All types of modelling are intended as an input to the decision-making process; accessibility modelling provides another input.
Accessibility indicators are central to most transport models. However the indicators are typically used within the model as part of the trip generation and distribution analyses. There are a number of reasons why the existing transport models do not meet the needs required of a potential new accessibility tool.
First, traditional transport models are designed to be most accurate for understanding the demand for travel, and demand is most important for motorised modes. This means the modelling of modes such as walking and cycling, for which demand is rarely a consideration (apart from crowded centres such as major railway stations), is rarely undertaken in detail.
Second, the representation of travel choices and behaviour is averaged down to a few dominant travel groups relevant to demand. However travel demand has many niche markets, not just people with poor mobility who have physical difficulties, but people with different levels of wealth, fitness, expectations and desires. For this reason it has been common for commercial transport operators not to rely on the transport demand models used in road planning, and instead to use accessibility models to understand the characteristics and needs of people in the catchments of bus stops and railway stations. This practice is also used by other commercial operators such as supermarket chains and major retailers who plan networks of stores using accessibility/catchment analysis.
In transport planning, accessibility models are therefore bridging the gap between the sophisticated demand models in current use by transport planners and the crude accessibility models used by market analysts. Figure 4.2 in Part 1 describes the traditional models in more detail. An accessibility modelling approach for New Zealand therefore starts from existing strengths but also recognised weaknesses, not met by either the market or transport demand models, including a lack of understanding of the following:
· How the needs of minority groups are met including physical, security, information, safety and other constraints on transport use.
· What opportunities for walk and cycle access are available since these modes are free at the point of use and capacity is seldom an issue.
· How forecasts of land use change affect travel. Although land use forecasts are used in current travel demand models there is limited analysis of the feedback loops. For example a retail impact assessment might show that a local grocer will close if a new supermarket opens but this level of detail is rarely included in land use transport interaction modelling.
· How modes of transport relate to each other. Models based on demand need to be calibrated based on supply/demand relationships but most trips are multi-modal and calibration of multi-modal trips is more complex.
The strengths of transport demand models are their mathematical modelling created to anticipate the required changes in the transport network when supply and demand changes. These types of models anticipate how many people ‘would’ choose a particular model of travel and consequently these types of models are most useful when identifying where to add capacity to roads, railways or bus systems.
However, assumptions in mathematical algorithms based on past trends are not so useful when looking at the sensitivity to factors not included in the calibration of supply and demand. Therefore it may be that power sockets and Wi-Fi on trains will lead to large increases in rail travel, but a demand model based on the calibration of systems where there is no data on power sockets or Wi-Fi gives little insight into the impacts. Additionally, the analysis of what people ‘would’ do does not recognise other travel options that may be only slightly less economically efficient or are currently not provided so are not possible to utilise.
Accessibility planning acknowledges the opportunity rather than just the ease of moving through the transport network. Modelling accessibility therefore considers what people ‘could’ do. Accessibility planning looks at how changes affect the opportunities for people. It is not always the case that better access capabilities result in more travel demand, but there is a close relationship between travel opportunities and travel demand.
In looking at the need for an accessibility modelling capability the key points are that there are policy sensitive issues not currently being modelled and a potential for new types of model to add value to and expand on the capabilities of existing analysis. Accessibility modelling can include all modes of transport since the focus is the traveller, not the modes they use. These models do not replace demand modelling but are complementary. Accessibility modelling is best used as a precursor to, or in development with, traditional modelling given a number of the initial inputs are the same.