In this work we consider entropy stable discontinuous Galerkin methods applied to nonconservative hyperbolic systems. We introduce a new class of entropy conservative fluctuations that allow us to construct entropy conservative schemes without any system-specific derivations. We demonstrate that a loss of entropy symmetrization for nonconservative systems restricts the design of entropy stable fluctuations and propose a novel blending procedure to construct entropy stable dissipation terms from general numerical viscosity matrices. The resulting methodology is applied to develop a high-order, entropy stable, and well-balanced approximation for the Saint-Venant-Exner system. Numerical tests are presented to verify the theoretical findings and demonstrate the performance and robustness of the proposed scheme.
Overset grid methods handle complex geometries by overlapping simpler, geometry-fitted grids to cover the original, more complex domain. However, ensuring their stability-particularly at high orders-remains a theoretical challenge: although overset grid methods perform robustly in extensive practical use, general stability proofs are not available. In this work, we address this gap by developing a discrete counterpart to the recent well-posedness analysis of Kopriva, Gassner, and Nordstr & ouml;m for continuous overset domain initial-boundary-value problems. To this end, we introduce the novel concept of sub-cell summation-by-parts (SBP) operators. These discrete derivative operators mimic integration by parts at a sub-cell level. By exploiting this sub-cell SBP property, we develop provably conservative and energy-stable overset grid methods for fixed one-dimensional overset domains that do not change with time or under grid refinement, providing a step toward stability proofs for overset grid methods based on the energy method.
While robots offer clear benefits in fulfilling functional tasks, service providers remain cautious when using robots in roles that include interaction with customers, as their social benefits may be questionable. This paper examines how the social presence of the robot makes service encounters with service robots satisfactory to customers. Across two online studies and one laboratory study, results show that social presence positively influences customer satisfaction with service encounters. Service innovativeness mediates this relationship, but only for more innovative customers. The results highlight the importance of balancing social presence and anthropomorphism in service robots, as robot anthropomorphism negatively moderates the relationship between social presence and satisfaction. The paper contributes to service robot research by explaining how social presence shapes customer responses and when its benefits vary. Managerially, the findings suggest that providers should implement robots high in social presence while carefully balancing their level of anthropomorphism to maximize their benefits.
We analyse the distribution of payments with respect to income, home and work location, commuting across the cordon, and other socio-economic characteristics in the two full-scale congestion charging systems in Stockholm and Gothenburg. We use unique population-wide Swedish micro-registry data linking all residents to their vehicles and to annual congestion charges paid during 2018–2020. The payments are weakly regressive, with Suits indices of −0.02 (Gothenburg) and −0.04 (Stockholm). However, the income gradient is several times smaller once we control for time-invariant individual-specific factors. There is substantial heterogeneity in annual payments within income groups—so large that the regressive/progressive classification becomes of limited practical relevance. Only a small group pays frequently, on a daily basis, while most individuals pay only occasionally, implying that equity assessments based on self-reported behaviour in a “normal week” or on single-day travel observations can be misleading. When controlling for individual-specific effects, commuting across the cordon is only weakly associated with annual payments.
Assume that a stick with unit length is broken totally randomly into n pieces. The kth largest piece (with positive integer k≤n) is called the kth largest uniform spacing induced by n uniformly distributed order statistics on (0,1). Exact large deviations of the kth largest uniform spacing away from its most probable size lnn/n are studied in this paper, together with an application in goodness-of-fit tests.
