ORAHS2024
Abstract Submission

266. A stochastic optimization approach for scheduling CT scans and reports

Contributed abstract in session TC-1: Poster session, stream Posters.

Tuesday, 14:00-15:30
Room: Auditorium

Authors (first author is the speaker)

1. Sara Cambiaghi
Matematica, Università di Pavia
2. Davide Duma
Dipartimento di Matematica "Felice Casorati", Università degli Studi di Pavia

Abstract

In a hospital setting, the organization of the CT scan service is highly complex due to the examination being divided into two tasks - scanning and report writing - which require coordination. We present the case study of Policlinico San Matteo in Pavia, Italy, where further complexity arises from the presence of three different patient categories - outpatients, inpatients, and emergency - all sharing the same resources. The main issue in this operational context lies in the predominance of emergency cases, comprising 73% of total patients. Consequently, strategic planning of outpatient and inpatient appointments is crucial to avoid overlaps with emergency cases. In this study, we focus on scheduling appointments for outpatients, proposing a multi-objective optimization problem. Firstly, we aim to obtain predictions of the duration of examinations and reporting for outpatients. To achieve this, we relied on data from examinations conducted between June 2021 and November 2022, employing various machine learning and deep learning techniques. Furthermore, we introduce a stochastic programming model aimed at minimizing direct waiting times for outpatients and completion time for emergency patients. A genetic algorithm is proposed to efficiently solve the problem within a reasonable timeframe. Finally, a quantitative analysis is conducted to evaluate the effectiveness of the proposed optimization and machine learning approach in enhancing the efficiency of the CT-scan service.

Keywords

Status: accepted

Back to the list of papers