 |  |  |
South East Dorset Transport Study
Phase 2 - Model Build, including data collection and preliminary testing.
A detailed traffic model for South East Dorset has been in existence for over thirty years. The model has been used to forecast the traffic effects of major land-use changes and support the case for associated transport improvements e.g. Twin Sails bridge. The current version of the model was jointly funded by the four highway authorities, but is now well out of date and is insufficiently robust for future transport planning work.
Phase 2 requires data collection and the construction of a multi-modal model for the study area , capable of both assessing the impact of various development and transport strategy scenarios, and providing a technical base for the future preparation of Major Scheme Bid business cases.
There are 2 key elements to Phase 2:
(i) Data Collection
To develop a comprehensive transport model for the Poole/ Bournemouth conurbation and surrounding settlements requires a significant programme of data collection surveys. Further details can be found by following the links below.
- Roadside Interview and Traffic Count Surveys
- Car Park and Other Interview Surveys
- Public Transport Surveys
- Journey Time Surveys
Roadside Interview and Traffic Count Surveys
As part of this study an extensive traffic survey programme will take place during the spring and autumn of 2008, involving around 60 roadside interview surveys on many roads within South East Dorset and west Hampshire.
Dorset Engineering Consultancy, with uniformed Police officers in attendance, will conduct these surveys at key locations on the highway network. This will involve stopping and interviewing a sample of drivers over 12 hour periods for information on their trip Origin Destinations, journey purpose, type of vehicle, vehicle occupancy and basic demographic information. The surveys are to be undertaken in groups across a ‘screenline’ to capture all vehicular trips from one area to another. The roadside interview surveys need to be undertaken in the spring and autumn 2008, to avoid major road maintenance works already planned for other times of the year.
The surveys themselves will obviously delay those interviewed, and cause some local traffic tailbacks and delays. This is especially true for high-speed roads, particularly during busy periods, when there is potential for significant queuing as a result of the surveys and associated traffic management in place. Experienced survey staff and the Police will monitor any congestion, with arrangements in place to release traffic through the survey site in the event of excessive queuing. It is vital to collect as true a picture of transport movements as is possible on survey days, and we seek the publics’ co-operation in gathering the data by providing accurate answers and returning questionnaires (where this method is used).
Car Park and Other Interview Surveys
The proposed RSI surveys would intercept strategic traffic movements within the study area, but would not cover shorter distance intra-urban traffic movements within Bournemouth, Poole and Christchurch. It is recognised that additional interview surveys should be undertaken at selected town centre car parks. 4000 interviews should provide a 15% sample of drivers using these town centre sites. Similar surveys in adjacent towns such as Ringwood, Ferndown, Verwood, Wimborne Minster, and Wareham, together with interviews at selected supermarkets and shopping centres would potentially increase the total number of interviews required to about 6000.
Similarly, interviews or self-completion questionnaire surveys should be conducted at selected employment sites, to supplement Origin~Destination data collected on cordons and screenlines and at long stay commuter car parks. With about 150,000 work places in Bournemouth, Poole and Christchurch, surveys of 7500 workers would provide a 5% sample of employees. Selected employers could also be asked to provide home postcodes for their employees.
Public Transport Surveys
Interview surveys at railway stations and selected bus stops are required to provide information on passenger origins and destinations, supplementing ticketing data which would only record movements between stations and fare zones.
For calibration and validation of the Public Transport model, bus passenger counts on selected screenlines, and rail passenger counts at stations will be required. These surveys would be determined in detail following a review of service patterns and frequencies, and passenger movements.
Journey Time Surveys
Journey time surveys are required on selected roads for input to the model calibration and validation process. These surveys would cover the modelled time periods using the GPS tracking equipment on a moving observer, with at least 6 runs in each direction during each time period.
(ii) Model Development
A recent review of the existing SATURN highway model drew attention to the age of the input Origin~Destination data and, the fact that it did not include a public transport model [i.e. was not multi – modal]. The model also criticised for its failure to meet the validation criteria set down in the Design Manual for Roads and Bridges (DMRB), Volume 12. Other deficiencies of the current model relate to its forecasting capability.
The four highway authorities with responsibilities in the SED area (Bournemouth, Dorset, Poole and the Highways Agency [HA]) are aware of the existing SATURN model’s deficiencies. They have agreed that a new 2008 model should be built with the following key characteristics.
- The model would need to cover in detail the whole of the SE Dorset Joint Study Area (JSA) as defined for the Regional Spatial Strategy (RSS), which is roughly equivalent to the joint Local Transport Plan area. This includes the main urban areas of Poole, Bournemouth and Christchurch, as well as the outlying towns of Wareham, Wimborne Minster, Ferndown, West Moors, Verwood, and St Ives. Data from the 2001 Census Journey to Work sample suggests that about 89% of employment places in SE Dorset and about 82% of employment places in the conurbation are taken by residents of these areas. A model covering this area would therefore include the majority of commuting trips to the sub-region. However, Ringwood, Bransgore and other adjacent Hampshire towns would also need to be included within the main study area, given their proximity to SE Dorset.
- The model would need to be multi-modal, including passenger movements by motorised modes, such as car, bus, rail and park and ride (P&R). We do not envisage that it would include trips by slow mode (walk and cycle), other than is necessary to model transfers between PT services, at P&R sites, or at the beginning and end of journeys, for access. However, the potential transfer of short car trips to walking and cycling and the effect of this on congestion would need to be considered as part of the sub-regional transport strategy study.
- The transport model would need to be responsive to future changes in the generalised costs of travel, including road pricing on key routes, congestion charging in the urban areas, parking charges, journey time and vehicle operating costs for motorists; and fares, in-vehicle time, walking time, waiting time and transfer penalties for public transport users.
- The transport model would need to forecast the impacts of future land-use developments on travel patterns and traffic flows in the study area, in order to facilitate strategic transport assessments of LDF (2016) and RSS (2026) housing and employment allocations, and more localised traffic impact assessments of specific development proposals.
- The SE Dorset model should be sufficiently robust at the strategic level to allow for the testing, assessment and prioritisation of a wide range of transport policy and scheme options for the study area as a whole. It also needs to be sufficiently robust at the local level to facilitate more detailed assessments of individual schemes and measures by the promoting authorities. It may be necessary for certain schemes and/or development proposals (as yet unspecified) to develop more detailed traffic models or micro-simulation models of specific areas, or extend existing microsimulation models.
- The model would be required to represent average weekday conditions (by time period) in a neutral month (April-June and/or September-October). A model of the AM peak hour or period would be necessary to cover the worst case for commuting, and to facilitate the forecasting of P&R and congestion charging effects, decisions about which are frequently taken by drivers (particularly commuters) in the inbound direction. Given the need for economic assessment of options, an inter-peak model would also be required. A PM peak hour model would also be required to provide robust data for assessing the traffic impacts of proposed developments and future transport strategies, and for consistency with the existing SEDTM.
 |  |  |  |