Document Details

Document Type : Thesis 
Document Title :
ASSESSING THE EFFECTIVENESS OF INTERACTIVE TRAFFIC ACCIDENT MAPS AS A PLANNING TOOL FOR DECISION MAKERS IN SAUDI ARABIA
تقييم فاعلية الخريطة التفاعلية للحوادث المرورية كأداة تخطيط لمتخذي القرار في المملكة العربية السعودية
 
Subject : Faculty of Computing and Information Technology 
Document Language : Arabic 
Abstract : The rapid increase of road traffic accident (RTA) risk causes significant concerns for decision-makers and researchers on traffic safety. The diversity, rarity, and interconnectivity of historical data on factors causing car accidents point to the need or more focused studies for analyzing, predicting, and visualizing the risk of accidents over the short and long term for preventive purposes. In the field of prediction, previous studies mentioned the long short-term memory (LSTM) as a flexible technique to deal with the time series problems to consider many factors by using a deep learning algorithm instead of usual linear regression techniques as autoregressive integrated moving average (ARIMA) and Vector Auto Regression (VAR). Some studies have explored how these methods can be used to construct prediction models and tools for different risk situations to support decision-makers. The present study investigates the possibility of using LSTM to enhance the analysis and predictability of RTAs risk for Saudi Arabia. Univariate and multivariate models were used in three cases to predict RTAs risk with comparison by ARIMA and VAR models. Based on the future forecasting models, the present study created an interactive map as a planning tool to visualize the actual and predicted values up to 2030 for all Saudi regions. The research displays a tutorial to explain the proposed framework process and discuss the results based on some evaluation metrics for forecasting. The performance of ARIMA models still progressing other models in forecasting short term time series data of RTAs risk. This is don’t mean LSTM has the worse ability among models, but it can work better with big data for a long time. Whereas VAR registered good accuracy for multivariate time series for the short term. The study also discusses how to invest the study limitations in future researches to support the road traffic safety field. 
Supervisor : Dr. Mahmoud Ibrahim Kamel Ali 
Thesis Type : Master Thesis 
Publishing Year : 1441 AH
2020 AD 
Added Date : Wednesday, July 8, 2020 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
مرام محمد الراجحيAlrajhi, Maram MohammedResearcherMaster 

Files

File NameTypeDescription
 46604.pdf pdf 

Back To Researches Page