How Big Data Drives Intelligent Transport? Explained in 3 Examples
Editor’s Note: The following guest post comes from Dr. Wei Liu, Principal Asset Management Engineer at GHD, Waikato, New Zealand. This is the second part of Dr. Liu’s articles on the topic of Big Data for transportation. In the first part, Dr. Liu discussed the main data sources. In this article, he is going to share with us some case studies.
As mentioned in my previous article, data holds much promise for transport innovation by using the growing amount of mobility-related data. When properly captured and analyzed, these Big Data can provide fresh insights and actionable information to improve transport planning and optimize transport operation.
All key fields of transport, including transport system monitoring and management, asset management, and safety and security, are areas where Big Data and data-related insights can improve transport policy.
Enabling Intelligent Incident Management
In the City of Rio de Janeiro, after a series of floods and mudslides claimed the lives of 72 people in April 2010, city officials recognised the need to overhaul city operations more significantly in preparation for the 2014 World Cup and 2016 Olympics (United States Environmental Protection Agency, 2014).
In collaboration with IBM, the City of Rio de Janeiro launched the Rio de Janeiro Operations Centre (ROC) in 2010 with the initial aim of preventing deaths from annual floods. This centre was later expanded to include all emergency response situations in Rio de Janeiro.
The Rio de Janeiro Operations Centre uses a mobile application (as shown in Figure 1) to warn citizens about heavy rain, strong wind, fog, energy shortages, traffic signal malfunctions, mudslides, fire, smoke and points of flooding. It also receives information from the public.
Figure 1: Views of Waze, a community-based traffic and navigation application
(interface showing the reported incidents and traffic suggestions)
On an average day, Rio’s transport planners receive aggregated views from 110,000 drivers and reports on 60,000 traffic incidents. Since 2013, Rio de Janeiro has been the first city in the world to collect real-time data both from drivers who use Google’s navigation application Waze, and pedestrians who use the public transportation app Moovit. The crowd-sourced data is overlaid with real-time information from various sensors and cameras. In the future, the city plans to start monitoring how cyclists move around the city using cycling app.
Building Intelligent Network
A recent HubCab initiative was carried out by the MIT SENSEable City Lab in partnership with Audi and General Electric (GE). HubCab analyses taxi trips to explore the benefits and impacts of vehicle sharing in New York City. The data was derived from the records from over 150 million trips made by 13,586 registered taxis in Manhattan during 2011. (Santi, et al., 2014) The GPS-enabled taxis reported on the geographic coordinates (longitude and latitude) and time of each trip’s origin and destination, creating a map of pick-up and drop-off points as shown in Figure 2.
Figure 2: Pick-up and drop-off points of all 170 million taxi trips over a year in New York City
The resulting data set could be used to study more conventional queries, such as the location of the nearest taxi or most efficient route for a single trip. The innovative aspect of the HubCab project, however, is in its capacity to model and optimise trip-sharing opportunities through what is referred to as the “shareability network”.
At a conceptual level, “shareability network” involves building a network of links between nodes. In the HubCab initiative the nodes represent individual taxi trips and the links connect trips that can be combined. Taxi routes are recalculated in real-time to pick up new passengers based on their current location and desired destination. (Santi, et al., 2014).
Conclusions drawn from this exercise have shown potential impacts of taxi sharing both at the level of the city and of the individual. In cities like New York, taxi services account for a major mode of travel. Hence, reducing the number of taxi trips would lead to dramatic reductions in air pollution and traffic congestion.
In 2014, the researchers at MIT SENSEable City Lab sought to move beyond linking traffic accidents to congestion towards more predictive visualisation tools for traffic management centres. Their “Traffic Origins” study introduces time-lapse visualisations to observe how congestion propagates from a point source, or, when used in after-action reviews, to understand the effectiveness of mitigation measures (see Figure 6). Despite having targeted expert users, the study emphasises the effectiveness of simple, aesthetically pleasing visualisations to communicate complex relationships to both traffic controllers and laypersons (Anwar, et al., 2014)
Managing Intelligent Asset
Transportation Asset Management is a strategic approach to the optimal allocation of resources for the management, operation, maintenance, and preservation of transportation infrastructure. One of the key aspects of the development of Asset Management is data collection and usage. The way in which transportation agencies collect, store, and analyze data has evolved along with advances in Big Data technology, such as mobile computing, advanced sensors, distributed databases, and spatial technologies. These technologies have enabled data collection and integration procedures necessary to support the comprehensive analyses and evaluation processes needed for Asset Management.
A new framework (Roshandeh et al. 2014) as illustrated in Figure 3 for bridge asset management and health monitoring has been developed by incorporating Big Data and Cloud computing technologies with principles of transportation asset management strategy.
Figure 3: Big Data Framework for Bridge Data Management and Health Monitoring (Source: Roshandeh et al. 2014)
The conventional framework for bridge data management and health monitoring was upgraded. In this framework, Big Data application and Cloud computing were incorporated with transportation asset management fundamentals. There are two phases involved in this framework – phase 1 is “Cloud” based and phase 2 is “decision” based.
- First phase is comprised of three layers including data collection, data transfer and data storage as well as visualization. These layers could get benefit of cloud computing technology and handle the processing of datasets using cloud resources.
- In the second phase, which is mainly coming from asset management strategy, experts and decision makers viewpoints are closely embedded into the framework.
Summary
The transportation industry has been experiencing unprecedented amounts of data captured from different sources such as on-board sensors and data collection points introduced by passenger counting systems, vehicle location systems, ticketing and fare collection systems, and scheduling and asset management systems. Those big data sets contain unprecedented potential for deriving insights into planning and managing transportation networks.
Case studies presented in this article indicate that, through adopting Big Data technologies, transportation network can be more effectively monitored and managed, innovative models can be developed and implemented to enhance safety and resilience of the network, and better decisions can be made on asset management.
Resources:
- United States Environmental Protection Agency, 2014. U.S.-Brazil Joint Initiative on Urban Sustainability. at: http://www.epa.gov/jius/projects/rio_de_janeiro/rio_operations_center.html
- Santi, P., Resta, G. & Ratti, C., 2014. Quantifying the Benefits of Taxi Trips in New York through Shareability Networks. ERCIM News 98.
- Anwar, A., Nagel, T. & Ratti, C., 2014. Traffic Origins: A Simple Visualization Technique to Support Traffic Incident Analysis.. s.l., IEEE Pacific Visualization Symposium.
- Arash M. Roshandeh, Rashed Poormirzaee, and Forough Sheikh Ansari. Systematic Data Management for Real-Time Bridge Health Monitoring Using Layered Big Data and Cloud Computing. International Journal of Innovation and Scientific Research ISSN 2351-8014 Vol. 2 No. 1 Jun. 2014, pp. 29-39.
- How Big Data Drives Intelligent Transport? Explained in 3 Examples - November 30, 2016
- Top 3 Big Data Sources Driving Breakthrough Transport Intelligence - November 28, 2016
