TY - JOUR
T1 - Predicting serious injuries due to road traffic accidents in Malaysia by means of artificial neural network
AU - Radzuan, Nurul Qastalani
AU - Hassan, Mohd Hasnun Arif
AU - Abdul Majeed, Anwar P.P.
AU - Musa, Rabiu Muazu
AU - Mohd Razman, Mohd Azraai
AU - Abu Kassim, Khairil Anwar
N1 - Funding Information:
Acknowledgements. The authors would like to acknowledge ASEAN NCAP, FIA Foundation, Global NCAP, OEMs, Malaysian Institute of Road Safety Research (MIROS) and the Society of Automotive Engineers Malaysia (SAE Malaysia) for funding this study under the ASEAN NCAP Holistic Collaborative Research (ANCHOR II) grant. Also, the authors are thankful to the Universiti Malaysia Pahang for providing the facilities to conduct the study.
Funding Information:
The authors would like to acknowledge ASEAN NCAP, FIA Foundation, Global NCAP, OEMs, Malaysian Institute of Road Safety Research (MIROS) and the Society of Automotive Engineers Malaysia (SAE Malaysia) for funding this study under the ASEAN NCAP Holistic Collaborative Research (ANCHOR II) grant. Also, the authors are thankful to the Universiti Malaysia Pahang for providing the facilities to conduct the study.
Publisher Copyright:
© Springer Nature Singapore Pte Ltd. 2020.
PY - 2020
Y1 - 2020
N2 - Malaysia has recorded a steady increase in the number of road traffic accidents from year to year at an alarming rate of 5%. Serious injuries due to the accidents, which could lead to permanent disability, might cause a long-term problem to the nation economy-wise. Predicting the number of serious injury cases in the future is important in understanding the trend of road traffic accidents to help policymakers in proposing a countermeasure. Time-series model has been employed to predict the occurrence of road traffic crashes including fatalities. Nonetheless, the prediction of serious injury cases, which should not be taken lightly due to its potential impact, has not been proposed especially with regards to Malaysian road traffic accident data. This study attempts to employ artificial neural networks (ANN), a machine learning algorithm, to predict the number of serious injury cases in Malaysia based on the road traffic accident data of the past 20 years. Machine learning has increasingly been adopted in recent years owing to its ability to predict as well as catering for the non-linear behaviour of the data examined. A single-hidden ANN model was developed based on seven features, namely the number of registered vehicles, population, length of federal road, length of FELDA road, length of federal institutional road, length of federal territory road, and length of the expressway in order to predict the number of serious injuries. It was established from the present investigation that the developed ANN model is capable to predict the number of serious injuries from 1997 until 2017 with a mean absolute percentage error of only 3%. This demonstrates the capability of the developed machine learning in road traffic accident prediction, and it could be useful in outlining an action plan to mitigate the number of serious injuries in Malaysia.
AB - Malaysia has recorded a steady increase in the number of road traffic accidents from year to year at an alarming rate of 5%. Serious injuries due to the accidents, which could lead to permanent disability, might cause a long-term problem to the nation economy-wise. Predicting the number of serious injury cases in the future is important in understanding the trend of road traffic accidents to help policymakers in proposing a countermeasure. Time-series model has been employed to predict the occurrence of road traffic crashes including fatalities. Nonetheless, the prediction of serious injury cases, which should not be taken lightly due to its potential impact, has not been proposed especially with regards to Malaysian road traffic accident data. This study attempts to employ artificial neural networks (ANN), a machine learning algorithm, to predict the number of serious injury cases in Malaysia based on the road traffic accident data of the past 20 years. Machine learning has increasingly been adopted in recent years owing to its ability to predict as well as catering for the non-linear behaviour of the data examined. A single-hidden ANN model was developed based on seven features, namely the number of registered vehicles, population, length of federal road, length of FELDA road, length of federal institutional road, length of federal territory road, and length of the expressway in order to predict the number of serious injuries. It was established from the present investigation that the developed ANN model is capable to predict the number of serious injuries from 1997 until 2017 with a mean absolute percentage error of only 3%. This demonstrates the capability of the developed machine learning in road traffic accident prediction, and it could be useful in outlining an action plan to mitigate the number of serious injuries in Malaysia.
KW - Artificial neural network
KW - Machine learning
KW - Prediction
KW - Road traffic accident
KW - Serious injuries
UR - http://www.scopus.com/inward/record.url?scp=85071289301&partnerID=8YFLogxK
U2 - 10.1007/978-981-13-9539-0_8
DO - 10.1007/978-981-13-9539-0_8
M3 - Conference article
AN - SCOPUS:85071289301
SN - 2195-4356
SP - 75
EP - 80
JO - Lecture Notes in Mechanical Engineering
JF - Lecture Notes in Mechanical Engineering
T2 - 2nd Symposium on Intelligent Manufacturing and Mechatronics, SympoSIMM 2019
Y2 - 8 July 2019 through 8 July 2019
ER -
