A PLS/SEM Approach Risk Factor Analysis in Road Accidents Caused by Carelessness
Mohammad Reza Elyasi1, Mahmoud Saffarzade2, Amin Mirza Boroujerdian3

1Mohammad Reza Elyasi, Roads and Transportation Department, Tarbiat Modares University, Tehran, Iran,.
2Mahmoud Saffarzade, Transportation Planning Department, Tarbiat Modares University, Tehran, Iran.
3Amin Mirza Boroujerdian, Roads and Transportation Department, Tarbiat Modares University, Tehran, Iran.

Manuscript received on 13 August 2016 | Revised Manuscript received on 20 August 2016 | Manuscript Published on 30 August 2016 | PP: 85-93 | Volume-5 Issue-6, August 2016 | Retrieval Number: F4689085616/16©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Many developed countries in line with the increase in road transport, and consequently an increase in the rate of accidents, are searching for effective ways to reduce road accidents. In the area of traffic safety, in order to identify factors contributing to accidents, conventional methods which generally based on regression analysis are used. However, these methods only detect accidents in different roads, but cannot clearly identify the cause of accidents and define the relationship between them. In addition, the methods used have two major limitations: 1- Postulate the structure of the model, and, 2- Observability of all variables. Due to the limitations discussed and also due to the complex nature of human factors, and the impact of road conditions, vehicle and environment on human factors, the aim of this study is to provide a useful tool for defining and measuring road, traffic and human factors, to evaluate the effect of each of them in accidents which caused by carelessness, directly and indirectly by using structural equation modeling with the partial least squares approach. Compared with the regression-based techniques or methods of pattern recognition that only a layer of relationships between independent and dependent variables is determined, the SEM approach provides the possibility of modeling the relationships between multiple independent and dependent structures. Moreover, the ability to use unobservable hidden variables, by using observable variables would be possible.
Keywords: Human Factors, Road Safety, Road Factors; Accident Analysis; Partial Least Square (PLS); Structural Equation Modeling (SEM).

Scope of the Article: Risk Management