Yazar "Dogan, Erdem" seçeneğine göre listele
Listeleniyor 1 - 16 / 16
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Analysis And Comparison Of Long Short-Term Memory Networks Short-Term Traffic Prediction Performance(FAC TRANSPORT SILESIAN UNIV TECHNOLOGY, 2020) Dogan, ErdemLong short-term memory networks (LSTM) produces promising results in the prediction of traffic flows. However, LSTM needs large numbers of data to produce satisfactory results. Therefore, the effect of LSTM training set size on performance and optimum training set size for short-term traffic flow prediction problems were investigated in this study. To achieve this, the numbers of data in the training set was set between 480 and 2800, and the prediction performance of the LSTMs trained using these adjusted training sets was measured. In addition, LSTM prediction results were compared with nonlinear autoregressive neural networks (NAR) trained using the same training sets. Consequently, it was seen that the increase in LSTM's training cluster size increased performance to a certain point. However, after this point, the performance decreased. Three main results emerged in this study: First, the optimum training set size for LSTM significantly improves the prediction performance of the model. Second, LSTM makes short-term traffic forecasting better than NAR. Third, LSTM predictions fluctuate less than the NAR model following instant traffic flow changes.Öğe Analysis of the relationship between LSTM network traffic flow prediction performance and statistical characteristics of standard and nonstandard data(WILEY, 2020) Dogan, ErdemThe effectiveness of road traffic control systems can be increased with the help of a model that can accurately predict short-term traffic flow. Therefore, the performance of the preferred approach to develop a prediction model should be evaluated with data sets with different statistical characteristics. Thus a correlation can be established between the statistical properties of the data set and the model performance. The determination of this relationship will assist experts in choosing the appropriate approach to develop a high-performance short-term traffic flow forecasting model. The main purpose of this study is to reveal the relationship between the long short-term memory network (LSTM) approach's short-term traffic flow prediction performance and the statistical properties of the data set used to develop the LSTM model. In order to reveal these relationships, two different traffic prediction models with LSTM and nonlinear autoregressive (NAR) approaches were created using different data sets, and statistical analyses were performed. In addition, these analyses were repeated for nonstandardized traffic data indicating unusual fluctuations in traffic flow. As a result of the analyses, LSTM and NAR model performances were found to be highly correlated with the kurtosis and skewness changes of the data sets used to train and test these models. On the other hand, it was found that the difference of mean and skewness values of training and test sets had a significant effect on model performance in the prediction of nonstandard traffic flow samples.Öğe Application of Smeed and Andreassen accident models for Turkey: Various scenario analyses(Gazi Univ, Fac Engineering Architecture, 2008) Akgungor, Ali Payidar; Dogan, ErdemIn this study, accident prediction models for Turkey were developed by using the historical data, between 1986 and 2005, including population (P), the number of vehicles (N), accidents (C), injuries (I), and fatalities (D). In the models development, the structural form of Smeed and Andreassen Models were employed. However, Smeed Model was developed in different point of view so that this model was called as Smeed Similarity Model. The number of accident, injury and fatalities in Turkey which is on the way of a full member of European Union, were tried to estimate under different three scenarios until the year of 2010. Both models were compared in terms of percent difference (PD), mean absolute percent errors (MAPE), and root mean square errors (RMSE). Despite that Andreassen model for the years between 1986 and 2005 had errors lower than Smeed Similarity model, for future estimates latter gave more plausible results with scenarios.Öğe Comparison Of Different Approaches In Traffic Forecasting Models For The D-200 Highway In Turkey(Fac Transport Silesian Univ Technology, 2018) Dogan, Erdem; Korkmaz, Ersin; Akgungor, Ali PayidarShort-term traffic estimations have a significant influence in terms of effectively controlling vehicle traffic. In this study, short-term traffic forecasting models have been developed based on different approaches. Seasonal autoregressive integrated moving average (SARIMA), artificial bee colony (ABC) and differential evolution (DE) algorithms are the techniques used in the optimization of models, which have been developed by using observation data for the D-200 highway in Turkey. 80% of the data were used for training, with the remaining data used for testing. The performances of the models were illustrated with mean absolute errors (MAEs), mean absolute percentage errors (MAPEs), the coefficient of determination (R2) and the root-mean-square errors (RMSEs). It is understood that all the models provided consistent and useful results when the developed models were compared with the statistical results. In the models created separately for two lanes, the R2 values of the models were calculated to be approximately 92% for the right lane, which is generally used by heavy vehicles, and 88% for the left lane, which is used by less traffic. Based on the MAE and RMSE values, the model developed by the ABC algorithm gave the lowest error and showed more effective performance than the other approaches. Thus, the ABC model showed that it is appropriate for use on other highways in Turkey.Öğe Energy Demand Estimation in Turkey According to Road and Rail Transportation: Walrus Optimizer and White Shark Optimizer Algorithm-Based Model Development and Application(Mdpi, 2024) Korkmaz, Ersin; Dogan, Erdem; Akgungor, Ali PayidarTransport energy demand (TED) forecasting is a crucial issue for countries like Turkey that are dependent on external resources. The accuracy and effectiveness of these forecasts are extremely important, especially for the strategies and plans to be developed. With this in mind, different forms of forecasting models were developed in the present study using the Walrus Optimizer (WO) and White Shark Optimizer (WSO) algorithms to estimate Turkey's energy consumption related to road and railway transportation modes. Additionally, another objective of this study was to examine the impacts of different transport modes on energy demand. To investigate the effect of demand distribution among transport modes on energy consumption, model parameters such as passenger-kilometers (P-km), freight-kilometers (F-km), carbon dioxide emissions (CO2), gross domestic product (GDP), and population (POP) were utilized in the development of the models. It was found that the WO algorithm outperformed the WSO algorithm and was the most suitable method for energy demand forecasting. All the developed models demonstrated a better performance level than those reported in previous studies, with the best performance achieved by the semi-quadratic model developed with the WO, showing a 0.95% MAPE value. Projections for energy demand up to the year 2035 were established based on two different scenarios: the current demand distribution among transport modes, and a demand shift from road to rail transportation. It is anticipated that the proposed energy demand models will serve as an important guide for effective planning and strategy development. Moreover, the findings suggest that a balanced distribution among transport modes will have a positive impact on transport energy and will result in lower energy requirements.Öğe Estimating The Number Of Traffic Accidents, Injuries And Fatalities In Turkey Using Adaptive Neuro-Fuzzy Inference System(Scientific Research Center Ltd Belgrade, 2016) Akgungor, Ali Payidar; Korkmaz, Ersin; Dogan, ErdemThis study proposes Adaptive Neuro-Fuzzy Inference System (ANFIS) models to estimate the number of accidents, injuries and fatalities in Turkey. In the model development, population (P) and the number of vehicles (N) are selected as model parameters. Three different ANFIS structure models are developed using the data covering from 2000 to 2014. Developed models' results are statistically compared to observed values for training and test data in terms of root mean square errors (RMSE), mean absolute percentage errors (MAPE) and coefficient of determination (R-2). The results of the ANFIS models showed that they was suitable to estimate the number of accidents, injuries and fatalities. To investigate the performance of ANFIS models for future estimations, a ten-year period from 2015 to 2024 is considered. Thus, future values of population was obtained from the projection of Turkish Statistical Institute (TSI) and the vehicle ownership rate is expected to reach 0.4 by 2024. Therefore, population and the number of vehicles are considered to reach approximately 85 and 34 million, respectively. The results obtained from future estimations reveal the suitability of ANFIS approach for road safety applications.Öğe Estimation Of Car Ownership In Turkey Using Artificial Bee Colony Algorithm(Scientific Research Center Ltd Belgrade, 2016) Korkmaz, Ersin; Dogan, Erdem; Akgungor, Ali PayidarThis study proposes Artificial Bee Colony (ABC) models to estimate the number of cars in Turkey. In other words, car ownership is defined the number of cars per 1000 people. In the models development, population (P), per capita Gross Domestic Product (GDP) as dollars and fuel prices as Petrol, Diesel and Lpg were selected as model parameters. Two different ABC models were developed using the data covering from 2004 to 2015. According to fuel type, the coefficients of models were determined for each fuel type. Therefore, the sum of number of cars for each fuel type presented car ownership in Turkey. Developed models' results were statistically compared to observed values in terms of root mean square errors (RMSE), mean absolute percentage errors (MAPE) and coefficient of determination (R-2). The results of the ABC models showed that they were suitable to estimate the number of cars. To investigate the performance of ABC models for future estimations, a ten-year period from 2016 to 2025 was considered. Thus, future values of population were obtained from the projection of Turkish Statistical Institute (TSI) and the projections of other parameters, per capita GDP and fuel price, were executed according to current growth curve. The results obtained from future estimations reveal the suitability of ABC approach for determination of car ownership.Öğe Estimation of delay and vehicle stops at signalized intersections using artificial neural network(Univ Rijeka, Fac Engineering, 2016) Dogan, Erdem; Akgungor, Ali Payidar; Arslan, TuranDelay and number of vehicle stops are important indicators that define the level of service of a signalized intersection. Therefore, they are usually considered for optimizing the traffic signal timing. In this study, ANNs are employed to model delay and the number of stops estimation at signalized intersections. Intersection approach volumes, cycle length and left turn lane existence were utilized as input variables since they could easily be obtained from field surveys. On the other hand, the average delay and the number of stops per vehicle were used as the output variables for the ANNs models. Four-leg intersections were examined in this study. Approach volumes including turning volumes are randomly generated for each lane of these intersections, then the traffic simulation program was run 196 times with each generated data. Finally, average delay and the number of stops per vehicle were obtained from the simulations as outputs. In this study, various network architectures were analyzed to get the best architecture that provides the best performance. The results show that the ANNs model has potential to estimate delays and number of vehicle stops.Öğe EXAMINING THE SAFETY IMPACTS OF TRANSIT PRIORITY SIGNAL SYSTEMS USING SIMULATION TECHNIQUES(Fac Transport Silesian Univ Technology, 2024) Dogan, Erdem. Transit Priority Signal (TPS) systems are increasingly used to improve traffic efficiency and reduce passenger waiting times. However, such systems may carry potential safety risks. This study aims to investigate the safety effects of TPS at intersections. Our study utilized the SUMO traffic simulation program to create a road network model containing nine signalized intersections. Subsequently, the TPS system was applied to selected bus routes within the road network, and the cases with and without TPS implementation were compared in terms of safety and performance. In safety-oriented comparisons, surrogate safety measures were employed, including number of conflict and Time to Collision (TTC). Signalized intersection performances were measured and compared in terms of the number and duration of stops. The analysis results indicate that TPS enhances safety and transportation performance for buses, but adversely impacts safety and transportation performance for passenger cars. This study underscores the importance of considering safety aspects in the implementation of TPS aimed at improving passenger transportation efficiency. These findings may contribute to the enhancement of public transportation infrastructure and the implementation of appropriate safety measures.Öğe LSTM training set analysis and clustering model development for short-term traffic flow prediction(Springer London Ltd, 2021) Dogan, ErdemLong short-term memory (LSTM) is becoming increasingly popular in the short-term flow. In order to develop high-quality prediction models, it is worth investigating the LSTM potential deeply for traffic flow prediction. This study has two objectives: first, to observe the effect of using different sized training sets in LSTM training for various and numerous databases; second, to develop a clustering model that contributes to adjusting the training set size. For this purpose, 83 datasets were divided into certain sizes and LSTM model performances were examined depending on these training set sizes. As a result, enlargement of the training set size reduced LSTM errors monotonic for certain datasets. This phenomenon was modeled with the state-of-the-art clustering algorithms, such as K-nearest neighbor, support vector machine (SVM), logistic regression and pattern recognition networks (PRNet). In these models, statistical properties of datasets were utilized as input. The best results were obtained by PRNet, and SVM model performance was closest to PRNet. This study indicates that enlarging the training set size in traffic flow prediction increases the LSTM performance monotonically for specific datasets. In addition, a high-precision clustering model is presented to assist researchers in short-term traffic forecasting to adjust the size of the training set.Öğe Modelling Weekend Traffic With Weather Conditions Using Various Equation Type And Differential Evolution Algorithm(Scientific Research Center Ltd Belgrade, 2016) Dogan, Erdem; Akgungor, Ali Payidar; Korkmaz, ErsinIn weekends, amount of passenger car traffic is usually higher than weekday because of the activity-based traveling on some highways. Forecasting of this traffic, might help to local authorities to take safety precautions decisions on a road segments. This study aims to compose models to forecast weekend traffics using weather conditions and average weekday traffic variables. For this aim, two main models were composed: The Saturday traffic model and the Sunday traffic model. The Saturday traffic model variables are mean weekday daily traffic, maximum temperatures of Saturday and precipitations. The Sunday model is a linear model with only one variable: the predicted traffic values from the Saturday traffic model. In the modeling Saturday traffics, six-month (from January to June) data, which belongs to year 2015 and Ankara Kinkkale highway in Turkey, were used and 2014-March data were used for testing the models. The used temperatures were normalized and the precipitations data were involved as logical (0 or 1) inputs in models. To find best equation type for Saturday traffic model, four various equation forms were selected: (1) Linear, (2) polynomial-1, (3) polynomial-2, (4) multiplicative equation from. The linear and polynomial-1 have three, multiplicative has four, and polynomial -2 equation has five coefficients need to be determinate. Differential evolution algorithm was utilized to determinate best fitted values for these coefficients. Performance of the models were calculated using mean square error and coefficient of determination. The model with the polynomial-2 equation has minimum errors for the modelling stage and R-2 value is around 0.80. The model with the polynomial-2 showed the best performance on testing stage (R-2=0.96). These results show that the weekend traffic is related to weather conditions and it can be modeled convenient equation form and differential evolution algorithm.Öğe Optimizing a fuzzy logic traffic signal controller via the differential evolution algorithm under different traffic scenarios(Sage Publications Ltd, 2016) Dogan, Erdem; Akgungor, Ali P.This study aims at optimizing fuzzy logic controller (FLC) triangle membership functions (MFs) for different traffic volumes via differential evolution (DE). To achieve this goal, a new FLC with a red time limiter, which actually calculates green time and the extension time of traffic movement phase, is developed to control an intersection. Subsequently, this FLC is optimized with two levels, namely Level-1 and Level-2. Level-1 searches each fuzzy class's minimum and maximum values ( and ) that generate the lowest average delay per vehicle with DE. Using DE Level-2 inherits Level-1 ranges and reshapes the MFs to explore lower delay values computed by Level-1. The proposed method is tested with nine different traffic scenarios. For each scenario, 15 different headways are applied for a four-leg isolated intersection. The results indicate that the intersection average performance is increased up to 52%, 48%, and 14% at 800, 1600, and 2400 veh/h total intersection volumes, respectively, after Level-1 optimizations. They also reveal that intersection control produces higher delay values in only four scenarios after Level-2 procedures. Consequently, it is shown that the DE has significant potential to optimize FLCs at the intersection signal control. In addition, tuning fuzzy class ranges is found to be more critical than the MF reshaping process in traffic control via FLCs.Öğe PERFORMANCE ANALYSIS OF LSTM MODEL WITH MULTI-STEP AHEAD STRATEGIES FOR A SHORT-TERM TRAFFIC FLOW PREDICTION(Fac Transport Silesian Univ Technology, 2021) Dogan, ErdemIn this study, the effect of direct and recursive multi-step forecasting strategies on the short-term traffic flow forecast performance of the Long Short-Term Memory (LSTM) model is investigated. To increase the reliability of the results, analyses are carried out with various traffic flow data sets. In addition, databases are clustered using the k-means++ algorithm to reduce the number of experiments. Analyses are performed for different time periods. Thus, the contribution of strategies to LSTM was examined in detail. The results of the recursive based strategy performances are not satisfactory. However, different versions of the direct strategy performed better at different time periods. This research makes an important contribution to clarifying the compatibility of LSTM and forecasting strategies. Thus, more efficient traffic flow prediction models will be developed and systems such as Intelligent Transportation System (ITS) will work more efficiently. A practical implication for researchers that forecasting strategies should be selected based on time periods.Öğe Robust-LSTM: a novel approach to short-traffic flow prediction based on signal decomposition(Springer, 2022) Dogan, ErdemIntelligent transport systems need accurate short-term traffic flow forecasts. However, developing a robust short-term traffic flow forecasting approach is a challenging task due to the stochastic character of traffic flow. This study proposes a novel approach for short-term traffic flow prediction task, namely Robust Long Short Term Memory (R-LSTM) based on Robust Empirical Mode Decomposing (REDM) algorithm and Long Short Term Memory (LSTM). Short-term traffic flow data provided from the Caltrans Performance Measurement System (PeMS) database were used in the training and testing of the model. The dataset was composed of traffic data collected by 25 traffic detectors on different freeways' main lanes. The time resolution of the dataset was set to 15 min, and the Hampel preprocessing algorithm was applied for outlier elimination. The R-LSTM predictions were compared with the state-of-the-art models, utilizing RMSE, MSE, and MAPE as performance criteria. Performance analyses for various periods show that R-LSTM is remarkably successful in all time periods. Moreover, developed model performance is significantly higher, especially during midday periods when traffic flow fluctuations are high. These results show that R-LSTM is a strong candidate for short-term traffic flow prediction, and can easily adapt to fluctuations in traffic flow. In addition, robust models for short-term predictions can be developed by applying the signal separation method to traffic flow data.Öğe Short-Term Traffic Flow Prediction Using Artificial Intelligence With Periodic Clustering And Elected Set(SVENCILISTE U ZAGREBU, FAKULTET PROMETNIH ZNANOSTI, 2020) Dogan, ErdemForecasting short-term traffic flow using historical data is a difficult goal to achieve due to the randomness of the event. Due to the lack of a solid approach to short-term traffic prediction, the researchers are still working on novel approaches. This study aims to develop an algorithm that dynamically updates the training set of models in order to make more accurate predictions. For this purpose, an algorithm called Periodic Oustering and Prediction (PCP) has been developed for use in short-term traffic forecasting. In this study, PCP was used to improve Artificial Neural Networks (ANN) predictive performance by improving the training set of ANN to predict short-term traffic flow using selected clusters. A large amount of traffic data collected from the US and UK motorways was used to determine the PCP ability to increase the ANN performance. The robustness of the proposed approach was determined by the performance measures used in the literature and the mean prediction errors of PCP were significantly below other approaches. In addition, the studies showed that the percentage errors of PCP predictions decreased in response to increasing traffic flow values. Considering the obtained positive results, this method can be used in real-time traffic control systems and in different areas needed.Öğe SIMULTANEOUS MEASUREMENT AND ANALYSIS OF NOISE LEVELS IN FLEXIBLE AND RIGID PAVEMENTS(Pamukkale Univ, 2014) Yildirim, Hakan; Acik, Selin; Akgungor, Ali Payidar; Dogan, ErdemAlthough concrete roads have been used worldwide for years, the same improvement could not exist in Turkey and therefore, bituminous (hot mix) asphalt roads were preferred instead. In this paper, the vastly built HMA (Hot Mix Asphalt) roads and rarely preferred concrete roads were compared based on their level of noise. For this purpose, the concrete road at the length of 2 kilometers between Afyonkarahisar and Emirdag; also the transition point to the HMA road (which continues after the concrete road) was observed. Both concrete and HMA road ends of this transition point was equipped with a noise measurement device and a camera was installed separately for providing minimum level of noise interference. Consequently, change in the noise levels depending on the building material of roads was recorded simultaneously. These factors was analysed and various models related to the sort of coating was provided. At this stage, the distance between set up points was kept as short as possible in order to prevent different results in the density of traffic and also in the flow rate. Result of various measurements and analysis provided the noise levels of concrete roads being 4 dB(A) less than HMA roads in the comparison based on the same level of vehicle composition and traffic flow. Encouraging the construction of concrete roads in our country is emphasized accordingly.
