Using Data Mining Techniques to Road Safety Improvement in Spanish Roads

Abstract

Crashes are events that involve the interaction of different components: road, driver, vehicle and environment. Nevertheless, road is an essential component and improvements on road conditions are directly related to increased traffic safety. From 2008 to 2010 a road safety inspection project was developed, whose aim was to identify and collect information about hazardous points on the Complementary Road Network of Andalusia, Spain, and build a database with this information. These elements were technically called Susceptible Elements of Improvement (ESM), which are defined as elements on the road that show worse road conditions than the ideal road safety standards. The main objective of this paper is to study the relationship between ESMs, number of crashes and hazardous sections, by analysing the information gathered in this database with advanced data mining techniques. Economically, this project is rather beneficial, since the resources of governments are limited, and therefore, it is necessary to intervene in those sections that have a higher cost-effectiveness ratio. Therefore, these relationships between roads conditions and crashes will be identified by analysing the information available in this data set of the Government of Andalusia, which has not been previously used.

Vessel traffic safety in busy waterways: A case study of accidents in western shenzhen port

“…With the rapid development of information and computer technology, big data theory and advanced techniques were also introduced to study the mechanism of safety and accident. Martín (Martín et al, 2014) applied data mining technique to study the relationships between road conditions and crash accidents to improve road safety. Verma (Verma et al, 2014) utilized association rule mining to identify safety patterns in steel plant.…”

Identifying Seriousness of Car Accidents using Data Mining in Jordanian Roads

“…When it comes to resource and economic costs, road accidents have resulted in a largely negatively impact; because of this, a number of studies have been constructed so as to examine this issue. One such example of this is that of Martin et al [5], who studied the data concerning dangerous points on Andalusia, Spain's Complementary Road Network, whereby a link was, indeed, founded, and the Andalusian roads improved. In a similar vein, in Ethiopia, Tesema et al [6] utilised an adaptive regression tree model in order to construct a decision support system for the road accidents on the Ethiopian roads; similarly, in Belgium, Geurts et al ( [7] utilised association rule mining technique to comprehend the many circumstances that come about within high-frequency accident locations.…”

Identifying Efficient Road Safety Prediction Model Using Data Mining Classifiers

“…Identified and collected information about hazardous point on the complementary road network of Andalusia and Spain and data mining techniques are applied [11] to determine road safety . Road accidents in India at National, State and Metropolitan city level has been analysed with its issues and challenges [12]. Historical traffic accident data of California, severity factor analysed with Naïve Bayes classifier and Decision Tree classifier [13].…”