Optimizing Traffic Light Timing Using Graph Theory: A Case Study at Urban Intersections DOI Creative Commons

Darmaji Darmaji,

Utama Khalid Lubis,

Riska Fitriani

и другие.

Interval Indonesian Journal of Mathematical Education, Год журнала: 2024, Номер 2(2), С. 149 - 163

Опубликована: Дек. 17, 2024

Purpose of the study: This study aims to optimize traffic light timing at Usman Salengke-Poros Malino-K.H. Wahid Hasyim intersection using a graph theory approach. By modeling compatible flows and calculating optimal signal durations, seeks reduce congestion, minimize delays, improve efficiency. Methodology: utilized manual volume data collection methods with direct field observations intersection. It employed Webster's method for cycle calculation MATLAB software simulation. Tools included measuring tapes (Stanley), stopwatches (Casio), sheets recording flow. Surveys captured vehicle types peak hour volumes. Main Findings: The duration was calculated as 95 seconds, reducing original time 128 seconds. Peak observed 1,383 pcu/hour (Usman Salengke North). green increased North 39 seconds Poros Malino 28 Total average waiting decreased by 33.3%, improved throughput 20%. Novelty/Originality this introduces practical application optimizing timing, flow simplify analysis. Unlike adaptive systems requiring expensive technology, approach relies on data, offering cost-effective solutions. advances existing knowledge providing simplified, scalable congestion enhancing efficiency in urban settings.

Язык: Английский

Research on Traffic Accident Severity Level Prediction Model Based on Improved Machine Learning DOI Creative Commons
J. K. K. Tang, Yao Zhi Huang, Dingli Liu

и другие.

Systems, Год журнала: 2025, Номер 13(1), С. 31 - 31

Опубликована: Янв. 4, 2025

Traffic accidents occur frequently, causing significant losses to people’s lives and property safety. Accurately predicting the severity level of traffic is great significance. Based on accident data, this study comprehensively considers various influencing factors such as geographical location, road conditions, environment. The data are divided into accident-related categories, weather-related road- environment-related categories. machine learning method improved through integration for prediction. In experiment, effective preprocessing measures were taken problems imbalance, missing values, encoding categorical variables, standardization numerical features. unbalanced distribution “Severity” was under-sampling over-sampling techniques. Firstly, we adopted a multi-stage fusion strategy. A multi-layer perceptron (MLP) used preliminary prediction, then its result combined with original features form new feature. Decision tree, XGBoost, random forest algorithms, respectively, applied secondary analysis results show that model significantly superior single in overall performance. “MLP + forest” performs well evaluation indicators accuracy, recall rate, F1 value. accuracy rate high 94%. prediction different levels (minor, moderate, severe), also generally shows better performance stability. research have broad prospects field intelligent driving. It can realize real-time early warnings, provide decision support drivers autonomous driving systems. provides scientific basis planning management departments improve conditions reduce probability accidents.

Язык: Английский

Процитировано

3
Traffic flow Modelling of Vehicles on a Six lane Freeway: Comparative Analysis of Improved Group method of Data Handling and Artificial Neural Network Model DOI Creative Commons
Isaac Oyeyemi Olayode, Alessandro Severino, Frimpong J. Alex

и другие.

Results in Engineering, Год журнала: 2025, Номер unknown, С. 104094 - 104094

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

2
AI-Driven Digital Transformation and Sustainable Logistics: Innovations in Global Supply Chain Management DOI Creative Commons

Ghazaleh Kermani Moghaddam,

Mostafa Karimzadeh

Research Square (Research Square), Год журнала: 2025, Номер unknown

Опубликована: Фев. 27, 2025

Abstract The global supply chain has progressed beyond conventional logistics, incorporating digital technology, sustainability, and automation. It involves interrelated processes that convert raw resources into finished goods. rising complexity from cross-border legislation, currency volatility, evolving market demands requires decision-making driven by AI, Big Data, This study does a Systematic Literature Review of 65 journal papers (2010–2024) to analyze developments in logistics via innovation, sustainability. In contrast models characterized static decision-making, emerging frameworks integrate AI-driven optimization, blockchain transparency, real-time data for predictive forecasting. Furthermore, autonomous freight transportation, encompassing self-driving trucks, drone-assisted last-mile delivery, hyperloop cargo systems, is transforming logistics. Findings underscore significant transformations strategy, focusing on sustainable mobility, carbon footprint mitigation, integrated analysis delineates research deficiencies proposes avenues future investigation systems management.

Язык: Английский

Процитировано

1
From Baseline to Best Practice: An Advanced Feature Selection, Feature Resampling and Grid Search Techniques to Improve Injury Severity Prediction DOI Creative Commons

Soukaina El Ferouali,

Zouhair Elamrani Abou Elassad, Sara Qassimi

и другие.

Applied Artificial Intelligence, Год журнала: 2025, Номер 39(1)

Опубликована: Янв. 23, 2025

Язык: Английский

Процитировано

0
GINSER: Geographic Information System Based Optimal Route Recommendation via Optimized Faster R-CNN DOI Creative Commons

S.D. Anitha Selvasofia,

B. Sivasankari,

R. Dinesh

и другие.

International Journal of Computational Intelligence Systems, Год журнала: 2025, Номер 18(1)

Опубликована: Апрель 7, 2025

Язык: Английский

Процитировано

0
A review on Control momentum Gyroscopic Stabilization for intelligent balance Assistance in Electric Two-wheeler DOI Creative Commons

Prithvi Raj Pedapati,

C. Ramesh Kumar

Results in Engineering, Год журнала: 2025, Номер unknown, С. 105069 - 105069

Опубликована: Апрель 1, 2025

Язык: Английский

Процитировано

0
Machine learning based adaptive traffic prediction and control using edge impulse platform DOI Creative Commons
Manoj Tolani,

G E Saathwik,

Ayush Roy

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Май 17, 2025

Traffic congestion and delays are two major challenges in modern vehicle traffic control systems. These issues can be mitigated through an efficient autonomous scheduling system. The objective of the proposed methodology is to automate system based on density vehicles approaching signal without any human intervention. Unlike conventional systems that rely preset timers which often unsuitable for unpredictable conditions. Therefore, approach dynamically adjusts timings real-time data. utilizes proximity sensors strategically placed at a predetermined distance from detect vehicles. speed monitored readings these sensors. A Edge-Impulse-based machine learning model predict arrival time signal. Using algorithms, forecast future conditions optimize by significantly reducing delays. Moreover, automating process, help reduce error improve safety road users. has potential transform existing systems, making them more intelligent, efficient, autonomous. rigorously tested validated ensure its reliability accuracy real-world scenarios.

Язык: Английский

Процитировано

0
Application of Machine Learning for Road Safety Modeling of Selected South-West Highway in Nigeria DOI

O. Bayode,

Olumuyiwa Samson Aderinola,

Bamitale Dorcas Oluyemi-Ayibiowu

и другие.

Deleted Journal, Год журнала: 2025, Номер 3(3), С. 202 - 213

Опубликована: Май 22, 2025

Road crash prediction has proven to be an effective means of improving highway safety. In recent years, machine learning (ML) models have been embraced as efficient for the road accident frequency. This study applied two occurrence on selected South-West in Nigeria. Accident data were obtained period 10 years from 2013 2022 Federal Safety Commission (FRSC) Nigeria under and traffic operations determined site using manual counting stopwatch approach. Machine Learning including Support Vector (SVM) Extreme Gradient Boosting (XGBoost) also used statistical model safety with consideration identified contributing factors. The performance was compared both training testing dataset coefficient determination (R2), Mean Absolute Error (MAE) Root Square (RMSE). showed consistency ML R2 0.99 SVM, 0.97 XGBoost data, 0.93 SVM 0.76 data. are easy fast implement. result this supports use a predictive tool evaluation. knowledge attained will benefit transportation planners, engineers, policymakers implement measures aimed at reducing occurrence, thereby enhancing overall efficiency.

Язык: Английский

Процитировано

0
Computational Methods for Automatic Traffic Signs Recognition in Autonomous Driving on Road: A Systematic Review DOI Creative Commons
Hui Chen, Mohammed A. H. Ali, Y. Nukman

и другие.

Results in Engineering, Год журнала: 2024, Номер 24, С. 103553 - 103553

Опубликована: Дек. 1, 2024

Язык: Английский

Процитировано

1
Optimizing Traffic Light Timing Using Graph Theory: A Case Study at Urban Intersections DOI Creative Commons

Darmaji Darmaji,

Utama Khalid Lubis,

Riska Fitriani

и другие.

Interval Indonesian Journal of Mathematical Education, Год журнала: 2024, Номер 2(2), С. 149 - 163

Опубликована: Дек. 17, 2024

Purpose of the study: This study aims to optimize traffic light timing at Usman Salengke-Poros Malino-K.H. Wahid Hasyim intersection using a graph theory approach. By modeling compatible flows and calculating optimal signal durations, seeks reduce congestion, minimize delays, improve efficiency. Methodology: utilized manual volume data collection methods with direct field observations intersection. It employed Webster's method for cycle calculation MATLAB software simulation. Tools included measuring tapes (Stanley), stopwatches (Casio), sheets recording flow. Surveys captured vehicle types peak hour volumes. Main Findings: The duration was calculated as 95 seconds, reducing original time 128 seconds. Peak observed 1,383 pcu/hour (Usman Salengke North). green increased North 39 seconds Poros Malino 28 Total average waiting decreased by 33.3%, improved throughput 20%. Novelty/Originality this introduces practical application optimizing timing, flow simplify analysis. Unlike adaptive systems requiring expensive technology, approach relies on data, offering cost-effective solutions. advances existing knowledge providing simplified, scalable congestion enhancing efficiency in urban settings.

Язык: Английский

Процитировано

0