I-Corps: Real-Time Update System for Hospitals

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is a significant increase in the efficiency of hospital procedure scheduling. The project analyzes the sources of inefficiencies in the planning, staffing, coordination, and execution of the various phases of surgeries and associated procedures. Preliminary analysis of hospital data shows that the throughput of surgical units can be increased significantly with the same staffing while reducing the delays and improving the working conditions of the personnel. Delays in obtaining and communicating updates on the status of surgeries and on actions that personnel should perform are major causes of inefficiency, as is the randomness of the duration of tasks. We expect the methodology to be widely applicable to the coordination of teams in services and transportation industries. This I-Corps project is based on a machine learning approach to the optimization of real-time messaging tuned using actual hospital data. The technical novelty is a formulation of the real-time policies that incorporates the messages between the main actors and enables the use of machine learning to optimize the policies. The approach combines new parametric models of real-time scheduling, stochastic gradient descent, and infinitesimal perturbation analysis. In this formulation, perturbation analysis computes the gradient of the objective function with respect to the timing of messages and results in an efficient algorithm. The algorithm discovers the best time to send messages to optimize a combination of operating room efficiency and patient waiting times.