Real-Time Passenger Information Systems in Rural Areas
There is a recognition that real-time passenger information (RTPI) systems play a critical role in making public transport more accessible. Delivering information, such as vehicle location, estimated arrival time, and service disruptions in real-time can support passengers making journey decisions and have a positive influence on passenger experience.
Recent advances in Information and Communication Technologies (ICT), such as smartphones and social media, have resulted in a growing passenger demand for this information to be delivered via multiple channels, e.g. websites, smartphone apps, SMS, and social media feeds, along with more traditional methods, e.g. variable message displays at bus stops or in-vehicle.
The Informed Rural Passenger Project
When considering RTPI systems for bus services in rural areas, several challenges must be considered: there is typically a lack of infrastructure, such as GPS on vehicles and variable message displays, hail-and-ride sections making it more challenging to accurately predict arrival times, and routes tend to be longer and services less frequent and so the impact of disruptions can be greater.
The Informed Rural Passenger (IRP) project at the dot.rural Digital Economy Hub at the University of Aberdeen, UK, is investigating how these, and other challenges, can be addressed for rural RTPI systems through capitalising on advances in ICT and a growingly popular concept called crowd-sourcing. Crowd-sourcing is usually defined as using a group of people to perform some task that would otherwise be performed by an expert; the IRP project adapts this approach to have passengers provide information about their journey on public transport. This information can come from the passenger themselves or from sensors on their smartphone (e.g. GPS) and is uploaded to the IRP servers where it is integrated with different types of data (timetable, route details, geographical) from several providers (open data from government, online communities, other users). This is then used as the basis for the RTPI provided to users. The system has been designed to accommodate a number of input and output channels. To date, GetThereBus, a prototype Android smartphone app has been trialled, with SMS (text messaging) channels in development.
The GetThereBus app has been developed whilst working with FirstGroup, the operator of several key routes in the Scottish Borders. The app was developed for rural areas and initial trials took place in Aberdeenshire. As part of promoting the GetThereBus application in the deployment area two Bus Stop Tours took place in July and August 2013. These ‘tours’ involved project team members spending 2-3 hours each time at key bus stops in the area (Galashiels, Hawick, Selkirk and Peebles). Whilst at the bus stop the GetThereBus app was promoted, useful information on mobile phone ownership gathered and many informal interviews took place. In total, from eight tour stops, over 900 people were spoken to about the GetThereBus app. The bus stop tour, as well as being a useful public engagement activity, gathered field intelligence that continues to influence the project team’s work
Using Open Data
Open data, the idea of making data available that can be “freely used, reused, and redisributed by anyone” [http://opendefinition.org/], has been growing in popularity in recent years. Potential benefits of open data include economic (growth and job creation), improved public services, and more transparent and accountable government.
When developing GetThereBus, the IRP team opted to use open data wherever possible to meet the data requirements of RTPI systems. This included the National Public Transport Access Nodes (NaPTAN) dataset published by UK Department for Transport which provides details of every access point (e.g. bus stop, bus station) to public transport in the UK; and OpenStreetMap, an alternative to Google Maps, which makes not just the graphical map tiles available, but also the data they present (e.g. details of road segments). However, at the time of development, several key datasets were not openly available, namely bus timetables (although this has now changed), details of routes travelled by buses, and real-time bus locations. Timetable and bus route information was provided by Traveline and FirstGroup, with the aforementioned crowd-sourcing approach used to obtain real-time bus locations.
These datasets form the basis of a transport information ecosystem developed to support the GetThereBus RTPI system. A major challenge here was integrating the heterogeneous datasets in a manner that would allow updates to be automatically included in the system. As part of addressing this, an approach based on Linked Data was used. Linked Data is a technique for using web technologies to connect data that is currently not linked – for example, linking the real-time location provided by users to the timetables provided by Traveline, which are themselves linked to the UK Government’s NaPTAN dataset. This approach also allows the addition of semantics to the data, i.e. explicit descriptions of its meaning and the nature of the links between different pieces of data. This allows machines to process and reason with it to, for example, discover further related data which can then be used to improve the system.
The GetThereBus smartphone app provides users with access to the GetThere RTPI system. Users start by selecting the bus route (and direction of travel) that they desire information for. The bus route, location of bus stops, estimated bus locations (based on the timetable), and real-time bus locations are then overlaid on a map. To overcome the problem of poor mobile data coverage in rural areas, the necessary map tiles are stored within the app. The time of the next and previous buses at a stop can be retrieved by tapping on a bus stop. As buses on the routes covered by GetThereBus are not typically equipped with GPS, when travelling on a bus GetThereBus users are asked to share their location with the system, which uses the user’s location as a proxy for the bus’ real-time location provided to other users.
Having passengers provide bus locations potentially introduces low quality location information into the system. GetThere features a quality assessment mechanism that is capable of evaluating passenger-provided locations against a number of criteria, each of which consider a different aspect of the location. For example, relevancy considers how close the reported location is to the expected route of travel; accuracy considers the error margin associated with the GPS reading. This is calculated on-demand when a user taps on the real-time location on the app, with results of the system’s analysis presented to the user to help them decide if they trust it or not.
As is often recognised, RTPI is particularly valuable during disruption events, yet it can be the case that the first people to know about a disruption such as a road accident or flooding are those travelling on the affected roads. When such a disruption is encountered, users are encouraged to use the GetThereBus app to report it, and then later to report that it is cleared. Along with being displayed on the map for the benefit of other users, disruptions are also integrated with the ecosystem allowing other components (e.g. quality assessment) to use them in their reasoning.
Does Access to Real-Time Passenger Information Change Behaviour?
To investigate the effects of providing RTPI via a mobile device to rural public transport users, the IRP team conducted a travel behaviour study in the Scottish Borders, which includes areas ranging from urban to remote rural. Previous work identified three main types of passengers: those using the bus service as a cheap alternative to the train service; local residents without regular access to a car and that use the service for short trips, including commuting, shopping, and entertainment; and students that commute between Galashiels and various villages, towns and cities (visited) by the bus services.
For the purpose of this study 15 participants were recruited (nine female, six male), from the student passenger group. Nine of the 15 participants were frequent users, making more than five trips week; the average age of the participants was 28. At the beginning of the study participants’ existing travel behaviour was categorised by variables that were previously identified from the literature and explored through semi-structured questions and ride-alongs.
Participants were then given the GetThereBus app and asked to use it for 20 days. Following this, their travel behaviour defined by the same variables was explored through semi-structured interviews. This qualitative approach enabled the team to explore the travel behaviour based on the common experiences and the individual stories of the participants. The variables that were affected by use of the GetThereBus app include: increased feelings of journey control, bus service perceived as easier to use, increased willingness to pay, perceptions of improved bus service, waiting time decrease, and increased overall satisfaction. In contrast, four variables were only minimally affected by the app: decreased uncertainty, passenger’s changing their journeys, impact on decision making, and feelings of safety.
The GetThereBus app is just one way in which information can be provided from the system to users; the team are also investigating the use of SMS for people who do not own or have access to a smartphone. The team are designing a solution “BusM8” with older adults that allows them to “have a conversation” with the system via SMS to obtain information.
An SMS experience prototype was created, called BusM8, that enabled the passenger to go through the steps required to obtain real-time travel information using SMS texts. To evaluate the prototype, and also gather further material relevant to the IRP project, a series of co-design workshops were organized to engage with a user group consisting of older adult females who use bus services in Aberdeenshire. There were 3 workshops with 8 participants in total. The workshops involved a journey mapping activity, followed by a demo and use of the BusM8 prototype and then a discussion on the topics relevant to the design.
There were a range of findings from the workshop activities related to the experience of bus travel in rural areas, use of mobile phones, and usage of travel information. A key finding was that there must be a consideration of alternative ways of delivering real-time information to bus users who are uncomfortable interacting with a mobile phone. However, there is potential to encourage learning between those who are comfortable using the phones and those who have phones but are less able to exploit the phone’s full potential.
The experience prototyping approach was particularly successful at demonstrating the SMS service to the users. Some participants even asked if this service was available already or inquired if the bus operators would be taking it on. The BusM8 user group (currently older female adults) may be considered challenging users to work with in terms of new technology adoption. However, as the BusM8 work has shown, it is possible to engage with this user group and create a meaningful dialogue around the design.
The next stage of the project involves investigating existing and potential roles of social media in the context of RTPI. This includes: working with relevant parties (such as FirstGroup) to understand how social media, such as twitter, is currently used, particularly for rural areas; designing novel approaches for the use of social media in this setting; and developing technologies to integrate social media feeds into the existing GetThereBus system.