Analysis of Rural Roads in Developing Countries Using Principal Component Analysis and Simple Average Technique in the Development of a Road Safety Performance Index

Authors: Muhammad Tufail, Jawad Hussain, Hammad Hussain, Imran Hafeez, Naveed Ahmad

Abstract:

Road safety performance index is a composite index which combines various indicators of road safety into single number. Development of a road safety performance index using appropriate safety performance indicators is essential to enhance road safety. However, a road safety performance index in developing countries has not been given as much priority as needed. The primary objective of this research is to develop a general Road Safety Performance Index (RSPI) for developing countries based on the facility as well as behavior of road user. The secondary objectives include finding the critical inputs in the RSPI and finding the better method of making the index. In this study, the RSPI is developed by selecting four main safety performance indicators i.e., protective system (seat belt, helmet etc.), road (road width, signalized intersections, number of lanes, speed limit), number of pedestrians, and number of vehicles. Data on these four safety performance indicators were collected using observation survey on a 20 km road section of the National Highway N-125 road Taxila, Pakistan. For the development of this composite index, two methods are used: a) Principal Component Analysis (PCA) and b) Equal Weighting (EW) method. PCA is used for extraction, weighting, and linear aggregation of indicators to obtain a single value. An individual index score was calculated for each road section by multiplication of weights and standardized values of each safety performance indicator. However, Simple Average technique was used for weighting and linear aggregation of indicators to develop a RSPI. The road sections are ranked according to RSPI scores using both methods. The two weighting methods are compared, and the PCA method is found to be much more reliable than the Simple Average Technique.

Keywords: Aggregation, index score, indicators, principal component analysis, weighting.

References:

[1] WHO, “Global status Report on Road Safety,” 2018.
[2] Pakistan Bureau of Statistics Islamabad, “Traffic Accidents Data of Pakistan,” 2018.
[3] M. Saisana, S. Tarantola, and A. Saltelli, “Uncertainty and sensitivity techniques as tools for the analysis and validation of composite indicators,” J. R. Stat. Soc., vol. 168, no. 2, pp. 307–323, 2005.
[4] E. Hermans, F. Van den Bossche, and G. Wets, “Combining road safety information in a performance index,” Accid. Anal. Prev., vol. 40, no. 4, pp. 1337–1344, 2008.
[5] M. Tešić, E. Hermans, K. Lipovac, and D. Pešić, “Identifying the most significant indicators of the total road safety performance index,” Accid. Anal. Prev., vol. 113, no. January, pp. 263–278, 2018.
[6] G. Leduc, “Road Traffic Data : Collection Methods and Applications,” EUR Number Tech. Note JRC 47967, vol. JRC 47967, no. May, p. 55, 2008.
[7] F. Wegman, D. Lynam., and G. Nilsson, SUNflower: a comparative study of the developments of road safety in Sweden, the United Kingdom, and the Netherlands, no. January 2002. 2002.
[8] G. Al Haji, Towards a Road Safety Development Index (RSDI) Development of an International Index to Measure Road Safety Performance, no. 1174. 2005.
[9] E. Hermans, F. Van den Bossche, and G. Wets, “Impact of methodological choices on road safety ranking,” 2007.
[10] E. Hermans, F. Van den Bossche, and G. Wets, “Uncertainty assessment of the road safety index,” Reliab. Eng. Syst. Saf., vol. 94, no. 7, pp. 1220–1228, 2009.
[11] F. Wegman and S. Oppe, “Benchmarking road safety performances of countries,” Saf. Sci., vol. 48, no. 9, pp. 1203–1211, 2010.
[12] V. Gitelman, E. Doveh, and S. Hakkert, “Designing a composite indicator for road safety,” Saf. Sci., vol. 48, no. 9, pp. 1212–1224, 2010.
[13] J. T. Bastos, Y. Shen, E. Hermans, T. Brijs, G. Wets, and A. C. P. Ferraz, “Traffic fatality indicators in Brazil: State diagnosis based on data envelopment analysis research,” Accid. Anal. Prev., vol. 81, pp. 61–73, 2015.
[14] T. Brijs, Y. Shen, G. Wets, E. Hermans, Q. Bao, and W. Wang, “Inter-national benchmarking of road safety: State of the art,” Transp. Res. Part C Emerg. Technol., vol. 50, pp. 37–50, 2014.
[15] I. S. M. R., H. Hamid, L. Teik Hwa, and A. Farhan, “Identification of Hazardous Road Sections: Crash Data versus Composite Index Method,” Int. J. Eng. Technol., vol. 6, no. 6, pp. 481–486, 2014.
[16] M. Nardo, M. Saisana, A. Saltelli, S. Tarantola, A. Hoffman, and E. Giovannini, Handbook on constructing composite indicators, no. 03. 2005.
[17] M. A. Hakkert, A.S, Gitelman, V. and Vis, “Institutional Repository Road Safety Performance Indicators : Theory. Deliverable D3. 6 of the EU FP6 project SafetyNet” p. 166, 2007.
[18] M. A. and V. G. Vis, “Road Safety Performance Indicators: Country Profiles. Deliverable D3.7b of the EU FP6 project SafetyNet.” 2007.
[19] G. Pakistan, “Big Data Analysis Series Paper 2,” 2000.
[20] World Health Organization, “Global Status Report on Road Safety : Motorcycle helmets: the facts,” p. 2015, 2015.
[21] Google Earth, “Google Earth 2020,” Google Earth, 2020. (Online). Available: https://www.google.com/maps/dir/33.8406144,72.9057462.
[22] P. Thomas and V. Safety, “Deliverable D3.8: Road Safety Performance Indicators Manual,” 2007.
[23] N. Michela, S. Michaela, S. Andrea, and T. Stefano, “Tools for Composite Indicators Building Prepared,” Eur. Communities, 2005.
[24] B. J. Babin and R. E. Anderson, on Multivariate Data Analysis Joseph F. Hair Jr. William. C. Black Seventh Edition. 2014.
[25] M. A. Akaateba, “Comparing Road Safety Performance of Selected Eu and African Countries Using a Composite Road,” vol. 2, no. 8, pp. 31–46, 2012.
[26] L. L. Chan and N. Idris, “Validity and Reliability of The Instrument Using Exploratory Factor Analysis and Cronbach’s alpha,” Int. J. Acad. Res. Bus. Soc. Sci., vol. 7, no. 10, pp. 400–410, 2017.