Afet sonrası operasyonları iyileştirmek için ambulans sevk etme politikalarının incelenmesi: Olası İstanbul depremi vaka analizi

In this study, different ambulance dispatching policies are tested by a simulation modelfor the post-disaster ambulance dispatching problem. After an expected earthquake inIstanbul, many people would be trapped under the collapsed buildings. While the rescueefforts continue, a massive number of patients would need to be transported from the affectedareas to the hospitals by ambulances. Consequently, ambulances would be overwhelmedby the transportation requests. Our objective is to assess the performance of differentambulance dispatching policies, by simulating the ambulance dispatching operations in sevendistricts of Istanbul and their neighborhoods. There are two performance criteria. The firstone is the overall average response time. Response time is divided into two. First one is theelapsed time from the rescue of the patient till the arrival of the patient to the emergency unitat a hospital (RTH), while the second one considers the time between rescue of the patientand leaving time of emergency department of the patient (RTT). The second criterion is theoverall average service level. It is also divided into two based on reaching to hospital (SL1),and transferring the patient out of the emergency department (SL2). The rescue timesof the patients, their first-treatment durations and travel times are represented as randomvariables. Real road data is used for setting the distances between neighborhoods andneighborhood isolation risks are also considered while establishing travel times. The testedambulance dispatching policies are first-called-first-served (FCFS), shortest-distance-first(SDF), and most-critical-patients-first (MCPF). SDF dispatching policy works by assigningthe ambulances to the nearest patients based on expected travel times. MCPF policyinitially sorts people based on one of the three injury types, and assigns the ambulance to thenearest patient if there is a tie from the injury types. In the base case, we run the simulation model assuming the availability of real-time information. The results show that SDF policyperforms the best in terms of average service time but has problems with fairness. Weanalyze several what-if scenarios to investigate the effects of periodic information update,decreased travel times and inclusion of hospital emergency treatment times in separatecases. Our results indicate that intensive communication between ambulance drivers andhospital emergency coordinators is necessary to improve performance. Also, there is aneed for additional ambulances to ensure effective transportation of victims to the hospitalsand it is shown that average service times are highly sensitive to the balance betweennumber of patients and number of ambulances. Moreover, opening temporary emergencyhospitals in some neighborhoods would serve as a beneficial strategy increasing fairnessamong neighborhoods.