In August 2025, the research paper "Deep Stacking Kernel Machines for the Data-Driven Multi-Item, One-Warehouse, Multiretailer Problems with Backlog and Lost Sales" by researcher Chen Zhenyu (first author) from the School of Business Administration was formally published in INFORMS Journal on Computing UTD24 journal. The research paper was jointly completed by Researcher Chen Zhenyu from the School of Business Administration, NEU and Professor Sun Minghe from the University of Texas at San Antonio.
The single-warehouse, multiretailer problems have attracted significant attention from scholars in supply chain management. This research tackles the dual challenges of computational efficiency and resource scheduling performance in single-warehouse, multiretailer problems from a data intelligence perspective. At the methodological level, this research proposes the Deep Stacked Kernel Method (DSKM) and Weighted DSKM to address the single-warehouse, multiretailer problems under two scenarios: backlog and lost sales. These methods handle both complete and censored demand data, respectively. Two solution algorithms are developed, and their global and local optimality properties are explored. At the applied research level, this research demonstrates the proposed model algorithm's robust performance across diverse complex scenarios—including new product management, high-dimensional data and censored demand data analysis—using datasets from two major retail chains. It further explores its potential as a decision-support tool for management practice.
INFORMS Journal on Computing is a bimonthly publication of the Institute for Operations Research and the Management Sciences (INFORMS), focusing on publishing high-quality research at the intersection of operations research and computational science. This journal is recognized by the University of Texas at Dallas as one of the 24 top-tier academic journals for business schools (abbreviated as UTD 24). It enjoys high academic standing within the global management community and serves as a key reference for evaluating management disciplines both domestically and internationally.
Since the start of the 14th Five-Year Plan Period, the School of Business Administration has established scientific systems to safeguard international collaboration and talent development. Adhering to the principles of "quality as the foundation, interdisciplinary integration, benchmarking against authoritative standards, and showcasing mainstream excellence," the School of Business Administration completed revisions to its academic journal classification methodology in April 2025. This initiative guides faculty members and students how to publish high-quality academic papers by focusing on international frontiers and addressing national needs. Since 2021, the School of Business Administration has published a total of seven papers in UTD 24 journals (including three published in 2025), continuously growing its international academic influence.
