribution mining and conventional scheduling methods with fixed or normalised distributions. Additionally, the study discusses the implications of these results, offering insights into enhancing prediction models and techniques. Use cases illustrate the practical applications of these findings in improving route scheduling processes and advancing the integration of machine learning into simulation-based models.

Abstract. When recording street scenes, the privacy of the people concerned must always be taken into account. Previous work has focused heavily on directly identifiable features such as faces or license plates. However, the visibility of indirectly identifiable objects can also limit the protection of citizens. This is especially true for cars. Regular recordings could be used to infer when a person is at home or where else they have been. The obvious solution would be to simply make cars unrecognizable. However, previous work has shown that this has a negative impact on downstream tasks such as traffic sign recognition. Therefore, in this work, we use generative models to synthetically modify cars. We compare models from the generative adversarial network (DP-GAN, OASIS) and diffusion model communities (Kolors, SDXL) to see which ones are best suited in terms of anonymity, image integrity, and performance. For the GAN-based models, we use an image-to-image translation approach to modify only image sections with cars, while for the diffusion models, we develop two methods that use text-guided image inpainting. We compare the developed methods using established metrics and perform a survey with test subjects. In terms of anonymity, all models achieve convincing results, while diffusion models that generate each car mask individually produce particularly realistic images. In return, GAN-based methods process images more than twice as fast creating a trade-off between image integrity and performance.

Abstract The rapid shift to hybrid work settings has raised concerns about decreased workplace interactions and communication ties. Based on Need‐to‐Belong Theory, this study extends previous research by adopting the concept of ‘concern about relationship loss’ to explore when and how hybrid work leads to the retention of work‐related communication ties. Our research, comprising a quasi‐experimental study and a pre‐registered replication, both conducted in the US, shows that hybrid work increases worries about the potential loss of work‐related relationships (‘concern about relationship loss’). This concern prompts employees to engage in proactive interpersonal and organizational citizenship behaviours (OCBs), but only when organizational identification is high. Notably, the impact of hybrid work and organizational identification is more pronounced for retaining existing communication ties than for losing or forming new ones. Our findings underscore the need to differentiate between concerns about relationship loss and actual isolation. They also emphasize the active role of employees in maintaining communication ties, highlighting the importance of organizational identification and OCBs. This study extends the conventional view of hybrid workers as passive towards a more dynamic understanding of how employees navigate work‐related communication challenges in hybrid work settings.

Supply chains are a critical resource for achieving growth targets of new ventures. Despite this importance, the strategies ventures use to evolve and scale their supply chains in nascent industries have received relatively little attention. This in-depth multiple case study explores how six new ventures develop their supply chains in the emerging plant-based meat industry. Our results demonstrate how supply chains and the entrepreneurial scaling process co-evolve, which is reflected in three characteristic features: the difficulty of managing demand in power-asymmetric relationships, ambidextrous tensions and the multifaceted role of supply chain management. Beyond these commonalities, our analysis identifies three distinct supply chain scaling modes adopted by new ventures: vertical scaling, horizontal scaling, and replication scaling. While growth aspirations drive entrepreneurial engagement in these relationships, they are not the only motivating factor. As both the firm and industry co-evolve, new ventures dynamically adjust their supply chain design to limit supply dependencies, improve core capabilities and minimize distribution cost. Our results offer growth-seeking entrepreneurs a portfolio of actionable supply chain strategies. The findings of our study also contribute to the relatively under-researched intersection of supply chain management and entrepreneurship.

Artificial intelligence is rapidly transforming how research is conceived, executed, published, and shared. This editorial examines the “elephant in the room”: the integration of AI across every stage of the research life cycle and publication pipeline. We trace AI’s expanding footprint on the author side, from sparking novel research ideas, mapping literature, and designing studies, to simulating data, analyzing results, and drafting manuscripts. We also consider AI’s growing role on the journal side, including automated manuscript triage, AI-assisted peer review, decision synthesis, and revision checks. And we discuss AI’s impact on research dissemination. Throughout, we highlight not only the unprecedented opportunities for creativity, efficiency, and accessibility, but also the ethical risks, such as epistemic homogenization, challenges to accountability, and the loss of scholarly craft. We urge researchers, editors, and institutions not to fall into the false binary of blind optimism or blanket skepticism. Instead, we call for deliberate engagement: a principled, transparent, and reflexive partnership between human scholars and machine collaborators. The question is no longer whether we want AI to shape the future of scholarship—it already is. The challenge now is to ensure that what it amplifies is not only our productivity, but our judgment, imagination, and collective responsibility in knowledge creation and dissemination.

Our study explores how production location decisions impact consumer preferences, particularly willingness to pay (WTP), and how to use sustainability perceptions and values as underlying mechanisms. Our research highlights the importance of perceived quality and environmental and social sustainability in shaping consumer behavior, with local production often viewed more sustainable and favorably. We employ a multi-method approach to test our conceptual framework. Study 1 utilizes an online survey with a choice-based conjoint (CBC) analysis to measure stated preferences for products with varying origins. The study covers multiple product categories and captures perceptions of origin, personal values, and demographic data. Study 2 complements these findings through a controlled lab experiment, introducing real monetary incentives to capture revealed preferences and address social desirability bias. Together, these studies provide robust insights into consumer preferences for local and sustainable products. Our findings show that product origin significantly influences consumer decision-making, with local products consistently associated with higher WTP. This preference is driven by product-specific perceptions of higher quality and/or environmental and social sustainability and by personal values. By integrating consumer perspectives into production location decisions, our research highlights the dual benefits of aligning operational strategies with sustainability goals: enhancing profitability through higher WTP and supporting more sustainable supply chains that contribute to broader environmental and social objectives.

Leadership is a complex process involving more than just the actions of one individual. While traditional studies focus on the leader- follower relationship, leadership in modern organizations is shaped by a wide range of formal and informal interactions, emotional expressions, and key partnerships. Leaders influence their teams, but they are also influenced by the people around them, including through emotional signals, social skills, and support from trusted partners, such as personal assistants. Understanding this broader leadership system is key to learning how leaders inspire others, build connections, and succeed in today’s complex workplaces. This symposium presents new ideas about how leaders affect their teams and their own effectiveness, while emphasizing the important role of personal assistants. By expanding the focus beyond just the leader-follower relationship, it provides a deeper understanding of how formal and informal leadership processes interact. The symposium highlights how leadership is not just about direct influence but also about relationships and partnerships. Personal assistants play a critical role as trusted partners who support leaders emotionally and practically, helping manage relationships within the organization. These insights broadens our understanding of leadership by showing how leaders work together with others to achieve organizational success How Leader Expressions of Pride Shape Followers’ Organizational Socialization Author: Alexandra Androulidaki; Kühne Logistics University Author: Prisca Brosi; Author: Manuel J. Vaulont; Northeastern University Exploring Exploitative Leadership in Digital Collaboration: A Qualitative Focus Group Study Author: Melanie Vilser; University of the Federal Armed Forces Munich Author: Lisa Wein; University of the Federal Armed Forces Munich Machiavellianism, Political Skill and Leadership Success: The Socioanalytic Camouflage Hypothesis Author: Svenja Lehr; University of Bonn Author: James A. Meurs; Kennesaw State University Author: Gerhard Blickle; Connecting the Dots: How Do Personal Assistants Exhibit Relations-Oriented Leadership? Author: Hannah M. Baum; WHU - Otto Beisheim School of Management Author: Anna Van Der Velde; WHU - Otto Beisheim School of Management Author: Fabiola Heike Gerpott; WHU - Otto Beisheim School of Management Author: Niels Van Quaquebeke; Kühne Logistics University How Personal Assistants’ Networks Affect Leadership Effectiveness Through Brokerage Behaviors Author: Cécile Emery; University of Exeter Author: Andreas Wihler; University of Exeter Author: Niels Van Quaquebeke; Kühne Logistics University