|
|
|
|
|
          
|
Personalised journey planning
Evidence on performance Travel Blending® Travel Smart® Indimark® in Germany TravelSmart in Bristol, UK TravelSmart in Gloucester, UK
Context “The Travel Blending® Program was initially developed as part of a major public initiative called ‘Clean Air 2000’ which aims to reduce pollution caused by car travel in Sydney prior to the year 2000 Olympics. …After the pilot study had been completed in Sydney, the Department of Transport in South Australia (TransportSA) initiated a trial which took place in Adelaide” (Rose and Ampt, 2001). Travel Blending® in Adelaide is known as ‘TravelSmart’ Adelaide. As a PJP programme, Travel Blending® emphasises the “How to rather than … [the] Should do” (Rose and Ampt, 2001). Travel Blending® consists of two one week travel diaries completed by all members of participating households. Individual participants were recruited through the workplace; the individual then co-opted the rest of their household. The first travel diary allowed:
The second diary:
Travel diaries recorded “all travel outside the home with details obtained of destination, place and purpose, start and end time of each trip, travel mode and for car driver trips, the odometer reading at the start and end of the trip” (Rose and Ampt, 2001). The diaries covered seven days as week day and weekend journeys can be very different; people may be more able to travel blend at the weekend than during the week, or vice versa. It was found that people did complete the full seven day diaries; possibly because they included a built in reminder system (Rose and Ampt, 2001). Travel Blending® does not merely promote replacing motor vehicle travel with other modes or means of communication, it encourages “thinking about activities and travel in advance (i.e. in what order can activities be done, who should do them, where should they be done etc.), and then blending modes (i.e. sometimes car, sometimes walk, sometimes public transport etc.), or blending activities (i.e. doing as many things as possible in the same place, or on the same journey [i.e. trip chaining]), or finally blending over time (i.e. making small sustainable changes over time on a weekly or fortnightly basis)” (Rose and Ampt, 2001). The key message is “to blend travel choices in a manageable but sustainable way to reduce motor vehicle use … [whilst] allowing people to participate in the same activities that they currently undertake” (Rose and Ampt, 2001). Further information regarding the detailed design of Travel Blending® can be found in Rose and Ampt (2001). Impacts on demandRose and Ampt (2001) report details of the Sydney pilot study in qualitative terms due to the small sample size, and the Adelaide study in quantitative terms. Sydney
The above households changed their travel patterns as a result of Travel Blending®. Two others made fairly dramatic changes because one of their vehicles was off the road. Other participants had plans to change in the longer term, including:
Adelaide The Table A1 below indicates the changes in car use as a result of Travel Blending®. The results of a Z test to test the hypothesis that the means are equal for diary one and diary two, against the alternative hypothesis that the mean for diary two is less than that for diary one are also included. Table A1 Travel Behaviour Change Amongst Adelaide Travel Blending® Participants
Source: Rose and Ampt (2001) Rose and Ampt (2001) produced aggregate results for the population as a whole (see Table A2) by including non-participants in the analysis. This was done by assuming “that each person who refused to participate in rounds one and two travelled in the same way as the average for all persons in diary one (i.e. before they had received feedback) in both rounds, and that any person who participated in dairy one and not diary two was assumed to have made no change between the two diaries” (Rose and Ampt, 2001). Table A2 Estimates of Aggregate Reductions in Car Use
Impacts on supply Other impacts
Contribution to objectives
Case Study B: Travel Smart® The Travel Smart® PJP was developed by SocialData under the name Indimark®. Travel Smart® PJP starts by identifying individuals who are prepared to think about reducing their car use through telephone surveys. Those completely resistant to the idea do not receive any further communication. Those who already use alternatives a lot receive some form of reward, which is found to increase use of alternative modes further. Those who are prepared to think about reducing their car use and participate provide information about their journeys and receive targeted suggestions to reduce their car use. This is done through the post or a home visit where appropriate (Brög and Schädler, 1999). A pilot study was undertaken in South Perth in 1997, with approximately 400 randomly selected households. The pilot comprised a benchmark survey in August 1997, intervention in September/October 1997 and an evaluation survey in November 1997. A second and third evaluation survey was undertaken in September 1998 and February 2000 respectively. A large scale application also occurred between February and June 2000, but that is not reported in detail here. Monitoring of the large scale application was also undertaken using the electronic bus ticketing system in the area subject to PJP. Impacts on demand The percentage changes in travel behaviour resulting from the pilot study are presented in Table B1. Table B1 Percentages Travel Behaviour Change from Travel Smart®
*No figure avilable. The fare box monitoring undertaken with the large scale application revealed a 27% increase in bus patronage between the period March to June 1999 and the same period in 2000. Over the wider network, there was a 1.5% increase in patronage, thus the net increase of 25% was attributed to Travel Smart® PJP. Impacts on supply The implementation of PJP through Travel Smart® did not affect the supply of road space or public transport infrastructure. It is possible that supply of public transport services may increase in response to demand. Other impacts The Travel Smart® analysis notes a number of cross cutting benefits. Many of these fall within the contribution to objectives below, but additionally, there are preventative health outcome due to increased levels of physical activity. Contribution to objectives
Case Study C: Indimark® in Germany ContextBrög and Schädler (1999) report Indimark® in Europe, which takes the same form as that developed for Travel Smart® in Australia, described above. Following the Australian success, the ‘Switching to Public Transport’ demonstration project was initiated in Europe (in Germany) by the International Association of Public Transport (UITP). Indimark® was in this case applied with the specific aim of increasing public transport patronage, and has since been adopted by a number of operators as part of their marketing strategy. Many applications in Europe have included upwards of 2000 participants, one the largest being in Leipzig, where there were 75,452 participants. Impact on demandLike the Travel Smart® results, those in Germany (see Table C1) indicate that changes are sustained over two years. Even where there is a slight reversal in changes in travel behaviour, there is still less car use and more public transport use two years on than before intervention. Table C1 Changes in modal share following introduction of Indimark
B – before Indimark; A – immediately after Indimark, 1yr – one year after Indimark Source: (Brög and Schädler, 1999). Impact on supply Again, implementation of PJP does not change the supply of road space or public transport. However, there may be increases in public transport supply resulting from increases in demand, especially since the Indimark® reported here was implemented to increase patronage. Other impacts Indimark® resulted in an improved image of public transport amongst the target group. Perceptions also improved amongst a control group, but by a much smaller degree.
Source: (Brög and Schädler, 1999). Participants who received a 'test ticket' (a free ticket that could be used for trial public transport use) reported improved perceptions of public transport and increased intentions to use public transport. However, a trial of test tickets in one German city, without prior contact and dialogue indicates that experience alone does not change behaviour. The dialogue is essential.  Source: (Brög and Schädler, 1999). Additionally, in an Austrian application of IndiMark®, one city public transport operator sent out conventional, untargeted information packages to a group, in addition to the Indimark® group. The results suggested that conventional information has little impact on public transport patronage, but that Indimark® does have an impact. [For “follow” read “following” in this diagram.]
*Car driver and motorcycle journeys Contribution to objectives
Case Study D: TravelSmart in Bristol, UKThe case study reported here is taken from Anable J, Kirkbride A, Sloman L, Newson C, Cairns S and Goodwin P (2004) ‘Smarter Choices Volume 2: Case study reports’, Department for Transport, London. ContextTravelSmart using the SocialData IndiMark methodology has been implemented in two locations in Bristol, a medium sized city in south west England. The locations were Bishopsworth/Hartcliffe (approximately 5km south of the city centre), and Bishopston (approximately 2.5km north of the city centre). Bishopsworth/Hartcliffe has a high deprivation index, with low car ownership and comparatively high use of public transport. In contrast to this, Bishopston has relatively high levels of car ownership, resulting in a high number of short car trips. Both applications of TravelSmart involved Bristol City Council, Sustrans and SocialData. For the Bishopsworth/Hartcliffe scheme First Bus also provided information and trial tickets. TravelSmart (or IndiMark as it is referred to in some parts of Europe) includes a number of distinct phases: “initial contact is made with households which are then segmented into groups according to their use of different transport modes and whether they are interested in making more use of alternative travel modes. Those who would like to use public transport more, or walk or cycle more, are provided with personalised information about transport alternatives. Some of these households receive further support (the “convincing” phase) to encourage walking, cycling or to use public” (Anable et al, 2004). The initial contact stage in both locations included a target to approach a minimum of 5,000 people in each area. In Bishopsworth/Hartcliffe however, TravelSmart was applied in two rounds, with 2,500 people approached initially, then another 2,500 a year later. This approach was used to allow for “implementation of the 76/77 service showcase bus route, which passes through the area” (Anable et al, 2004). This was done to compare results of TravelSmart without improved public transport with TravelSmart plus improved public transport. The first round of the Bishopsworth/Hartcliffe scheme started in September/October 2002. 1,192 households (comprising 2,500 individuals) were targeted, of which 1,081 were approached (the rest were uncontactable). 867 households responded, “of which 46% [399 households] expressed an interest in receiving information on alternative travel modes” (Anable et al, 2004). Of these 399 households, 284 received information and further advice/support (52 of whom only received rewards for already using alternatives to the car as much as possible). It is not clear what happened to the other 115 households who had expressed an interest. At the “convincing” phase a range of incentives were provided to participants including:
Round two of the Bishopsworth/Hartcliffe scheme started in September/October 2003, but response rates were not available to Anable et al at the time of writing. Costs Impact on demandChanges in mode choice were monitored through household and individual travel surveys [most likely diaries] sent to a sample of the target group before and after intervention. A control group not contacted as part of the marketing intervention were also surveyed. Anable et al note that whilst round 1 of the Bishopsworth/Hartcliffe scheme was not affected by the implementation of the 76/77 service showcase bus route, which passes through the area, it was influenced by other wider public transport improvements in the general area that coincided with it. Results for round 1 of the Bishopsworth/Hartcliffe scheme (see Tables D1 and D2) are reported in Anable et al, 2004. Table D1 Mode shares before and after round 1 of Bishopsworth/Hartcliffe TravelSmart
Control group only benefited from public transport improvements. Table D2 Relative change in average number of trips per person per year from round 1 of Bishopsworth/Hartcliffe TravelSmart
The increases in public transport use and reductions in car passenger travel and walking in the control area are thought to be largely the result of public transport improvements. It is thought that the public transport improvements will have influenced the TravelSmart results, particularly the apparent modal shift to public transport apparent in the control group. No evidence is available on the scale of the influence public transport improvements had. Preliminary results for the Bishopston TravelSmart (see Table D3) are reported in Anable et al (2004) as follows. Table D3 Bishopston preliminary results
Anable et al (2004) note that in addition to the public transport improvements that affect both of the TravelSmart areas, a variety of Smarter Choices measures were implemented throughout Bristol, which will have generated a positive synergy with TravelSmart. No evidence on the impact of these Smarter Choices measures is available. Impact on supplyTravelSmart itself will not have changed supply of road space or any individual modes. However, simultaneous public transport improvements and Smarter Choices measures will have improved the quality of supply of public transport if not the quantity, and will have increased the alternatives to conventional car use available to residents of Bristol. Other impactsIncreases in walking and cycling resulting from TravelSmart will have positive health benefits for individuals and could if achieved on a large scale and sustained over time result in reduced spending on health services. Given the range of transport improvements being made in Bristol, wider community benefits may also accrue. Contribution to objectives
Case Study E: TravelSmart in Gloucester, UKThe case study reported here is taken from Anable J, Kirkbride A, Sloman L, Newson C, Cairns S and Goodwin P (2004) ‘Smarter Choices Volume 2: Case study reports’, Department for Transport, London. ContextTravelSmart has been implemented in Quedgley, a suburb four miles south of the centre of Gloucester, in the UK. Quedgley was chosen over other areas of Gloucester as “it has good local facilities (including primary and secondary schools, a library, and a supermarket), and a good bus service into Gloucester city centre (every 15 minutes)” (Anable et al, 2004). Further, the distance from the city centre made this journey a cycleable length. Combined with the fact that Quedgley has higher than average car use and peak hour congestion on main roads, it was felt that there was real potential for modal shift, especially for short journeys. Other factors that favoured Quedgley as an initial trial area in Gloucester were the fact that bus services were not full to capacity as they were in other locations around the city, car use was relatively unconstrained due to proximity to a motorway junction, and lack of on-street parking problems and general congestion made car use easier than other in other areas of the city which did experience these problems. Given the ease of using a car it was felt that if TravelSmart could work in Quedgley, then it could work elsewhere provided local destinations and real alternatives to the car were available. In Quedgley, TravelSmart was implemented in two phases, a pilot with 515 people (running April 2001 to May 2002), and a large-scale roll out involving 10,000 (representing all 4,631 households in Quedgley). The project was managed by Sustrans, with the work being carried out by SocialData using their IndiMark methodology. In addition to the local authorities (Gloucestershire County Council and Gloucester City Council) who commissioned the work, Sustrans and SocialData, a number of other organisations were also involved with TravelSmart in Quedgley: Stagecoach and Swanbrook Transport (both public transport operators), Quedgley Parish Council, Vision 21 (the Local Agenda 21 forum), local cycle retailers and outdoor pursuits shops. Pilot
These categorisations are the standard groupings allocated in IndiMark/TravelSmart. Individuals in the ‘R with’ and ‘I’ groups (217 people) received a “Service Sheet” listing information which they could choose to receive, which they were required to return to the marketing team. 187 (80 households) people returned their “Service Sheet”. Information was then hand-delivered to households in personalised packages. Those in the ‘R’ groups (‘R without’ and ‘R with’) received a gift by way of a thank you for already using environmentally friendly modes. Large-scale roll out The households not contacted by phone or post (approximately three quarters of the 4,631 initially targeted) were then approached by knocking on doors. For these households the initial contact, Service Sheet and information provision stages were combined into one, as households selected the information they would like there and then, and this was provided on the spot from a supply the marketing teams carried with them. “Of those people at home when the call was made, 90% were interested in receiving information materials – a higher take-up rate than normally expected. Once people could see what was available, they were keen to receive information” (Anable et al, 2004). Ultimately, of the 4,631 households targeted, 271 were found to be uncontactable due to occupants moving away or having deceased, giving a possible 4,360 contacts. Of these, 93% (4,069) were contacted – a contact rate comparable with other TravelSmart applications. The contacted households fell into the following categories:
In total 2,120 households received information or rewards for already using environmentally friendly modes; 102 (‘R without’) received rewards only. The information requested most often was bus-stop specific timetables (requested by approximately 70% of households). Walking and cycling information was requested by approximately 50% of households. 977 households requested discount cards (it is not clear what these were for, but they were most likely to be for local cycle and outdoor shops selling goods relating to walking and cycling) and/or a home visit to provide personal advice), and 89 received home visits; 56 related to public transport, 20 related to cycling and 13 related to walking. Costs and funding The budget for the pilot project was £30,000, comprising £12,000 for the marketing campaign and £18,000 for before and after monitoring surveys. This represented a comparatively high £58.25 per person approached (based on the 515 people approached) when compared with the costs of the large-scale roll out, highlighting the economies of scale achieved through larger schemes. The budget for the large-scale roll out was £168,600, comprising:
This £168,600 did not include the cost of staff time within the local authorities (for the pilot this was estimated to be £3,000). The cost per person approached was £17 (based on all 10,000 targeted). Sustrans have estimated that a project targeting 30,000 would cost £30 per household, or £13 per person based on an average household size of 2.3 people. These figures cover the costs of marketing, a before and two after surveys, and promotional materials. The figures do not cover the costs of information materials (as existing materials are usually utilised), or test tickets and public transport related home visits, which are normally provided in kind by the public transport operator. Impact on demandData and key pieces of text directly from Anable et al, 2004. “Evidence of the effect of the pilot project on car use is based on a ‘before’ survey carried out in September 2001 and an ‘after’ survey in January / February 2002. For both before and after monitoring, a control group from another part of Quedgley was surveyed as well as all those people involved in the marketing exercise. The net ‘before’ survey sample was 871 people (a response rate of 66%); and the net ‘after’ survey sample was 624 people (a response rate of 76%)” (Anable et al, 2004). Results from the pilot are presented Table E1. Table E1 Mode share before and after for target and control groups
The data reported in Table E1 has not been adjusted or weighted. Further analysis of the results are reported in Anable et al (2004), including an adjustment of the target group before data to allow for the fact that changes seen in the control group would most likely affect the target group as well. The after data for the target and control groups is also weighted by trip purpose such that distribution of trip purposes is the same in the before and after data. The after target group data is further weighted to correct for the proportion of ‘I’, ‘R’ and ‘N’ respondents, such that it is the same as observed during the marketing campaign. Such analysis of Travel Smart data is standard practice, but in this case, made little difference to the results. “Car mileage is also reported to have fallen by 9%, from 21 km per person per day to19 km per person per day. Again, this figure is averaged across the entire target population: that is, it represents the behaviour change for both people who responded and those who did not respond, and for those who requested information and materials as well as those who did not. Changes in car use were seen across all times of day (both peak and off-peak). Much of the increase in bus use was off-peak, when capacity was already available. There was no information about impacts on weekdays as compared to weekends. The number of activities per day remained the same, and so did the number of trips per day. This suggests that people did not respond to the project by trip consolidation (combining different trip purposes into one journey). Car use was affected for all journey purposes apart from education: that is, car use For the large-scale roll out only preliminary results (see Table E2) were available to Anable et al (2004) as follows. Table E2 Large-scale roll out preliminary results
Impact on supplyTravelSmart itself will not have changed supply of road space or any individual modes. However, increased demand for public transport could result in increased supply in the long term. Other impactsIncreases in walking and cycling resulting from TravelSmart will have positive health benefits for individuals and could if achieved on a large scale and sustained over time result in reduced spending on health services. Contribution to objectives
The primary weakness in the evidence presented stems from PJP being a relatively recent development. Consequently, there is a scarcity of evidence. This is exacerbated by some reporting of PJP lacking transparency and consistency. The UK Department for Transport (DfT, 2009) cite this lack of robust evidence for recent caution in forecasts of the impacts (in terms of carbon emission reductions) that may result from Smarter Choices (which includes PJP).
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Context
