Opening up opportunities to benefit the individual

Open data provides opportunities for exploration and creativity. Such innovation will open new doors in the public sector, and enable new private sector possibilities

­When the data.gov.uk website launched in January 2010 I did not even notice, to be honest. A failure on my part as a data ‘geek’, I admit.

However, in April 2010, following a more trumpeted arrival (The Guardian’s ‘free our data’ campaign, for example), I was the proverbial kid in a sweet shop ordering as much data as my download facility could handle. ‘Free’ Ordnance Survey data had arrived and they were ‘giving away’ all the data I could ever want for the price of an internet click, not to mention free DVDs to boot.

From memory, I believe this data previously had a value of £0.25m ‘on the street’ for complete UK coverage – so it was a good start! I would have liked OS MasterMap Topography and ITN data as well but you can’t have everything…

With awareness of data.gov.uk now switched to the ‘on’ position, in August 2010 the Department for Transport landed on data.gov.uk, in the form of the publication of the National Public Transport Data Repository (NPTDR) – passenger transport data for all of Great Britain, all free and all available to use without restriction, whereas previously access was through and only for public sector agencies and/or projects.

As a network and accessibility planning specialist, through the OS and DfT open data releases I now had at my disposal all the data I needed to do my work for free (I could even use open source software if needed to).

To gauge use by other transport planners at Halcrow (all of whom have Geographical Information Systems experience), for this article I conducted a quick poll of open data use. The results showed that most (75 per cent) were aware of open data, particularly Ordnance Survey data), with 75 per cent of these people actively using the data in some guise. In the majority of cases most use was attributed to using OS imagery files (250,000 and StreetView rasters) as background figures in reports, typically overlaid with outputs from transport models.

My use has been slightly different and the remainder of this article describes what I have done over the last six months.

By way of background, prior to joining Halcrow in 2007 I spent six, formative, years at GMPTE (now Transport for Greater Manchester) observing a steady but rapid rise in the use and management of Geographical Information Systems (GIS) and data. As Ordnance Survey and Census liaison, part of my role, getting data, was not difficult. Granted, someone in the organisation paid for this somewhere, most of the time, but for me it was still free and available to use.

Picture me as I ran through the green fields of Greater Manchester (maybe best not to, actually) – I could explore and see what was achievable. When I joined the private sector, however, I no longer had access to such data, or at least no one would pay me to explore the green, green grass anymore, but at least I had paid work to do.

Turning back to open data, what opportunities has it given me so far? Well, it has given me the chance to explore once more, at no cost – excluding my time – with no public sector client required. The green fields have returned, although maybe they should be considered more of a brownfield redevelopment due to the current economic climate.

How have I used the data made available? Or, to use one of those annoying phrases, how have I ‘added value’ to produce something more than what is being offered; and done so in a cost efficient manner.

With Ordnance Survey data (background images at many scales, postcode locations, road networks) and NPTDR data, my aim has been to build a GIS tool that brings the two together, in the form of a multi-modal network analyser tool called ‘Solaris’.

In the past I have spent many a day/week/month ‘free-hand’ plotting bus networks in GIS, cross-referencing with timetables for routes, never quite getting my mapped nodes to match up. What I needed was an automated tool that did this for me, for any part of Great Britain, and did so so neatly and quickly, and consistently to the road network – which became the initial successful aim of ‘Solaris’.

At the cornerstone of ‘Solaris’ is an Excel/GIS tool, utilising Visual Basic and MapBasic programming scripts, that works by firstly manipulating the NPTDR data by arranging it into rows of origin-destination rows (public transport node to public transport node). For the bus network, the tool maps the OD pairs against the quickest routes between the two points to create the mapped full network; for non-road modes of public transport a pre-drawn network is required, split by station.

The pertinent point to make here is that this was done with no prior VB/MapBasic programming experience on my part, so open data really was a catalyst to experimenting and this application of open data is possible for anyone to do.

The tool is hardly groundbreaking or a game changer to the transport planner but it does allow a network to be drawn fast and efficiently and, most importantly, gathers intelligence about change scenarios dictated through an Excel interface in a cost-effective manner.

Through the interface, routes can be quickly switched on and off, minimum frequency of service and time-periods set (a facility relevant with many transport authorities juggling subsidies to bus networks in the quest for for efficiency savings). Intelligence is gathered by weaving data in from a third open data source, the humble postcode centre mapped against Census and deprivation data (was the Census of 2001 the first open data source placed on the internet from central government?).

A further aim of ‘Solaris’ involved the creation of additional GIS tools to understand public transport operators and how they interact temporally and spatially with each other and, potentially of more use and importance for my work, the creation of an accessibility planning application.

The image below is an example output from ‘Solaris’ and shows the bus routes running within the city centre of Manchester and the operators running those routes. The entire bus network for Greater Manchester was created in little under half a day (computer time), with the output shown and associated analysis being relatively instantaneous to complete.

The accessibility planning section of the tool has been developed to support our use of mainstream accessibility tools (AccessionTM for example) for projects, rather than replace them. In ‘Solaris’ the accessibility tool looks at one destination only but has functionality that:

• Builds pictures of travel times temporally by day;

• Allows the user to specify perception weightings (to walk and wait time, plus interchange penalties);

• Restricts calculated access time to either direct or direct plus one interchange journeys only and;

• Creates, in GIS, direct or direct plus one interchange route networks, so that the user has information on how access is possible (including frequency of service).

Temporal accessibility allows the creation of accessibility maps for one destination over a 24-hour period (set at 15 minutes, 30 minutes or hourly time-periods). The purpose of these maps is to build a picture for a full day (a practical use of this being hospital appointments, which could be set according to when is best to travel for the patient).

By allowing the user to set perception weightings, as per traditional transport models, it allows accessibility to take into consideration impediments to travel (excessive walk distance, wait times and interchange). Further refinement could be made using open data from the UK’s police forces on the perception of crime, for example (disclaimer: the merits of doing this are open to debate, but not one that the author is opening).

By restricting access to only direct journeys, or direct plus one interchange journeys, it gives a more realistic view on accessibility levels. Existing accessibility tools do not restrict levels of interchange. Overall, the accessibility tool within ‘Solaris’ could be considered less of a black box with the above considerations, albeit restricted to one destination.

An example of this application is found in the image above, which shows travel times to Leeds city centre in the weekday am peak period (travelling between 8am and 9am).

To conclude, open data has given me the chance to explore and innovate, which will in the future help ‘pay the bills’, hopefully. In the public sector and my core work, open data is not so significant to me as a contractor, as the data has always been there freely through contractor licenses, as project costs to the client will not change.

However, it is hoped that innovation, such as described here, using open data, will open new doors in the public sector. Furthermore, it allows me now to work more in the private sector, whereas before prohibitive data charges for say postcode, road and background imagery data made my work (GIS and accessibility planning) too expensive, or at least ‘nice to have’, but not essential.

Now all I want is MasterMap data to be free, and a complete GIS rail network.