A number of GIS tools that focus on, or are built solely for public transport, have been developed for the purpose of assessing accessibility. These GIS-based modelling tools, most of which have been developed in the UK, may suit the needs of an assessment tool in New Zealand. They provide insight into the key functionalities that will probably need to be built into an accessibility tool for use in New Zealand.
Several accessibility tools that estimate travel time and cost in order to produce an accessibility indicator are described in Part 1 of this report. Accessibility modelling tools are separated into two categories. The first model type is a demand model with accessibility indicator functions typically based on traditional transportation model or the ‘would’ approach. These demand models are highly complex and similar to network transport models. They attempt to create simulations of demand based on the land use of an area. Accessibility is assessed from this demand model, so the analysis is of public transport frequency estimates as opposed to actual public transport timetables. The financial and time cost for demand-based accessibility modelling is high.
The second model types are those that have been developed without demand modelling capabilities and are based on the ‘could’ approach. These tools are less expensive to develop, more diverse and are often highly customised to a specific area. The majority of these custom tools assess whether or not groups are able to reach destinations, and then attempt to measure factors such as time and cost that are involved in reaching the destinations.
Several accessibility tools have already been developed, most of which have been produced in the UK by consultants, academic institutions or local government. A list of these tools is provided in table 4.2. All models include three core components:
2 a relational database
3 customised development or manipulation of GIS tools.
The tools analyse spatial data and public transport timetabling data in order to produce an estimation of time and travel costs for a particular set of OD locations.
The tools produced all involve each of the three components to varying degrees of complexity. Some models run on the back of desktop GIS software, and others are complete self-installing packages that include GIS functionality. It is crucial to note these tools have been developed around policy, and are customised tools for producing results according to accessibility planning specifications. Consequently, it would be difficult to simply adopt these tools in their current format and use them for accessibility planning in New Zealand. Rather, these existing tools and their functionalities should be seen as a guide to the development of a tool based on New Zealand accessibility indicators (once they have been developed) and the level of available data. It is also important to note the majority of the existing tools are used to measure access to public transport or to measure accessibility using public transport. Accessibility assessment in the context of this report will measure accessibility by all modes of transport, including walking, cycling, public transport and private motor vehicle. This is different from the majority of existing international customised accessibility assessment tools.
Common GIS formats, such as ESRI Shapefiles or MapInfo Native Tab files, are used in accessibility tools, either through direct importation, or indirectly through conversion software.
Public transport timetabling and scheduling information also needs to be set in the correct format before it becomes importable into accessibility models. An XML-based data format named TransXchange (DfT 2011b) is being increasingly used by the UK DfT for both journey planning software and in accessibility planning tools. There is no comparable data transfer protocols to the TransXchange data format in New Zealand; however, IPTIS, the journey planning software developed by Jeppesen (a Boeing company) – the engine most used by New Zealand journey planning websites in 2008 – requires a standardised public transport scheduling system. Even so, there is a trend towards the use of a Google public transport standard (Google Transit Feed) and the Auckland and Wellington websites allow users to download the public transport data in this format.
The widespread adoption of a customised international accessibility assessment tool would require the adoption of the data standards used by the tool. All New Zealand data would then be required to run through a conversion process and be continuously maintained or alternatively standardised in an agreed format.
To develop a useful toolkit some tasks can be processed generically nationally (or internationally as Google Maps has shown) using server-based analytical tools. There are other tasks that need specific local knowledge and data and these should build on the general information available in the national tool.
The software supplier industry that seeks to sell large volumes of generic products has been involved in developing bespoke accessibility assessment tools around the world. However, generic products are more suited to generic problems like transport supply and demand, rather than the niche issues and locally specific challenges that accessibility planning seeks to tackle.
To help develop accessibility analysis capabilities, DfT partnered with a number of software developers in the UK to create tools for specific applications. These included the development of the Capital model for London, the development of the Accalc analysis for Scotland and subsequently for calculating the national accessibility indicators, and the investment in Accession for use by local authorities without their own modelling tools.
Accession is described in detail in section 10.5.1 and in appendix A but it is also worth noting that both Capital and Accalc are run on servers rather than PCs. The intention of the server-based approaches is that they can complement other sophisticated server based approaches like Google Maps and Bing Maps. Most data sharing systems are organised via servers with network maps such as Open Street Map increasingly providing global opportunities to create online data about networks and routes.
It was noted in the UK when developing accessibility planning that most planners needed an ‘expert system’ rather than a model. The modelling capabilities already existed in GIS, databases and transport models but support was needed for planners on the questions they should be asking, where they might find the answers, and in automating data collection and management. Accession goes some way towards assisting with the data management tasks but many local authorities have not used it, as intended by DfT, to ask questions about who might face gaps in transport systems (DfT 2004).