ORAHS2024
Abstract Submission

134. Dispatching strategy with lookahead for the real-time management of ambulances

Contributed abstract in session TD-5: Ambulance Management, stream Regular talks.

Tuesday, 16:00-17:30
Room: Room S6

Authors (first author is the speaker)

1. Alessandro Druetto
Dipartimento di Informatica, Università degli Studi di Torino
2. Roberto Aringhieri
Dipartimento di Informatica, Università degli Studi di Torino
3. Davide Duma
Dipartimento di Matematica "Felice Casorati", Università degli Studi di Pavia
4. Alberto Guastalla
Dipartimento di Informatica, Università degli Studi di Torino

Abstract

In the context of the Emergency Care Delivery System, several factors play a crucial role in the rate of morbidity in hospitals and mortality of patients. For example, the stochasticity of rescue requests can cause peaks in demand, increasing the arrival time within which ambulances are able to reach patients. These peaks can cause an uneven distribution of hospital workload, that negatively impacts the Emergency Department capability of processing all patients in time. On the other hand, a finely crafted and carefully studied management of Emergency Medical Service can make the difference to save more lives. The definition of static dispatching, routing and redeployment policies that are able to cope with those criticalities is sometimes not sufficient in order to guarantee a good utilization of ambulances while maximizing the number of emergency requests served. In the simulation and optimization model presented during ORAHS2019, the impact of predictive models in real-time management of ambulances was studied. While the former dispatching was based on static policies, this work provides a comprehensive dispatching strategy with lookahead based on online assignment models, which exploits pertinent information regarding ambulance status and, when accessible, oracle predictions for upcoming patient arrivals. Preliminary results of the new model show that such an online dispatch is able to significantly increase the number of rescue requests served within a time threshold.

Keywords

Status: accepted

Back to the list of papers