Robust ambulance base allocation strategy with social media and traffic congestion information

Abstract

At present, traffic congestion has become a major problem in many metropolitan areas globally, affecting the economy and social conditions in these areas. Of special concern, the response time of ambulances has increased, causing the death or disability of patients during emergencies. Various ambulance allocation strategies have been developed to locate a base that can achieve coverage of the demand point within a prescribed time or distance frame. However, real-time ambulance deployment is required to determine the number of ambulance and their bases. In particular, the dynamic relocation of an ambulance base is complicated, and each relocation does not guarantee that the next period will change again, thus increasing the workloads of the ambulance crew and potentially reducing their capability of responding to an emergency call. Furthermore, these models only considered covering all demand points but lacked the ability to consider uncertain factors, such as traffic congestion, patient conditions, public events, and population movement. Therefore, this study focused on formulating a covering model based on traffic congestion from web-based services and social media analysis, using the Markov-chain traffic speed assignment to allocate ambulance bases and trade-off the number of ambulance facilities between the current period and next period while considering the number of ambulance vehicles. Moreover, the proposed model was demonstrated using a case study of Bangkok emergency medical services. According to the results, obtained through collecting data on social media and traffic speed, the average traveling time of an ambulance can be improved by more than 70% with the trade-off between different periods if several emergency calls are received.