Development of District wise Crash Prediction Model in Bangladesh

Abstract

Bangladesh is an accident prone country. Every year a considerable number of road crashes occur in this country and these are categorized as fatal injury crash, grievous injury crash, simple injury crash and collision or no injury or property damage crash. Different independent factors are directly and indirectly associated with these crash occurrences. Thus, a crash prediction model is necessary to explore the adverse effects of various structural and non-structural elements that are linked with transportation systems on individual crashes in context of Bangladesh. Considering all districts of Bangladesh about 13 types of road infrastructures, all types of roads and highways, socio economic and demographic factors and weather conditions (rainfall, fog and temperature) have been taken as predictors to identify their significance on different crash occurrences. Negative binomial regression models have been developed for these 4 types of crashes with the available data. From our study we have found that any types of concrete made infrastructure (RCC Bridge, Box culvert, RCC girder and PC girder) are vulnerable to the fatal and grievous injury crashes in the national, regional and zilla (district) roads. Highway infrastructures designed with composite materials i.e. Steel beam and RCC slab (SBRS), Truss with RCC slab (TRS), Baily with steel deck (BSD) are found less vulnerable to any types of the crash. Specially, SBRS and TRS are found safe for national highways and zilla roads. National highways are found risk prone area for fatal and simple injury. Roads that are not surveyed yet have been found hazardous for grievous injury crash. On the contrary, zilla roads and village roads are found less vulnerable to fatal crash relative to national highway. Most of the roads in upazilla and union are of pavement type rather than earthen type. Flexible and rigid pavements are found increasing the fatal injury whereas brick pavements are found both positive and negative effect on grievous injury. Also, paved roads are found vulnerable to simple crash in the union roads in Bangladesh. Moreover, variations in the number of road structures and number of existing gap influence positively and negatively on different crashes. vi In socio economic and demographic characteristics, number of female decreases the fatal crash where as sex ratio (M/F) tends to increase it. Also, the more the household size, the more the probability of fatal crash. After considering weather conditions in Bangladesh, we have found that rainfall likely to decrease fatal crash and collision during the year of 2009-2013. Average minimum temperature is found negatively significant in 2009 and 2010 on grievous injury crash and collision but after that till 2013 it explored with positively significant. No significant effect of temperature is found on fatal and simple crash.