Introduction

Background

TransPosition was engaged by the Department of State Development, Infrastructure and Planning (DSDIP) to provide option testing analysis for the Queens Wharf Brisbane Project using Brisbane City Council's (BCC) SATURN model. This short report outlines the work done by TransPosition on the SATURN model since the handover from BCC including the major assumptions, limitations and disclaimers that come with the model that will be given to proponents.

Disclaimer from Brisbane City Council

BCC are the prime custodian of the model and have made it available for this project. The use of the model was subject to a Model Release Agreement. A section from that agreement is shown below.

  1. Warranty

(a) Council gives no warranties and makes no representation as to the condition, completeness, quality, suitability or fitness of the Model Product(s) for the Recipient’s requirements or that the Model Product(s) is provided in a form free of errors. (b) The Recipient is responsible for ensuring that the Model Product(s) meets its own individual requirements. (c) The Recipient acknowledges that the Model Product(s) does not necessarily represent Council policy.

  1. Indemnity

(a) The Recipient acknowledges and accepts that it shall use and rely upon the Model Product(s) at its own risk and agrees that Council shall not be liable for any loss, liability, expense, costs or damages due to or arising out of the use of or reliance upon the Model Product(s), directly or indirectly. (b) The Recipient shall at all times indemnify and hold harmless Council, its officers, employees and consultants from and against any loss, damages, expenses, costs or liability incurred or suffered by the Recipient including any of their officers, employees, contractors, agents, consultants or third parties, whether in negligence or otherwise.

BCC officers have acknowledged the limitations to the model and their intent to perform a significant update to the model in the near future. BCC's Cameron Sims has provided the following advice (in an email to TransPosition on the 7th of May, 2014). He refers to the license agreement that TransPosition entered with BCC in order to obtain access to the model.

The purpose of the agreement is to make it clear that we do not guarantee the accuracy of the model, and that you will need to ensure that the model is fit for purpose for your project. This model was last fully calibrated in 2009, and has been used in an ad-hoc manner for various projects since then. Some areas of the model have been updated for particular projects, and other areas of the model have not been updated where there have been no projects or investigations since 2009. So as a result the model is imbalanced and in need of a major update which has commenced but won't be completed until next financial year.

We are not able to provide guidance on the use of SATURN Software. We expect that users of the models we provide through model release agreements are proficient users of the software. But we can provide some clarification of specific issues and assumptions that exist in the model. The model was principally developed by others in different circumstances before my involvement also so there are many issues that I might not agree with or might not be relevant now. There are many changes that we propose to make in the next major update to the model.

TransPosition's Changes to the Model

Network update

Before using the model, TransPosition checked the road network for the Queens Wharf precinct which covers the area from Queens Wharf Rd to Edward St, and Alice St to Elizabeth St. This included site inspections and the use of aerial photography (through the Queensland Globe).

We also checked the road network/intersections on George St at Queen St and Adelaide St; North Quay at Adelaide St; and also the Victoria Bridge/Queen St Bus Terminal intersection.

We did not review the zone centroids in the model, in part because we wanted to maintain consistency with BCC's matrices and zoning system. A small number of centroid connectors were moved to better allow some network options to be tested. In particular, the centroid connectors on Victoria Bridge were moved off the bridge and placed onto either side of the bridge.

Review of signal phases and times

Using data provided by the BMTMC, TransPosition updated the base-case signal phases and green times for the following intersections:

  • Adelaide St at George St
  • Queens St, William ST, North Quay, Queens Wharf Rd and Victoria Bridge
  • George St at Queen St Ped
  • George St at Elizabeth St
  • George St at Alice St
  • George St at Margaret St
  • George St at Charlotte St
  • George St at Mary St
  • North Quay at Adelaide St
  • William St Ped at Stephens Lane
  • William St at Elizabeth St and REX off ramp
  • William St at Margaret St and REX off ramp
  • Alice St at William St

Future developments

Based on discussions and advice from Cambray Consulting, some anticipated future developments have been incorporated into the model by the way of car parks numbers. The major changes were the following:

2013

  • 1 William St is a construction site for the base 2013 model and so the car parks spaces were decreased to zero.
  • QUT has an extra 100 spaces.
  • Block bounded by Herschel St, North Quay and George St has an extra 248 spaces.

2021 & 2031

  • Includes points 2 and 3 in 2013 changes
  • 1 William St is projected to have a total of 318 car spaces.
  • 300 George St is projected to have a total of 484 car spaces.
  • 480 Queens St is projected to have an extra 271 car spaces.
  • Block bounded by Margaret, Mary and Albert St is projected to have an extra 980 spaces.
  • Two developments on Albert St between Alice and Margaret St will add an extra 477 car spaces to that block.

The Saturn model requires peak hour demand at each car-park, for the AM and PM peaks. We had no direct survey data to use to convert the parking space estimates shown above into hourly demand in and out of each car park. We have adopted the approach used by BCC for other car parks; using the Myer Centre as a basis for estimating the daily demand profile. The demand has been estimated on a pro-rata basis using the Myer Centre car park numbers. It might be expected that some of these car parks will have different demand profiles, as they will have a different mix of residential, commercial and retail use. But we were unable to produce more accurate estimates without more detailed information on the likely structure of the new developments, or surveys of daily car park demand.

The impact of the future development has been incorporated into the model by applying a Fratar process to BCC's future year demand matrices. This can lead to an inconsistency, since the future matrices are driven by demand estimates from the Brisbane Strategic Transport Model (BSTM), whereas the car parking numbers show supply constraints. To reduce possible underestimating of demand, any parking-based numbers which were previously higher in the future year demand matrices were kept. This ensures that the future development described above always lead to increase demand numbers in the model.

Option coding

All of the options that were to be coded were given to TransPosition by Cambray Consulting, following their discussions with DSDIP. The description of these options, and the modelling outcomes, are detailed in Cambray Consulting's report.

Limitations and Disclaimers

Apart from the intersections and the road network area discussed above, we have not checked the wider Saturn network. TransPosition was given only the model itself, with no report from BCC about the original inputs into the model, or background data used to derive the model inputs. Because of this, there are aspects to the model that we do not fully understand, and processes that we cannot replicate. In particular, the future year matrices (which we understand were based on inputs from the BSTM) have been provided to us "as-is", and so we have not been able to check their validity.

We have reviewed the traffic counts in the Queens Wharf precinct, based on analysis of data from the BMTMC. Given the concerns mentioned above; and due to time constraints, we were reluctant to repeat the matrix estimation process. Instead we checked the counts against those used by BCC in their last matrix estimation process and found that there were no significant differences. Thus we have used the forecast matrices provided by BCC with only minor changes (due to the Fratar adjustment at the car parks listed above).

The bus routes and frequencies through the city are used to provide an estimate of the bus demand moving through the network. We have not done any systematic update to the data that was included in the model provided to us. There were a number of services that needed to be rerouted for some network options, and these services were reviewed and compared with the model inputs. We found that there was reasonably good agreement, but noted that the bus loads seemed to include only scheduled services, and appeared to make no allowance for dead running.

The detailed coding of the network options has been guided by Cambray Consulting, and all tuning of phase diagrams and timings has been done in consultation with them.

In summary, TransPosition has relied on data and assumptions obtained from BCC, the Brisbane Metropolitan Transport Management Centre (BMTMC), Cambray Consulting and DSDIP. Although some data has been reviewed, given the range and complexity of this data it has not been possible to undertake a full audit of the model for data quality and reliability.

All models make simplifications of the present and assumptions about the future. Inevitably some of the assumptions will not be realised, and other unanticipated events and circumstances may occur. TransPosition cannot provide any assurance that the base year outputs from the SATURN model are correct or that the outputs of the model will be achieved.

All information and outputs prepared by TransPosition are provided on a non-reliance basis.