A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
A Scalable Approach to Inferring Travel Time in Singapore’s Metro Network using Smart Card Data
Knowledge of the travel time in each component of the Singapore’s metro network is an important building block in our ongoing construction of an analytics system that aims to generate insights to support transportation planning and business decision making. It is useful for understanding individual passengers’ route choices, estimating crowd flow across the network, and identifying network deficiencies to facilitate future train line designs for improving passenger travels. Existing work on inferring travel time in a metro network, however, tends to consider the problem on scales too small, requires data sources that are not easily available, or exhibit limitations in terms of computational efficiency. To address this problem, we propose an automated multi-stage method for inferring the time variable in various components of a metro network. This method requires only the input of a smart card dataset, and it can easily scale to data with millions of tuples. We evaluate the proposed method for a route planning application, using smart card data from Singapore, and compared our estimated results with ground truth values. Our experimental results show that the proposed method demonstrates a satisfactory inference accuracy, offers much higher accuracy than baseline methods based on prior work, and is scalable to large datasets.
A Scalable Approach to Inferring Travel Time in Singapore’s Metro Network using Smart Card Data
Knowledge of the travel time in each component of the Singapore’s metro network is an important building block in our ongoing construction of an analytics system that aims to generate insights to support transportation planning and business decision making. It is useful for understanding individual passengers’ route choices, estimating crowd flow across the network, and identifying network deficiencies to facilitate future train line designs for improving passenger travels. Existing work on inferring travel time in a metro network, however, tends to consider the problem on scales too small, requires data sources that are not easily available, or exhibit limitations in terms of computational efficiency. To address this problem, we propose an automated multi-stage method for inferring the time variable in various components of a metro network. This method requires only the input of a smart card dataset, and it can easily scale to data with millions of tuples. We evaluate the proposed method for a route planning application, using smart card data from Singapore, and compared our estimated results with ground truth values. Our experimental results show that the proposed method demonstrates a satisfactory inference accuracy, offers much higher accuracy than baseline methods based on prior work, and is scalable to large datasets.
A Scalable Approach to Inferring Travel Time in Singapore’s Metro Network using Smart Card Data
Lin, Xi (author) / Xiao, Xiaokui (author) / Li, Zengxiang (author)
2018-09-01
204452 byte
Conference paper
Electronic Resource
English
Inferring mobility of care travel behavior from transit smart fare card data
| DOAJ | 2024
| Taylor & Francis Verlag | 2020
Decomposing Metro-Bus Transfer Time with Smart Card Data
| TIBKAT | 2022