An experimentally calibrated MBD–DEM framework for ice fracture, fragmentation, and dispensing in a refrigerator ice maker
Year: 2026
Type: article
Abstract: In the highly competitive home appliance market, the dispensing and crushing performance of built-in ice makers has become a critical quality attribute affecting both user satisfaction and product competitiveness. However, conventional product development processes still rely heavily on reactive trial-and-error approaches, which may lead to substantial economic losses during the later stages of mass production. Ice is a complex material whose mechanical properties and failure behavior are highly sensitive to initial conditions, including temperature, geometry, and porosity. Therefore, a high-fidelity numerical model capable of accurately representing these characteristics is essential. This study proposes a discrete element method (DEM)-based modeling methodology for constructing an ice block model that captures realistic mechanical behavior. Physical property data were obtained from cubic ice samples produced using an actual ice maker, and the DEM model was subsequently refined through a systematic parameter calibration procedure. Furthermore, a coupled simulation framework integrating multibody dynamics (MBD) and DEM was established to quantitatively predict the performance of two primary operating modes: Cube Mode, in which ice is dispensed without fragmentation, and Crush Mode, in which ice is fragmented prior to discharge. By evaluating ice mass flow and crushing characteristics in a virtual environment representative of actual operating conditions, the proposed framework enables predictive performance assessment during the product design stage. The developed approach provides a robust analytical basis for improving the design and performance evaluation of ice dispensing and crushing mechanisms.
Source: Scientific Reports
Institutions: Kyungpook National University, LG (South Korea)
Language: English
Subfield: Computational Mechanics
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,690
Funder: Kyungpook National University
OpenAlex ID: W7165739108