Journal Articles
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/19884
Browse
3 results
Search Results
Item Forecasting public transit passenger demand: With neural networks using APC data(Elsevier Ltd, 2022) Halyal, S.; Mulangi, R.H.; Harsha, H.The implementation of Intelligent Transportation Systems (ITS) as a part of smart mobility is crucial for solving the current problems of the transportation industry. The setting up and maintenance of ITS requires not only the current passenger demand but also the future passenger demand. The future passenger demand can be obtained with time-series forecasting carried out with different techniques. With the advancements in the technological field, modern and more advanced methods of time-series forecasting using deep learning are being preferred over traditional forecasting techniques. However, the research carried out in this regard is quite limited, particularly considering the Indian scenario. Hence this research work focuses on exploring the performance of deep learning forecasting techniques considering the aspects mentioned previously. Here, the forecasting of passenger demand was done with Long Short-Term Memory (LSTM) using the three months Automatic Passenger Counter (APC) data of the Hubballi-Dharwad Bus Rapid Transit System (HDBRTS) as part of a case study. Then the forecasting of passenger demand was also done with Seasonal Autoregressive Integrated Moving Average (SARIMA), and the comparison of the forecasting accuracy of both methods was made using Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). Furthermore, to validate the results, novel approach has been adopted for the process, by following some more time-series resampled with different time intervals. Study shows that LSTMs will be used satisfactorily in the traffic conditions of developing counties, for forecasting passenger demand using APC data. Study also provides detailed guiding methodologies of advanced methods of passenger forecasting along with conventional ones. © 2022 World Conference on Transport Research SocietyItem Short-term passenger demand modelling using automatic fare collection data: A case study of hubli-dharwad brts(Aracne Editrice, 2023) Halyal, S.; Mulangi, R.H.; Harsha, H.M.A well-planned Intelligent Transport System (ITS), is the need of the hour for solving the problem of excessive traffic and unacceptable travel durations in public transit systems. However, for the successful operation of an ITS, the forecasting of passenger demand on a regular basis is essential. Obtaining reliable and accurate passenger data required for forecasting passenger demand is a genuinely tedious task for most of the researchers. The current study focused on short-term forecasting of passenger demand using reliable ITS-based data from Hubli-Dharwad Bus Rapid Transit System (HDBRTS) as a case study. The aggregate pattern of passenger demand data was visualized using one month of Automated Fare Collection System (AFCS) data. Subsequently, Autoregressive Integrated Moving Average (ARIMA) and Seasonal ARIMA models were used on the data of selected stations and tested for their forecasting accuracy. Outcomes of the current case study show that Seasonal ARIMA models are the most suitable for forecasting the passenger data over the conventional ARIMA models. The most suitable and reliant models were evidenced through lower values of Mean Absolute Percentage Error (MAPE), Akaike’s Information Criterion (AIC), and Bayesian Information Criterion (BIC). The current case study helps in having an inclusive view of the passenger flow characteristics for the further advancement of the system. Research work in the Indian context, especially on the usage of AFCS data, is quite limited, this study makes to explore the possibility of using AFCS data efficiently in forecasting passenger demand. © 2023, Aracne Editrice. All rights reserved.Item Spatiotemporal capacity estimation of bus rapid transit system based on dwell time analysis(King Saud University, 2024) Angadi, V.S.; Halyal, S.; Mulangi, R.H.The performance study of an urban transport system, particularly a Bus Rapid Transit System (BRTS), must report on its operations and reliability. Such study of BRTS comprises numerous facets, including capacity, which directly influences how the system practically operates and serves the commuters. Hubballi-Dharwad Bus Rapid Transit System (HDBRTS) has been operational since 2018. A performance study is necessary to evaluate the performance of HDBRTS, which aids in its upgradation and improvement. The current research uses the experimental technique through an innovative and inspired basis to comment on the HDBRTS's performance by estimating the corridor's operational capacity at different spatial and temporal fluctuations. The selected route of the HDBRTS comprises combined segregated (exclusive traffic environment) and unsegregated (mixed traffic environment) stretches. The current study mainly conducted a video graphics-based survey to acquire the necessary data on identified spatial and temporal trends at various HDBRTS bus stations. The essential data gathered consists of Dwell Time (DT)-based data at each station, summarising the total time a bus takes to serve passengers at a station. DT is inversely proportional to the capacity of the particular bus station, which is related to the Failure Rate (FR). FR values of all the bus stations of the route were analyzed using DT, and then capacity values were calculated at different spatiotemporal patterns. Study results show that the busiest stations of the identified routes with critical DT values have FR values in the range of 1–2%, contradicting previous studies. The variations in the capacity of the stations, both spatially and temporally, were graphically represented with the minimum capacity of the segregated stretch as 36 buses/hr and the unsegregated stretch as 31 buses/hr. Finally, the Level Of Service (LOS) of the chosen study corridor was developed using the K-Means clustering algorithm and validated using the Silhouette Coefficient technique. The silhouette coefficient values obtained range from 0.52 to 0.74, indicating a reasonable structure. © 2023 The Authors