China
Africa
Explore chapters and articles related to this topic
Smart card data and its use in public transport research
Published in Corinne Mulley, John D. Nelson, Stephen Ison, The Routledge Handbook of Public Transport, 2021
Nevertheless, the value of smart card data is quite limited when considered on its own, since the information derived directly from such data would be limited to the number of taps on and taps off (i.e., demand) at a particular station/stop, possibly segmented by time of day. To make it more useful, analysts have to bring in more data from other sources, such as service timetable data, automatic vehicle location data, census data, household travel survey data and meteorology data, to name just a few. Along the process of marrying the smart card data with secondary data lie various assumptions. For example, to estimate the waiting time by combing the smart card data with the timetable data, it is necessary to assume that services are punctual to the timetable if automatic vehicle location data (or GPS data of the vehicles) are not available. Similarly, to predict where transfers would occur in a closed-system (i.e., a system where passengers only leave trace at the entry and exit points) using some kind of passengers-to-train assignment models, some rules/assumptions must be used. These could be simple, with equal probabilities for each possible transfer locations, or more sophisticated, with some behavioural theory such as utility maximisation and Bayesian theory. The point is that there is no smart card system that provides all the data required that one might need, and thus all sorts of assumptions are needed.
Deep Learning for Prediction of Bus Arrival Time in Public Transportation
Published in Turan Paksoy, Çiğdem Koçhan, Sadia Samar Ali, Logistics 4.0, 2020
Faruk Serin, Suleyman Mete, Muhammet Gul, Erkan Celik
Public transportation network mainly consists of route, stop, and bus. A line between two sequential stops on a route is defined as a segment. A bus travel time on a segment is calculated using automatic vehicle location data as in (8) where tvb is bus, v, arriving time at beginning-station, b, of segment s; tve is bus, v, arriving time at end-station, e, of segment s. Travel time of all buses on segment, s, are arranged sequentially as time series as in (9). Finally, series are rearranged according to time window as in Table 1 (time window =3).
Wireless Sensor Networks for Intelligent Transportation Applications : A Survey
Published in V. Çağri Güngör, Gerhard P. Hancke, Industrial Wireless Sensor Networks, 2017
Kay-Soon Low, Marc Caesar R. Talampas
Wireless signals can be detected easily by data thieves or eavesdroppers. Thus, there are also security concerns over the deployment of wireless sensors [33]. Technology such as Zigbee uses 128-bit AES algorithm to provide strong authentication mechanisms that prevent unauthorized devices from joining and using the network key to control devices. However, the collected vehicle location data in traffic monitoring systems must be kept secure as well. This requires other schemes aside from strong authentication and encryption mechanisms. In [96], the location privacy of drivers is protected by anonymizing the IDs of cars. Their system uses short IDs which are derived from the vehicles full ID using periodically updated random patterns. The use of short IDs allow vehicles to be re-identified without revealing their full identities.
Digital technologies for energy efficiency and decarbonization in mining
Published in CIM Journal, 2023
A conceptual system to improve haul truck safety at surface mines comprises truck location and route mapping, cameras and AI algorithms for fatigue monitoring, and cameras for improving vision (blind spots, image-restoring algorithms in adverse climate conditions) integrated with warnings and alarms (Sun, Nieto, Li, & Kecojevic, 2010). Some of these technologies could lead to other fuel consumption, GHG emission, and productivity benefits. For example, in addition to helping avoid collisions, the vehicle location data could be used to enhance scheduling and limit vehicle idling, as was proposed by Quash (2019). Fleet management software integrating ML and cameras to capture data related to loading and hauling activities could also be beneficial.
ABAFT: an adaptive weight-based fusion technique for travel time estimation using multi-source data with different confidence and spatial coverage
Published in Journal of Intelligent Transportation Systems, 2023
Sara Respati, Edward Chung, Zuduo Zheng, Ashish Bhaskar
Figure 1a shows an example of a vehicle trajectory obtained from GPS traveling over link C1. GPS provides time and vehicle location data. Subsequently, the speed between two points can be calculated based on the distance traveled and the time taken, written as: where and are the speed, distance traveled, and travel time, respectively, of information i. Figure 1b illustrates the recorded information in detail.