Investigating trip generation behavior during Medical Emergency in Bangladesh

Abstract

Bangladesh being a densely populated country, the usage of private and public transportation is getting higher, thus increasing the rate of accidents day by day. The spike of road accidents creates an impact on road condition, vulnerability of different modes used by mass people and the chances of emergency dispatch im-mediately after the occurrence of emergency situations. Despite being the fastest growing country, Bangladesh is still lagging when it comes to ensuring road safety, emergency dispatch, infrastructure, and adequate treatment. Research done before shows the lack of infrastructures as well as medical facilities, problems regarding trip generation and modal choice behavior when it comes to dispatching any victim by their family members and bystanders. Difficulties at the time of choosing proper mode impacts on the travel time to reach the destination,thus making patient more vulnerable before reaching any medical facility. Traffic congestion or narrow lanes creates a barrier to reach desired mode to reach the accident spot timely. In previous research, the uniqueness due to socio-economic aspects were not considered at all casesfor a developing country like Bangladesh.The following research establishes a pathway with the socio-economic variables and creates dependency with the trip generation factors. To determine the components, a causal link has been created between factors that indicate both demographic and risk perception that creates an impact on trip generation in urban and rural cities in Bangladesh. The Bayesian Belief Networkwas used to create the network, which itself was supported by previous research findings and technical expertise. The conditional independence-based PC technique has been used to create the Bayesian network. The data set was gathered from metropolitan like Dhaka, Rajshahi, Barishal and nearby districts. Usage 2 of the expectation-maximization technique and sensitivity analysis, the Bayesian network was utilized to calculate the posterior probability considering all variables respectively to determine factors affecting trip generation. Result shows that the type of mode selected, time of waiting, type of injury, in- come or educational background of both bystander and patients impacts on thetrip generation of any emergency dispatch. The result from this research can be useful when it comesto decision making of a patient or bystander, companies working to ensure quick dispatch, policy makers, law enforcement agencies and medical service providers.