FATIGUE FACTOR ON MOTORCYCLISTS’ ACCIDENT; ANALYSIS USING BAYESIAN NETWORK

Lumba, Pada (2022) FATIGUE FACTOR ON MOTORCYCLISTS’ ACCIDENT; ANALYSIS USING BAYESIAN NETWORK. FATIGUE FACTOR ON MOTORCYCLISTS’ ACCIDENT; ANALYSIS USING BAYESIAN NETWORK, 29 (6). pp. 1-9. ISSN 0858849X, 25870009

Text FATIGUE FACTOR ON MOTORCYCLISTS’ ACCIDENT; ANALYSIS USING BAYESIAN NETWORK.pdf Download (2MB)

Abstract

This study aims to determine the possibility of accident for motorcyclists who take a break and those who do not take a break along their way due to fatigue. Data statistically from July to December 2015, around 70.93% of the crashes that occurred in Indonesia had involved motorcycles. The study area is located in Indonesia. The number of samples consist of 220 motorcyclists who had suffered accident. From 220 respondent, around 120 respondents were used to analyze data and 100 respondents were used to validate of model. Criteria of respondent are motorcyclists who had suffered accident. Data was collected by distributing questionnaire links on social media by asking something that are related to accidents experienced by respondents. Because the data that is obtained from survey in form probabilistic, thus the data is analyzed by the Bayesian Network Method. The results showed that the accident probability for motorcyclists who did not take a break on the way due to fatigue was 74%, while motorcyclists who took a break on the way due to fatigue had an accident probability of 26%. To obtain an accurate model, the model is validated by calculating the value of Mean Absolute Deviation (MAD). The results indicated that MAD value was 15.43%. This shows that the model has high accuracy, after that several scenarios are performed to determine the dominant variables that influence the risk of accidents in motorcyclists who take a break and those who do not take a break along their way.

Item Type:	Article
Subjects:	T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Fakultas Teknik > Teknik Sipil
Depositing User:	Dr Pada Lumba
Date Deposited:	03 Oct 2022 03:56
Last Modified:	03 Oct 2022 03:56
URI:	http://repository.upp.ac.id/id/eprint/1541

Actions (login required)

View Item