In the inaugural issue of the new INFORMS Journal on Data Science, editor-in-chief Prof. Galit Shmueli states what an ideal paper for the journal will look like by highlighting papers from different domains such as transportation, sharing economy, and big data. For the field of health care, Prof. Shmueli highlights the paper “Machine learning approaches for early DRG classification and resource allocation” authored by Daniel Gartner (Cardiff University), Rainer Kolisch (Technical University of Munich), Daniel Neill (New York University), and Rema Padman (Carnegie Mellon University).
In her editorial, Prof. Shmueli writes “[the paper]… incorporates machine learning classification into a mixed-integer programming (MIP)-based resource allocation model for allocating scarce hospital resources. The authors apply their combined ML–MIP methodology to real data from a hospital, showing improved diagnosis-related group (DRG) classification accuracy and practical impact measures (e.g., utilization rate of operating rooms) as compared with the hospital’s current approach. The authors highlight the importance of selecting a concise set of relevant attributes at all stages of care, and also emphasize the generalizability of their approach to similar DRG systems (e.g., in Germany).
The scope of INFORMS Journal on Data Science is to publish top innovative and potentially impactful data science methodologies contributing to decision making in business, management, and industry.