The Effects of Park and Ride Supply and Pricing on Public Transport Demand
6 Cross Forth case study
6.1 The strong competition between bus and rail based park and ride between Fife and Edinburgh is examined. There is a large existing travel market for Cross Forth trips between Fife and Edinburgh, whilst the future public transport strategy for this corridor could be affected by the proposed new crossing. The park and ride sites at Ferrytoll and Inverkeithing help to tackle congestion, particularly as the number of daily vehicles using the Forth Road Bridge exceeds 60,000. The diversion of some drivers onto public transport helps to support employment levels in Edinburgh city centre, and 'control' congestion levels for other motorists. Two case studies were examined:
- Ferrytoll: the site is located at Inverkeithing near the Forth Road Bridge. The site is open daily from early morning until after midnight. There is a high frequency bus service, with departures every 5 minutes towards Edinburgh at peak times, with other services to Edinburgh Airport, Gyle and Edinburgh Park. There is free parking with 1,040 spaces. Journey times to Edinburgh during the peak periods are about 40 minutes, with return daily fares of £4.70
- Inverkeithing: there is rail based park and ride to Edinburgh in the town. Access from the strategic road network is less convenient compared with Ferrytoll. There are at least 5 trains per hour towards Edinburgh, with journey times of about 25-30 minutes. Although the rail service is less frequent, journey times are faster, especially in the peak periods when buses are more readily affected by congestion. However, rail fares are more expensive (£7.50 for a return journey). There are 425 parking spaces and 943,000 single trips per annum23
6.2 The generalised journey time for rail and bus has been incorporated into the modelling framework to compare journey choices on a consistent basis.
Primary research - overview of the findings
6.3 Passengers surveyed have relatively short access times to Ferrytoll and Inverkeithing, as shown in Figure 6.1. Almost 50% have a journey time less than 10 minutes, with a further 35% with a journey time between 10 and 20 minutes, illustrating the concentration of trips originating close to these places.
Figure 6.1: Journey times from trip origins to Ferrytoll / Inverkeithing
Source: Arup analysis of Accent data, sample size shown
6.4 The length of journey times to the final destination from the railway station or the bus stop in central Edinburgh as shown in Figure 6.2 is even more concentrated compared with the data presented in Figure 6.1. For example, nearly 85% of respondents have a journey time of less than 10 minutes to their final destination, whilst a further 14% have timing between 10 and 20 minutes.
Figure 6.2: Journey times to final destinations from Central Edinburgh
Source: Arup analysis of Accent data, sample size shown
6.5 Similar to the results from the bus based park and ride surveys, there is a high proportion of trips with a destination in central Edinburgh. The destination for around 85% of rail respondents is Edinburgh city centre, with 11% of people surveyed travelling to Haymarket. Table 6.1 presents the results.
| Trip Destination | Ferrytoll | Inverkeithing | Total |
|---|---|---|---|
| Edinburgh | 82 | 53 | 135 |
| Gyle | 2 | 2 | 4 |
| Haymarket | 3 | 14 | 17 |
| Leith | 1 | 0 | 1 |
| Rest of Edinburgh | 2 | 0 | 2 |
Source: Arup analysis of Accent data
6.6 Commuting and shopping account for the highest proportion of trips presented in Figure 6.3. Commuting accounts for about 45% of the total, with about 28% of passengers making a shopping trip. The percentage of passengers making other types of trip was relatively small.
Figure 6.3: Journey purpose
Source: Arup analysis of Accent data, sample size shown
Results from the stated preference research
6.7 More detailed results of the analysis of the SP data is presented in Appendix A3. The models produce a good statistical correlation, although some individual coefficients are not statistically significant. Separate models have been developed for commuting and non-commuting markets for each mode.
6.8 There are several important conclusions to draw from these results:
- frequency: the impact of service frequency is very low and appears relatively unimportant, given there are regular departures
- out of vehicle time (OVT): this has a strong influence on modal choice, given its relationship relative to in-vehicle (IVT). The OVT is higher compared with IVT, so the time spent getting to the final destination comprises a larger component of the overall journey
- mode constant: the mode constant indicates the preference for one mode compared with the other. The size of this parameter indicates the choice between rail and bus is relatively fixed. This conclusion is also reinforced by the outcome from the sensitivity tests which indicating there was relatively few passengers switching modes in response to the alternative travel choices presented
- values of time: The implied values of time vary between about 2 and 4 pence per minute, somewhat lower than other guidance, for example, Department for Transport WebTAG guidance24. This implies park and ride users are choosing to use park and ride even though the journey times are slower to avoid paying to park in central Edinburgh
- bus journey times: times from Ferrytoll appear less important compared with other examples reviewed in Chapter 5, although the unique characteristics of Ferrytoll need to be acknowledged accordingly
6.9 The analysis presented in Chapter 6 highlights the main issues affecting the viability of rail and bus based park and ride serving the Cross Forth corridor. Primary research has been collected from a range of existing users at Ferrytoll and Inverkeithing and the following outlines some important issues to shape the development of the future public transport strategy for the corridor.
- objective 1 (changes to parking supply and pricing affecting public transport usage): the results from the primary research indicate passengers using Ferrytoll and Inverkeithing park and ride have a relatively low value of time. This implies users are choosing to switch to public transport before the Forth Crossing to avoid paying the high parking charges in Edinburgh, even though the overall journey time by bus or rail is longer compared with driving. As a result, changes to parking availability will have a limited impact on travel behaviour
- objective 2 (changes to parking supply and pricing influencing modal shift): the scope to encourage future modal shift to public transport appears to be influenced by future parking policy and the distribution of new employment in Edinburgh. This is an important consideration, since the modelling analysis highlighted the relatively unresponsive choices by passengers to alternative travel options, given the time spent travelling to the final destination is fixed. As a result, the scope to encourage passengers to switch between rail and bus (or vice versa) in response to service improvements appears limited. Therefore, the opportunities to grow the Cross Forth public transport market will be influenced by the distribution of future employment and its accessibility to public transport nodes in terms of egress times and the cost of parking
- objective 3 (relative importance of complementary factors): although the overall journey times by public transport are slower compared with driving, the cheaper overall cost helps to offset these impacts. Two factors have emerged from the primary research which influences the overall success of Ferrytoll and Inverkeithing. The importance of high frequency services is reflected in the modelling outputs, whilst the relative cost of the rail and bus services versus the alternative parking choices has clearly contributed to the number of passengers using the park and ride
- objective 4 (identification of undesirable outcomes): results from the primary research indicate Cross Forth public transport has generated limited negative impacts. The access times to Ferrytoll and Inverkeithing are typically relatively short, indicating the car distances using the network are relatively short. The number of passengers using the bus and rail park and ride towards Edinburgh has helped to control congestion levels on busy corridors including the A90. Whilst the current capacity constraints affecting the Forth Bridge effectively limit the number of car drivers crossing from Fife, the proposed new crossing could release some suppressed demand, though it should be noted that the plans include a substantial investment in public transport including bus only lanes which will enhance the attractiveness of public transport across the Firth of Forth. Some existing car drivers parking on the Fife side of the bridge may drive closer to their final destination, particularly if these capacity constraints are alleviated. As noted earlier, there are also a number of other public transport options serving the travel market west of Edinburgh. This may also influence the overall decision making process. This does represent a risk that some motorists could drive further towards their ultimate destination, hence increasing total car kilometres
- objective 5 (financial metrics to guide appraisal): most of the existing bus services calling at Ferrytoll operate as part of a longer distance route. Analysis of the load factors using these services was outside this commission, so further work is needed to determine whether there is a business case to provide additional seats on these bus routes, or whether future growth can be accommodated using the existing vehicles. Similarly, the business case for rail service improvements needs to be linked to other proposals. The procurement of additional rolling stock needed to support future growth from Inverkeithing, especially during the peak period may need to be linked to service improvements elsewhere in Fife to produce a robust business case
- objective 6 (optimum pricing policy): Park and ride users appear to have a lower value of time compared with typical values. This would imply time savings are not the main factor that encourages park and ride, and indeed it is doubtful the rail and bus alternatives offer a quicker door-to-door journey time. However, these results are consistent with relatively high cost sensitivities, implying park and ride is used to avoid high parking charges. The results indicate frequency is not an important factor in the choice of whether to use park and ride, although this conclusion is dependent on a reasonable service frequency being available. Optimal fare calculations are not relevant to this section