Transport Assessment – Issue 2.1 Northampton South SUE J:\28015 Collingtree\word\Reports\Transport\Transport Assessment\28015 Transport Assessment 140430 - Issue 2.1 - Planning Application.docx 160 Appendix 12.1 - Northamptonshire Strategic Transport Model Local Model Validation Report and correspondence
281
Embed
Appendix 12.1 - Northamptonshire Strategic Transport Model Local ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Transport Assessment – Issue 2.1 Northampton South SUE
Appendix 12.1 - Northamptonshire Strategic Transport Model Local Model Validation Report and correspondence
Northamptonshire Strategic Transport Model Local Model Validation Report
- November 2011 Update Northamptonshire Highways
February 2012
Northampton Riverside House Riverside Way Northampton NN1 5NX Tel: +44 (0)1604 654659 Fax: +44 (0)1604 654699 www.mgwsp.co.uk May Gurney Ltd, Registered Office: Holland Court, The Close, Norwich, NR1 4DY Registered Number 00873179 England WSP Management Services Limited Registered Office: WSP House, 70 Chancery Lane, London WC2A 1AF Registered Number 02454665 England
1.1.1 As part of the sustainable transport highway services contract for Northamptonshire County Council (NCC) the MGWSP joint venture was commissioned by NCC to combine and update the two existing strategic transportation models:
North Northamptonshire Highway Model (SATURN)
West Northamptonshire Multi-Modal Model (SATURN and EMME/2)
1.1.2 Under this commission the existing models have been combined and updated to produce a single County Wide Strategic Transport Model. The SATURN (Simulation and Assignment of Traffic in Urban Road Networks) suite of programs was used to build the observed highway model while EMME3 has been used to build the Demand Model and the Public Transport Model. The strategic model developed by WSP offers an integrated system for a range of transport modes, representing private and public transport as well as modal interchange behaviour such as walking. The model will have the ability to quantify the benefits of a policy change or new transport infrastructure in Northamptonshire, enabling the testing of future transport proposals and developments (including developer led) in an efficient, consistent and evidentially based manner.
1.1.3 The model specification was developed in accordance with current Department for Transport guidelines detailed in the Web based Transport Analysis Guidance (WebTAG), which is aimed at ensuring a consistent and reputable basis for forecasting. The highway model was developed in line with Design Manual for Roads and Bridges (DMRB) guidance.
1.1.4 This involved a multi-modal data collection exercise with the end result being a fully calibrated and validated Multi-Modal Transport Model of Northamptonshire County and the surrounding areas, hereinafter referred to as the Northamptonshire Strategic Transport Model (NSTM).
1.1.5 The purpose of the NSTM is to inform the assessment of the Local Development Framework (LDF) Core Strategy, and will subsequently be used to assess individual development proposals and highway infrastructure schemes.
2 NSTM - LMVR Highway Model Version 1.0
1.2 BACKGROUND
1.2.1 Northamptonshire’s location near the geographic centre of England has long made it an important focus for the nation’s transport system. In the west of the county, between the villages of Norton and Whilton, four historic transport corridors run in close proximity:
The A5, originally a Roman road linking London and Chester, known to the Saxons as Watling Street and later a major turnpike between London and Holyhead
The Grand Union Canal opened on 21 June 1796 as part of the Grand Junction linking Braunston in Northamptonshire with the River Thames at Brentford
The West Coast Main line railway, opened in 1838 by the London & Birmingham Railway as part of the first line linking London with the Midlands and the north
The M1, the first substantial piece of motorway opened in England on 2 November 1959
1.2.2 The national focus on town planning and urban redevelopment following World War 2, led to a number of towns in the county being designated as new or expanding towns:
Corby, designated as a new town in 1950, largely to accommodate the growing steel industry
Daventry, designated as an expanding town in 1962 to cope with overspill from Birmingham
Wellingborough, designated as an expanding town in 1964, to cope with overspill from London
Northampton, designated as a new town on 14 February 1968, to cope with overspill from London
1.2.3 In addition, the designation of the new city of Milton Keynes, which borders the county, in 1969, has had a major impact on areas of Northamptonshire. Most of the county remains rural in nature; although only a relatively small percentage of workers, even in the rural areas, now work in agriculture. Many of the people who live in the county’s villages, commute to work either in the larger towns within or just outside the county, or to London. Figure 1.1 is a key diagram showing the strategic transport network within the county and the major towns.
NSTM - LMVR Highway Model Version 1.0 3
Figure 1.1: Strategic Transport Network within Northamptonshire
Source: Northamptonshire County Council Local Transport Plan 2006 - 2011
1.3 MODEL REQUIREMENTS
1.3.1 The NSTM is required for wider strategic testing, localised testing, plan policy evaluation and scheme appraisal within Northamptonshire. This must be done in a fair and transparent manner to ensure that the impacts of all proposals can be compared honestly. Impacts to the transport network from individual developments must also be equally assessed in order to fairly calculate contributions required from developers. For this reason it was necessary to obtain a particularly well validated model of Northamptonshire.
4 NSTM - LMVR Highway Model Version 1.0
1.3.2 A Steering Group (SG) comprising technically able members from WSP, MGWSP, NCC, AECOM (as advisors to the HA) and the HA, was set up to agree modelling principles, methodologies, assumptions and scope. This has provided a high degree of control and minimised grounds for disagreement at a later stage in the study process.
1.3.3 The purpose of the Local Model Validation Report (LMVR) is to present the development of the NSTM and demonstrate that it was developed in a robust manner, meets WebTAG and DMRB criteria and is fit for purpose.
1.3.4 The LMVR for the original NSTM highway model was published in June 2010, and was updated in June 2011 to include greater detail in Corby and Towcester. Subsequent use of the NSTM highlighted a number of minor points which, when fixed, had a sufficiently large cumulative effect to require a revalidation. This LMVR sets out the results from the November 2011 revalidation of the NSTM.
1.4 STRUCTURE OF THE REPORT
1.4.1 The structure of this Local Model Validation Report (LMVR) – Highway Model is based on guidance contained in the Design Manual for Roads and Bridges (DMRB) Volume 12, Section 2, Part 1, Appendix B (reference 6). It is as follows:
Chapter 2 of this report gives a summary of the overall structure of the NSTM and the way in which this model has been developed is described
Chapter 3 summarises the development of the highway network
Chapter 4 discusses the development of the observed highway matrices
Chapters 5 and 6 describe the calibration and validation process of the highway assignment model
Chapter 7 concludes the report
NSTM - LMVR Highway Model Version 1.0 5
2 Northamptonshire Strategic Transport Model - Overview
2.1 BACKGROUND
2.1.1 The purpose of the Northamptonshire Strategic Transport Model (NSTM) is to inform the assessment of the Local Development Framework (LDF) Core Strategy, and will subsequently be used to assess individual development proposals and highway infrastructure schemes.
2.1.2 With this in mind, the NSTM has been constructed as a strategic highway model covering Northamptonshire and as a multi-modal transport model capable of assessing the impacts of future changes in journey patterns within Northamptonshire.
2.1.3 This section describes the model approach, the software applied and the data collated for use in developing the NSTM.
2.2 MODELLING APPROACH
2.2.1 The NSTM is designed to enable future land use planning and transport policy initiatives to be assessed in an integrated fashion. This integration aims to take account of:
The impacts of economic and demographic trends and of land use change on the spatial and modal pattern of the demand for passenger transport
The impacts of changes in transport costs and characteristics on the costs and spatial patterns of location of employment and households
2.2.2 The modelling specified is wide ranging, in that it includes all modes of travel and has the ability to assess a wider set of short term and longer term behavioural responses to policy initiatives than is normal in more traditional models.
6 NSTM - LMVR Highway Model Version 1.0
2.2.3 This model represents, at the strategic level, the main long-distance transport routes in the entire study area. There is a detailed representation of traffic and public transport service levels within the NSTM area. As a result the key abilities of the model are:
Assessing the transport impacts of significant developments
Reviewing the impact of development plans on the transport network
Forecasting changes in demand as a result of changes in the socio-economic characteristics
Allowing forecasting of demand by all modes for a period in line with current Local Development Framework timescales
Identification of schemes and measures for LTP3 and considering the impact of improvements on the strategic network
Identifying current and future congestion hotspots and assist in the modelling of the network management scenarios assisting the local transport authority to fulfil its requirements as required by the Traffic Management Act 2004
2.2.4 The model has been developed with due cognisance of WebTAG on model form and development to accommodate major infrastructure improvements and the sensitivity analysis required to achieve central government funding.
2.2.5 The NSTM is built on a behavioural basis that determines the travel demand from the underlying characteristics of the transport supply and characteristics of the travellers in the area. Key features of the model are:
The input of detailed planning / land use assumptions
The generation of trips by all modes of travel for the different segments of the population
The choice of mode of travel
The distribution of the different trip types to the various destinations available
The choice of routes of travel
An accurate representation of the observed base year travel patterns
Realistic representation of the observed base year, congestion and queuing in the AM Peak and PM Peak periods
2.2.6 This list of requirements is a mixture of the features typically found in strategic transport models, focusing primarily on travel demand choices, and the features of a local highway model, representing the local road network in detail incorporating junction delays and queuing.
NSTM - LMVR Highway Model Version 1.0 7
2.2.7 Local highway models typically use highway demand matrices derived from survey data which can be calibrated to accurately match traffic counts in the base year. However, the lack of any behavioural basis in the derivation of these matrices means that a relatively crude approach is required to modify these matrices through time in order to produce future year matrices. This approach cannot take into account any fundamental changes in trip patterns arising through changing population and attraction opportunities.
2.2.8 Strategic travel demand models derive synthetic trip matrices through a series of behaviourally-based relationships using population, employment and trip rate information. Unless these synthetic matrices are subsequently manipulated to match some observed targets, it is unlikely that a validation of resulting traffic flows would be as accurate in the base year as those obtained in a local model using matrices built from survey data.
2.2.9 To produce a satisfactory model for this study it is important to bring the best aspects of the two types of model together into a single modelling framework; synthetic travel demand matrices are developed based on the choices facing travellers and observed matrices calibrated with observed origin and destination information from survey data. By making the two sets of matrices (synthetic and observed) comparable, the synthetic model is able to forecast changes in model matrices which can be applied to the observed highway matrices.
2.3 MODEL FORM AND STRUCTURE
2.3.1 The NSTM consists of two main models:
Observed assignment model
Synthetic or Demand model, incorporating the traditional four stages of trip generation, mode choice, trip distribution and assignment which uses the observed assignment model outputs
2.3.2 Of the two models, the observed model is the more accurate representation of the current (base year) conditions simply because it is based on the best significant volumes of directly observed data. However, the future will bring complex changes to land use and transport infrastructure that will have a significant impact on travel demand. In this case the observed model becomes less reliable if taken forward in its own right. The Demand Model uses the synthetic approach to develop the best estimate of changes in demand over time and as such is the ideal tool for forecasting as it considers so many of the complexities that the observed model cannot, such as destination choice and mode shift between all motorised and slow modes.
8 NSTM - LMVR Highway Model Version 1.0
2.3.3 Current guidance on model development indicates that one of the main decisions required at an early stage is the form of the forecast model. Namely whether they are implemented as incremental models or whether they are absolute models. WebTAG Unit 3.10.3 Section 1.5 sets out the three main types of the model form:
Absolute models that use a direct estimate of the numbers of trips in each category
Absolute models applied incrementally that use absolute model estimates to apply changes to a base matrix
Pivot point models that use cost changes to estimate the changes in the numbers of trips from a base matrix
2.3.4 The Department for Transport (DfT) preference is for an incremental form of model whether pivot-point or based on incremental application of absolute estimates. The NSTM is an absolute model applied incrementally. The overall structure of the model is shown in Figure 2.1.
NSTM - LMVR Highway Model Version 1.0 9
Figure 2.1: Overall NSTM Structure
2.3.5 The synthetic demand model, developed in EMME3, follows the traditional 4-stage hierarchy. This model considers the travel behavioural reactions induced by changes such as land use, socio-economics, and transport supply in developing the transport demand.
10 NSTM - LMVR Highway Model Version 1.0
2.3.6 The base year and forecast trip patterns are produced by the synthetic demand model calibrated to fit as closely as possible to the known observed travel patterns. The absolute model estimates of the base and forecast trip patterns are then used to apply changes to a base observed matrix.
2.3.7 The base year synthetic model is validated to the known observed travel choices and is used to produce future forecasts. The changes, which take into account changes in generalised cost and zone attractiveness, between the base year and the future forecast in the synthetic model are then inputted into the base observed model to measure the actual change between the base year and the future forecast.
2.3.8 The Highway and Public Transport observed models are developed in SATURN 10.9.22 and EMME3 respectively in line with the Design Manual for Roads and Bridges (DMRB) Volume 12 standards.
2.3.9 Further information on the Public Transport Model is provided in the NSTM Local Model Validation Report – Public Transport Model, December 2009.
2.4 OBSERVED ASSIGNMENT MODEL
2008 BASE YEAR
2.4.1 AM peak (08:00-09:00) and PM peak (17:00-18:00) observed assignment models were developed for the base year. The observed base year matrices were developed for each existing mode type (i.e. bus, rail, and vehicles including private vehicle and goods vehicle). These matrices were partially developed from observed trip behaviour determined through a combination of Roadside Interviews (RSIs), Automatic Traffic Counts (ATC), Manual Classified Counts (MCC), Bus and Rail surveys. The unobserved trips were created using the previous version of the NSTM synthetic model to infill. All peak models have been calibrated and validated to the current acceptability criteria as detailed in the Design Manual for Roads and Bridges (DMRB) Volume 12.
OBSERVED MODEL DATA COLLECTION
2.4.2 The NSTM prior matrices were constructed from Roadside Interview (RSI) data used to create the previous two models. Original interviews were conducted in 2002 and 2005. To update the data to 2008, new information from Automatic Traffic Counts (ATCs) at the same sites was used. The updated ATCs were either undertaken as part of NCC’s rolling data collection programme, or were specially commissioned. There were a number of sites for which it was not possible to obtain new data; in these cases, TEMPRO was used to factor the original ATC to a 2008 total.
2.4.3 The 2011 update incorporated an additional 6 RSI sites surveyed in June 2009 for the A43 Corby Link Road model, and 5 RSI sites, also surveyed in June 2009, for the Atkins Isham Bypass model.
2.4.4 A total of 44 RSI sites were used to create the prior matrix. These are shown in Table 2.1 and Figure 2.2.
C2 A6003 Rockingham Rd, N of Kettering 2009 Tempro
12 NSTM - LMVR Highway Model Version 1.0
RSI Site Number Location Date 2008
ATC ID Source
C3 A427 Corby Rd, W of Corby 2009 Tempro
C4 A6003 Uppingham Rd, Corby 2009 Tempro
C5 Gretton Brook Rd, Corby 2009 Tempro
C6 A43 Stamford Rd, E of Corby 2009 Tempro
I1 A510 Woodford Rd, Kettering 2009 Tempro
I2 Finedon Rd, Wellingborough 2009 Tempro
I3 A6 Burton Rd, Burton Latimer 2009 Tempro
I4 A509 Kettering Rd, Great Harrowden 2009 Tempro
I5 A43 Hannington, N of Northampton 2009 Tempro
AUTOMATIC TRAFFIC COUNTS
2.4.5 MGWSP undertook Automatic Traffic Counts (ATCs) at those locations shown in Table 2.1. The ATC surveys collected data in both directions over a two to three week period. The ATC surveys were undertaken using a set of parallel pneumatic road tubes being installed at each location separated by a distance of 36.1 metres. The tubes were then connected to an Automatic Traffic Counter which was set to obtain classified directional flows in fifteen minute intervals.
2.4.6 Other ATC information collected by MGWSP has been used as part of the model calibration and validation. The 2011 updated incorporated additional ATC data used for the calibration of the A43 Corby Link Road model.
MANUAL CLASSIFIED LINK COUNTS
2.4.7 Manual classified link counts were undertaken as part of the existing RSI surveys and were undertaken on the same day as the RSI. These surveys recorded all vehicles travelling through the RSI locations and provided a classification of traffic volume by vehicle type and time period. This enabled the sample of the trip patterns derived from the interviews to be factored to match the total traffic volumes.
2.4.8 The traffic count data was collected in 15 minute intervals using the following classifications:
P/CYCLE: Pedal Cycles
M/CYCLE: Motor Cycle
CAR
LGV: Light Goods Vehicle
OGV1: Other Goods Vehicle (2 and 3 axles)
OGV2: Other Goods Vehicle (4+ axles)
PSV: Coaches and service buses
NSTM - LMVR Highway Model Version 1.0 13
MANUAL CLASSIFIED TURNING COUNTS
2.4.9 The 2011 model update incorporated additional count data from Manual Classified Turning Counts collected for the development of the A43 Corby Link Road model. The model also incorporated data from counts around Towcester.
TRADS DATA
2.4.10 The Traffic Flow Database System (TRADS) system is a traffic information database maintained by the Highways Agency, which provides access to traffic flow information collected from England’s motorway and trunk road network.
2.4.11 Monthly traffic flows were obtained for the major trunk roads in Northamptonshire for the months of October and November 2008. These sites are shown in Table 2.2 and Figure 2.3.
Table 2.2: Location of Sites for TRADS Data
ID Road Site Location
1 A5 North of Weedon
2 A14 South of Welford
3 A14 Rothwell
4 A14 Kettering Bypass
8 A43 West of Towcester, South of A5, North of A413
11 A43 TMU Site 6860/1 & 6861/1 between A5 & M1
13 A43 TMU Site 6859/1 & 6760/1 between A422 near Brackley (east) & A5
14 A45 Highham Ferrers Bypass
16 A45 Great Billing
17 A45 North of Hardingstone
18 A45 North of M1 J15
19 A45 TMU Site 6786/1 & 6786/2 between A6 & B663
20 M1 M1J14 – J15
21 M1 J15 – J15A
22 M1 J15A – 1J6
23 M1 J16 – J 17
24 A5 Towcester Racecourse
25 M1 J17-18
26 A14 TMU Site 6801/1 & 6802/1 within A508 junction
27 A43 Slips with M1 J15A
28 A14 W of Kettering
29 A428 TMU Site 6864/1 & 6865/1 between A5 Rugby and A5/M1
14 NSTM - LMVR Highway Model Version 1.0
2.4.12 The set of TRADS data was updated for the 2011 model update to resolve issues of double counting or conflicting counts arising from the inclusion of the other additional count data.
JOURNEY TIME SURVEYS
2.4.13 MGWSP commissioned Nationwide Data Collection (NDC) to carry out journey time surveys on key routes in the Northamptonshire area. Journey time surveys were undertaken in both directions for the following 21 routes:
Route 4 - A45 Daventry to Northampton
Route 7 - Wellingborough Cordon
Route 10 - A6013/A6900/A6003 Through Kettering
Route 12 - A6 West of Desborough
Route 16 - A6014/Studall Avenue
Route 17 - Rockingham Road/Phoenix Parkway
Route 18 - Cottingham Road, Corby
Route 23 - A428 East to Rugby
Route 24 - A508 North of Northampton
Route 26 - A43 North of Northampton
Route 27 - A519
Route 29 - Daventry Ring Road
Route 30 - North Northampton Ring Road
Route 31 - A509, Corby to Kettering
Route 33 - A423 through Corby to Oundle
Route 34 - A5101 into Northampton
Route 35 - A510 A14 of Northampton
Route 36 - Northampton Northern Ring Road
Route 37 - A508/A5123
Route 38 - Lower Street/Stamford Road, Kettering
Route 39 - Kettering to Great Doddington
2.4.14 These routes are shown in Figure 6.1. The journey time surveys were carried out in such a way to try and ensure that at least six observations in each direction during each time period were taken. The surveys were conducted using the “moving observer” method, where a set route was driven with a GPS receiver set to record the vehicle’s position every three seconds. Each vehicle was driven at the speed of general traffic flow while also observing speed limits in order to ensure journey times
NSTM - LMVR Highway Model Version 1.0 15
were representative of typical traffic conditions. The surveys were carried out between 4th-18th November 2008 for the 07:00-10:00 and 16:00-19:00 time periods.
BUS INTERVIEW SURVEYS
2.4.15 MGWSP commissioned Monisyst to carry out surveys on the main bus routes in and between Northampton, Kettering, Wellingborough, Corby and Daventry.
2.4.16 62 main bus routes throughout the county were surveyed during November and December 2008 over a twelve hour period (07:00-19:00) providing the following data:
Boarding and Alighting data on bus services
Face to face interviews on bus services and at Bus Depot
Self completion free post questionnaires given out on bus services, bus stops and bus depots
RAIL INTERVIEW SURVEYS
2.4.17 MGWSP commissioned Monisyst to carry out surveys at the three main Northamptonshire rail stations namely Kettering, Wellingborough and Northampton. The three train stations were surveyed between 17 November 2008 and 4 December 2008. The surveys were carried out over a 15 hour period between 06:00-21:00.
2.4.18 All services which stopped at these stations during the peak hours were included within the surveys.
2.4.19 The rail services which stop at these stations are:
Northampton: Milton Keynes, Watford Junction, London Euston, Rugby, Coventry and Birmingham New Street
Kettering & Wellingborough: Bedford, Luton, London St Pancras, Leicester, Nottingham, Derby, Sheffield and Manchester
2.4.20 Full detail of the data collection process can be found in the Northamptonshire Multi-Modal Transport Model Survey Report, March 2009.
2.5 SYNTHETIC DEMAND MODEL
2.5.1 A synthetic demand model has been developed for each peak period. Each demand model has been calibrated in accordance with WebTAG to represent travel movements based on socio-economic data. The synthetic demand model uses the same network and zoning system as the observed assignment model.
2.5.2 The guidance available from WebTAG contains advice on the most appropriate choice hierarchies to adopt for the mode and distribution stages within travel demand models. This guidance has been given significant weight in the design of this model.
16 NSTM - LMVR Highway Model Version 1.0
2.5.3 The WebTAG guidance states that the hierarchy of choice components, ranging from the choices at the top that are least sensitive to supply characteristics in the order of increasing sensitivity to the bottom, is as follows:
Trip frequency (or generation) which may be represented as inelastic if all modes of transport (i.e. walk and cycle) are included
Main mode choice (i.e. between car, public transport and walk/cycle) is an insensitive choice component
Destination Choice (or trip distribution) is a sensitive choice component as it is based upon the relative attractiveness and the travel cost to a destination zones
Route assignment is at the bottom of the choice hierarchy, being the most sensitive of all of the choices, to the measured cost and time characteristics of the route options available
2.5.4 Trip frequency and mode choice are significantly less sensitive than destination choice. The NSTM follows the hierarchy stated above for all trip purposes and is compliant with current guidance and advice. Further information on the synthetic model is provided in the NSTM Demand Model – Model Development and Validation Report, November 2009.
NSTM - LMVR Highway Model Version 1.0 17
3 Highway Network Development
3.1 INTRODUCTION
3.1.1 The highway network was developed in SATURN by combining North Northamptonshire and West Northamptonshire model networks. The combined network went through checks on link types, link speed and link lengths and it was extended so that the model covered the entire United Kingdom.
3.2 STUDY AREA
3.2.1 The NSTM study area is presented in Figure 3.1. It has been defined to allow for the robust testing of possible impacts on the local transport network of infrastructure schemes and development proposals. The traffic corridors influencing travel patterns within the county include major motorways and trunk roads. In order to reflect these movements the model area extends to include a representation of motorways and trunk roads to outlying areas of the UK which is shown in Figure 3.2.
3.2.2 The main area of interest is Northamptonshire, where the density of road links necessary to accurately model the movements intensifies. The main study area encompasses the major motorways of the M1 (junctions 14 to 19) and M45 (from Junction 17).
3.3 ZONING AND SECTORING SYSTEMS
3.3.1 The zone system reproduced in Figure 3.3 and Figure 3.4 is based on the 2001 Census Output Areas (COA) and wards, ensuring compatibility with Census household data. The COAs were aggregated or disaggregated into different land uses within Northamptonshire to represent the county at the appropriate level of detail required for the study. The model zone system is sufficiently detailed so that all major trip movements within the study area can be identified to a level consistent with existing transport models covering the study area, as well as inter-urban movements across the county and wider strategic movements.
3.3.2 In total there are 614 zones in the NSTM, of which 548 are allocated to the Northamptonshire county area. A list of the zones is given in Appendix No 1.
3.3.3 The public transport zone system is identical to that used in the highway model. The zones are categorised into 19 regional sectors presented in Figure 3.5 and Figure 3.6.
18 NSTM - LMVR Highway Model Version 1.0
3.4 NETWORK DEVELOPMENT
3.4.1 In drawing up the road network, the following principles were borne in mind so that:
All major and minor road links within the Northamptonshire area were included
Outside the town centres, links between suburbs and villages were included, where necessary, to ensure sensible routing of trips to and from these areas
3.4.2 In constructing the highway network the following highway link characteristics were considered for inclusion:
Link length
Speed/flow relationship
Link type
Link capacity
One way/two way operation
Vehicular restrictions
Provision of bus/cycle lanes
Traffic calming and its impact on speed/capacity
On street parking and its impact on speed/capacity
Number of (effective) lanes
Any observed turn delays/penalties
Location of public transport routes/bus stops
Location of taxi ranks
Car park locations
Access points
3.4.3 Link speeds were determined by the speed limits of each road. Link lengths were measured and checked using a GIS map based system and 1:25,000 scale maps.
3.4.4 Simulation modelling has been used throughout for the whole of the modelled road network within the study area. The external connections include an averaged distance for each longer distance zone.
3.4.5 In an urban area, the main delays to a journey result from the interaction of traffic at junctions. SATURN is structured to model these interactions, and it is usually assumed that traffic within the simulation network travels at a uniform speed between the nodes of the model, with delay occurring only at the junction.
NSTM - LMVR Highway Model Version 1.0 19
3.4.6 Speed/flow curves have been used within the model area on the main arterial routes, inter-urban routes and country lanes within the core study area. All links along such routes were included, to ensure consistency of travel behaviour along the route. Links in urban areas were not included, since capacity is determined primarily by junction control. They have been derived from the speed/flow relationships contained within DMRB Volume 13 (COBA manual). The speed/flow curves that have been used in the NSTM have been derived from those contained within DMRB Volume 13 (COBA manual) and are shown in Table 3.1. Table 3.1: Speed/flow relationship
S0 S1 S2 F C N Description
Rural - Dual
105 79 8760 S 2.75 1 D4M
105 79 6570 S 2.75 2 D3M
102 76 4380 S 2.90 3 D2M
102 76 2190 S 2.90 4 D1M
101 73 7600 S 2.75 5 D4AP
101 73 6030 S 2.75 6 D3AP
97 68 4020 S 2.70 7 D2AP
Rural – Single
92 55 1730 S 2.35 8 Single Lane Slip Road – Motorway
57 21 1700 S 0.80 24 Urban Non-Central 2-Lane 50% Dev
54 25 1641 S 1.67 25 Urban Non-Central 50% Dev
48 25 1560 S 1.56 26 Urban Non-Central 80% Dev
44 25 1480 S 1.48 27 Urban Non-Central 100% Dev
37 15 740 S 1.83 28 Urban Non-Central INT=2
34 15 630 S 1.73 29 Urban Non-Central INT=4.5
28 15 450 S 1.55 30 Urban Non-Central INT=9
64 25 1770 S 3.06 31 Small Town 35% Dev
56 30 1000 S 3.39 32 Small Town 60% Dev
46 30 880 S 2.45 33 Small Town 90% Dev
50 21 1200 S 2.15 34 Country Lane
64 64 99999 S 0 35 Suburban with JCN modelling-1
32 32 99999 S 0 36 Suburban with JCN modelling-2
97 97 99999 S 0 37 Suburban with JCN modelling-3
48 48 99999 S 0 38 Suburban with JCN modelling-4
20 5 100 S 1 39 Park & Ride Links
3.4.7 All junctions had their coding updated to reflect any changes that may have occurred with the information being provided from the network inventory survey and aerial junction photographs. Each junction simulated in the SATURN simulation area network requires several parameters as detailed below:
Lane allocations
Junction type
Saturation flows at traffic signals
Signal times, stages and phases
Circulation and saturation flows at roundabouts
Gap acceptance criteria at priority junctions
3.4.8 In total, there are 2884 simulation nodes within the network. These are separated into the following junction types:
128 external nodes
2221 priority junctions
292 roundabouts
137 signal-controlled junctions
NSTM - LMVR Highway Model Version 1.0 21
3.4.9 Saturation flows are consistent with advice contained in TRL Reports LR942 (“The Traffic Capacity of Roundabouts”), RR67 (“The Prediction of Saturation Flows for Road Junctions Controlled by Traffic Signals”) and SR582 (“The Traffic Capacity of Major/Minor Priority Junctions”). Default gap acceptance has been used. It was necessary to use a set of default values in order to code the very large number of junctions in a timely fashion. The saturation flows adopted are given in Table 3.2 below.
Table 3.2: Default Saturation Flows
Junction Type Movement Saturation Flow
Priority
Straight ahead 1965
Major-minor left turn 1500
Major-minor right turn 745
Minor-major left turn 700
Minor-major right turn 600
Roundabout
One lane 1650
Two lanes 3200
Three lanes 4500
Signals
Left turn 1740
Straight Ahead 1980
Right turn 1740
ROUTEING VALIDATION
3.4.10 The Forest function in the SATURN programme P1X plots all the routes used within the network for trips between two zones. By using the Forest function for many zone pairs throughout the study area, routes used by traffic were examined to ensure they were reasonable and resemble probable routes used by vehicles in the current traffic situation. The routes used in each situation are the lowest cost routes available between the two zones. It also displays the percentage of traffic using each different route, thus highlighting the most common path taken.
3.4.11 To check that the assigned routes in the model were feasible and in accordance with common sense, ‘trees’ were built between selected OD pairs. The resultant minimum path routes are shown in Appendix No 2 and illustrate that the assigned routes are sensible.
22 NSTM - LMVR Highway Model Version 1.0
NETWORK REVIEW AND UPDATE
3.4.12 A review of the previous model was carried out by MGWSP, which led to the key changes listed below:
Nodes in Northampton were coded with a gap acceptance parameter of 5 seconds, a legacy from the previous Northampton model which generated significant delays - the default was revised to 2 seconds;
Existing bus priority measures have been included;
Minor routing issues around Oundle have been fixed;
Centroid loading points have been reviewed;
Addition of minor links around the Weedon and Flore to allow future improvements to be tested;
Corrections to various nodes including changes to number of lanes, saturation flows, node ordering, and signal timings.
Following advice from NCC, schemes previously modelled as forecast improvements now included in the base as follows;
Black Lion Hill Interchange – The access to Northampton train station was altered to include a bus only access to Marefair and a signalised bus priority junction
A14 Kettering Bypass Widening – This is a Highways Agency scheme to carry out an online widening of the A14 around Kettering within the existing highway boundary. The scheme runs just east of Junction 7 (A43) to just west of Junction 9 (A509)
3.4.13 Applying the above changes to the network had a detrimental impact on the validation of the existing model, requiring a revalidation exercise to be undertaken to restore the model to its previous level of validation. This report sets out the results of the revalidation exercise.
3.5 SUMMARY
3.5.1 During the network building process the following activities have been undertaken:
Review of network coding warnings produced by the SATURN program SATNET
Network link distance checks
Review of junction saturation flows
Review of gap acceptance parameters
Detailed review of the coding of complex junctions
Checks on key traffic routings on the unloaded and loaded networks to check for consistency
NSTM - LMVR Highway Model Version 1.0 23
Check all-or-nothing routes – the network was thoroughly checked to ensure the sensible routes are allocated between zones. This check was undertaken to check general network coding or, where junction coding is extremely complex, that the correct allowable turning movements have been coded
4.1.1 The Prior matrices were developed using data from 44 Roadside Interview (RSI) sites, plus trips taken from the initial synthetic base model. The RSI data was used to create an initial matrix based fully on observed data which is known as the RSI Matrix. Data from the synthetic model was used to infill trip patterns missed by the RSI surveys and this is known as the Infill Matrix. The combined RSI Matrix and Infill Matrix is known as the Prior Matrix.
4.1.2 Highway matrices for Car/LGV (Light Matrix) and HGV (Heavy Matrix) vehicle classifications were developed separately. All Light Matrix and Heavy Matrix volumes presented in this report are in Passenger Car Units (PCUs) rather than vehicles unless where stated.
4.2 RSI EXPANSION FACTORS
4.2.1 Roadside Interview (RSI) data has been used to construct the Observed Matrix based upon fully observed data. The surveyed movements represent the interview direction (ID), whereas the non-interview direction (NID) trips were derived from the transpose of the opposite time period. The origins and destinations of the AM peak ID records were reversed to represent the PM NID trips and vice versa. The transposed trips were scaled to the appropriate count – e.g. the transposed AM peak interview records were expanded to match the PM peak non-interview count.
4.2.2 Expansion factors were calculated to scale the interview samples to observed counts. It is desirable to maintain a well-distributed matrix by expanding the surveys for the three-hour peak period and then factor to match peak hour counts. Expansion factors were typically between 3 and 8, and on most occasions were less than 10. Further information can be found in the RSI data collection report.
4.3 DOUBLE COUNTING
4.3.1 The RSI matrices were combined and the multiple observations of a trip at more than one RSI site were removed by applying the following process. All trips from Zone A to B at different sites were summed up and then divided by the number of sites they appear in. It is understood that this technique might produce overestimates for the ‘double counted’ movements, but it is felt that, in the absence of further information, this is the best technique available. The effect has been closely monitored and the movements on the parallel routes have been thoroughly checked.
NSTM - LMVR Highway Model Version 1.0 25
4.4 JOURNEY PURPOSE MATRICES
4.4.1 RSI Matrices were produced for each of the following journey purposes:
Cars: Home Based Work (HBW)
Cars: Home Based Employers Business (HBEB)
Cars: Home Based Other (HBO)
Cars: Non Home Based Employers Business (NHBEB)
Cars: Non Home Based Other (NHBO)
Light Goods Vehicles (LGV
Other Goods Vehicles 1 (OGV1)
Other Goods Vehicles 2 (OGV2)
4.4.2 The Car and Light Goods Vehicle (LGV) journey purposes (HBW, HBEB, HBO, NHBEB, NHBO and LGV) were combined to form a Lights Matrix. The OGV1 and OGV2 journey purposes were combined to form a Heavy Matrix.
4.5 PCU FACTORS
4.5.1 The RSI surveys produced data in vehicle totals. The journey purpose matrices were factored by the following Passenger Car Unit (PCU) factors:
Cars and LGV: 1.0
OGV1: 1.5
OGV2: 2.3
4.5.2 These factors were applied to data collected by the RSI surveys to convert the data from vehicles to PCUs. Factors derived from the MCC data were used to convert the average HGV totals obtained from the ATCs into PCUs, on a site by site basis. The peak hour time periods were chosen to coincide with times of maximum traffic flow on the majority of links in the network.
4.6 RSI MATRIX
4.6.1 The above data was combined to form an RSI matrix for each period, representing those trips captured by the RSI sites, with potential double-counting taken into account. This matrix was assigned to the network and subsequent sensitivity testing of traffic routings and comparisons between observed and modelled flows were undertaken.
4.6.2 Select links were undertaken at each RSI site and on competing routes to determine the routing of observed traffic. Routes were subsequently checked to ascertain whether the routes selected were valid, given the uncongested nature of the network, or whether there was a network error leading to unlikely route choice. This ensured that the best possible RSI matrix, producing the optimum fit to the observed flows, was progressed to the Prior Matrix development. A comparison of modelled and
26 NSTM - LMVR Highway Model Version 1.0
observed counts from assignment of the AM RSI matrix is given in Appendix No 11, with the PM peak given in Appendix No 12.
4.7 INFILL MATRIX
4.7.1 The NSTM is a multi-user class model with 2 levels within the matrix i.e. a Lights Matrix (formed of Cars & Light Goods Vehicles) and a Heavy Matrix (formed of Heavy Goods Vehicles (HGV)) which is the RSI Matrix. The RSI Matrix had trips infilled to account for the unobserved trips. Only trips between those OD pairs that did not pass through one of the 44 RSI sites were infilled e.g. trips between Zone 1 and Zone 25 where there were zero observed trips have been extracted from the synthetic matrix and infilled for a greater level of accuracy within the matrix.
4.7.2 No viable information to infill the HGV matrix was available. It was felt that a flat seed value would result in the matrix estimation process creating trips between unlikely origin-destination pairs. Since the large majority of unobserved HGV trips were long-distance strategic trips, the matrix was manually seeded to reflect observed through-trips.
4.8 PRIOR MATRIX
4.8.1 The RSI Matrix and the Infill Matrix have been combined into a Prior Matrix. The development of the highway observed matrix or the Prior Matrix is shown in Figure 4.1.
Figure 4.1: Highway Observed Matrix Development
RSI Matrix (OD by JP)
Expansion to ATC counts
Double Counting
Prior Matrix
RSI Matrix
Synthetic Matrix
NSTM - LMVR Highway Model Version 1.0 27
4.9 MATRIX ESTIMATION
4.9.1 The model calibration process has used matrix estimation procedures as contained in the SATME2 program. The basic operation of this uses the best estimate of trip movements as contained in the prior matrix and adjusts the pattern of trip distribution and volumes in order to match a file of input traffic flows.
4.9.2 The Prior Matrix was assigned to a network representing the current highway network. SATME2 requires a PIJA file which represents the proportion (P) of trips between a particular origin-destination pair (IJ) which uses the counted link (A). The PIJA data is obtained through the program SATPIJA following a SATURN assignment using the SAVEIT option. This produced a PIJA output files for LGV and HGV which were used by SATME2 along with the Prior Matrix to produce updated 2008 estimated Light Matrix and Heavy Matrix which were then combined into a ‘stacked’ estimated matrix for assignment. No cells were frozen and there were no zonal constraints applied. The default number of iterations i.e. 30 was allowed for the estimation process.
4.9.3 Primary inputs to the calibration process were traffic flows used as target counts for the matrix estimation process. The number of sites used is shown in Table 4.1 with most sites comprising a traffic count in each direction of travel. The matrices have been built to the same standard for both peaks and this is reflected in the calibration results reported in Section 5.
Table 4.1: Number of ME2 Counts
Total Number of Calibration counts used
AM Peak PM Peak
404 404
4.9.4 A large number of sites were used as it was important to ensure that the traffic volumes were matched at controlling points across the network. The matrix estimation process enabled traffic flows to be updated by varying degrees across different spatial areas of the network. The data within the count file has used those counts deemed the most accurate in the order required to produce the calibration and validation results shown in Table 5.5 and Table 6.1
4.9.5 The AM Peak shows a reduction in the size of the Prior Matrix of 8,796 (0.98%) while the PM Peak shows a reduction in the size of the Prior Matrix of 8,892 (0.99%). A secondary round of matrix estimation was undertaken in order to improve the model validation. This is discussed in more detail in Section 6.3.3, but the details of the trip matrix are also presented here, labelled as “Final Trip Matrix”. The second round of matrix estimation results in a reduction in the size of the trip matrix of 8,390 (0.93%) in the AM peak and 7,228 (0.80%) in the PM peak. The trip matrix totals for the AM peak and PM peak are shown in Table 4.2 and Table 4.3.
28 NSTM - LMVR Highway Model Version 1.0
Table 4.2: AM Peak - Comparison of Trip Matrix Totals
AM Peak
Cars & LGVs
(passenger car units)
HGV
(passenger car units) Total
Pre ME Trip Matrix Totals 893,045 7,182 900,227
Post ME Trip Matrix Totals 886,143 -0.77% 5,288 -26.37% 891,431 -0.98%
Final Trip Matrix Totals 886,639 -0.72% 5,198 -27.62% 891,837 -0.93%
Table 4.3: PM Peak - Comparison of Trip Matrix Totals
PM Peak
Cars & LGVs
(passenger car units)
HGV
(passenger car units) Total
Pre ME Trip Matrix Totals 894,475 5,278 899,753
Post ME Trip Matrix Totals 887,609 -0.77% 3,252 -38.39% 890,861 -0.99%
Final Trip Matrix Totals 889,224 -0.59% 3,301 -37.46% 892,525 -0.80%
4.9.6 To ensure that the matrix estimation process was not fundamentally altering the composition of the prior matrices, two checks were undertaken. Firstly, a regression analysis was carried out on the Prior Matrix and Estimated Matrix. This involved compressing the matrices to a sector level comprising 19 sectors to determine if the trip distribution of the prior matrix was maintained following matrix estimation. These sectors are shown in Figure 3.5 and Figure 3.6.
REGRESSION ANALYSIS
4.9.7 For each time period, the total Prior Matrix and Estimated Matrix (Light Matrix and Heavy Matrix) were sectored on this basis and using the sectored version, the value of each matrix cell as a percentage of the matrix total was calculated. These values were then used to carry out a regression analysis on the two data sets, where an R² statistic is calculated that provides an indication of the degree of match between the data sets. A value of 1 would indicate a perfect match and generally values in excess of 0.9 are considered to indicate an acceptable level of compatibility.
4.9.8 The results of the analysis are provided in Table 4.4. The results show that all R² values are in excess of 0.9. This indicates that the distribution of the Prior Matrix has been maintained in the post matrix estimated matrices.
NSTM - LMVR Highway Model Version 1.0 29
Table 4.4: Total R2 values for Sector to Sector Analysis
AM Peak PM Peak Post Me Trip Matrix 0.9998 0.9998
Final Trip Matrix 0.9998 0.9998
4.9.9 Appendix No 3 shows the sector to sector comparison for the matrices, by user class, for the AM Peak and PM Peak.
COMPARISON OF TRIP LENGTH DISTRIBUTION
4.9.10 The second check involved comparing the trip length frequency distribution exhibited by the prior and post estimation matrices. The test compares the frequency of trip lengths between the prior and post estimated matrices to ensure that the overall integrity of the matrix is maintained in particular that matrix estimation process does not add too many short distance trips. Comparing the trip length distribution of the post ME matrix with that of the pre ME matrix gives some indication of the level of distortion caused by the matrix estimation process.
4.9.11 The trip length bands for the AM and PM peaks are shown below. Table 4.5: Trip Length Distribution
4.9.12 Appendix No 4 shows the trip length distribution for the AM Peak and PM Peak graphically. Overall the post ME matrix for the AM Peak and PM Peak maintain the pre ME matrix trip length distributions well. The number of trips in each range as a percentage of total number of trips was similar in the pre ME matrix and post ME matrix for both peak periods.
4.10 SUMMARY
4.10.1 The results of the regression analysis and the trip length distribution comparison confirm that the matrix estimation process is not altering the character or the overall pattern of trips in the prior matrices and that the trip length distribution is similar pre and post matrix estimation. The prior matrix and final base year matrix trip totals are shown in Table 4.2 and Table 4.3. The output matrices from the estimation process are robust and suitable for use in the SATURN assignment process. The Prior Matrix has been assigned to network and detailed calibration and validation carried out which is discussed in Section 5.
NSTM - LMVR Highway Model Version 1.0 31
5 Calibration - Highway Assignment Model
5.1 INTRODUCTION
5.1.1 The process of model calibration is designed to ensure that the parameters that control the model’s calculation of route choices and the resulting delays lead to accurate replication of traffic patterns in the network. The calibration of a model crucially relies on traffic entering the detailed model area at the correct points and on the correct routes.
5.1.2 A number of key calibration checks were made between the modelled flows and observed flows:
Matrix Calibration
Checking assignment model convergence (See Section 5.3)
Comparison of modelled flows against observed flows that were used explicitly in the creation of the Prior Matrix which include those used for developing the RSI Matrix and those used in matrix estimation (See Section 5.4, Appendix No. 5 and Appendix No. 6)
Comparison of sector to sector movements between the Prior observed model and the Post observed model after the matrix estimation process (See Section 4.8)
Trip Length Distribution (See Section 4.8)
Network Calibration
Adjustment and checking of the network to ensure plausible and realistic routing of traffic in the model through monitoring RSI matrix assignments (See Section 3.4)
Comparison of observed and modelled traffic flows (See Section 5.4, Appendix No. 5 and Appendix No. 6)
5.2 ASSIGNMENT PARAMETERS
5.2.1 Part of the calibration process is the identification of the assignment parameters which form the generalised cost function i.e. PPK (pence per kilometre) and PPM (pence per minute) which control the assignment algorithm.
5.2.2 Generalised cost parameters were input to the model networks to determine the minimum cost routes by which traffic assigns itself onto the network. For the vehicle journeys modelled in the NSTM, generalised cost is a function of both the in-vehicle travel time and the operating costs (related to distance travelled).
32 NSTM - LMVR Highway Model Version 1.0
5.2.3 DfT WebTAG guidance on modelling states that:
“It is important for models of behaviour to use realistic measures of generalised cost” (WebTAG, Unit 3.1.2 paragraph 5.1.11)
5.2.4 Generalised costs for use in the NSTM were calculated using values of time, GDP growth rates, purpose splits, and vehicle operating costs recommended by the DfT for use in economic appraisals of transport projects in England. These values are presented in WebTAG Unit 3.5.6 (April 2011), and values calculated for use in the base year models are shown in Table 5.1. Table 5.1: Generalised Cost Parameters
Time Period Pence Per Minute (PPM) Pence Per Kilometre (PPK)
Cars & Lights HGV Cars & Lights HGV
AM Peak 19.62 18.86 10.12 33.63
PM Peak 17.63 18.86 10.12 33.63
5.2.5 The generalised cost parameters were input into SATURN as separate values for time and distance. Time is calculated in Pence per Minute (PPM), and distance as Pence per Kilometre (PPK) within SATURN. Although the parameters are specified as ‘per minute’ and ‘per kilometre’ the actual units used by SATURN for time and distance are seconds and metres with the appropriate conversion handled by the program. The assignment within SATURN works in terms of ‘generalised cost’, defined by adding terms proportional to the link distance and/or monetary charge to the link travel time.
5.2.6 The generalised cost functions used for the AM Peak for each vehicle class are shown below:
Cars and Lights: GC = 1.00 t + 0.52 d
Heavy Goods Vehicles GC = 1.00 t + 1.78 d
5.2.7 The generalised cost functions used for the PM Peak for each vehicle class are shown below:
Cars and Lights: GC = 1.00 t + 0.57 d
Heavy Goods Vehicles GC = 1.00 t + 1.78 d
5.2.8 The assignment technique that has been used is Wardrop Equilibrium Assignment which is the default assignment procedure within SATURN. Wardrop’s Principle of traffic equilibrium is states that:
“Traffic arranges itself on congested networks such that the cost of travel on all routes used between each O-D pair is equal to the minimum cost of travel and all unused routes have equal or greater cost.”
NSTM - LMVR Highway Model Version 1.0 33
5.3 MODEL CONVERGENCE
5.3.1 An element of calibrating the model is ensuring a satisfactory convergence is achieved. Model convergence is needed to ensure traffic flows remain stable between successive iterations of the model.
5.3.2 In accordance with criteria set out in the DMRB, the parameters %Flow and Delta (d) have been monitored to determine the level of convergence. %Flow measures the proportion of links in the network with flows changing by less than 5% from the previous iteration and d is the difference between costs on chosen routes and costs on minimum cost paths.
5.3.3 The convergence criteria used to assess when a model is considered to have converged is shown in Table 5.2. The values shown below are those deemed to be acceptable however the ISTOP value used for both peak hours is 98%.
Table 5.2: Convergence Criteria
Measure of Convergence Acceptable Value ‘Delta’ Less than 1%
Percentage of links with flow changes < 5% Four consecutive iterations greater than 90% Source: DMRB Volume 12, Section 2, Part 1, Chapter 4, Table 4.1.
5.3.4 Table 5.3 indicates that a satisfactory convergence has been achieved in all three time periods. WebTAG Unit 3.10.4 suggests that delta (d) values of less than 0.2% are reasonable targets. As the table shows, all delta values are less than 0.2% therefore the base year models for both peak periods meet the required convergence standards.
5.4.1 A modified Chi² statistic known as the GEH (Geoffrey Edward Havers) statistic was used to determine the best fit. The statistic uses the formula shown below to calculate a value for the difference between observed (O) and modelled (M) flows.
5.4.2 The calibration was undertaken for a comprehensive set of traffic count sites for each time period. The calibration counts have been further compared as a whole and as a series of screenlines to further assess current traffic movements.
5.4.3 The criteria used to assess the acceptable performance of a traffic model are defined in UK DMRB Volume 12 and are shown in Table 5.4.
Table 5.4: Assignment Acceptability Guidelines
Criteria and Measure Assigned Model Hourly Flows compared with Observed Flows Acceptability Guideline
Flow Criteria
Observed flow < 700 vph Modelled flow within ±100 vph > 85 % of links
Observed flow 700 - 2,700 vph Modelled flow within ±15% > 85 % of links
Observed flow > 2,700 vph Modelled flow within ±400 vph > 85 % of links
2 Total screenline flows (normally >5 links) to be within ±5% All (or nearly all) screenlines
GEH Criteria
3 GEH Statistic for individual links < 5 > 85 % of links
4 GEH Statistic for screenline totals < 4 All (or nearly all) screenlines Source: DMRB Volume 12, Section 2, Part 1, Chapter 4, Table 4.2.
5.4.4 The GEH statistic takes account of the fact that when traffic flows are low the percentage difference between observed and modelled flows may be high but the significance of this difference is small. A GEH value greater than 10 indicates that closer attention is required as the match between observed and modelled flows is poor, while a GEH of less than 5 indicates a very good fit.
5.4.5 It is an important measure of model performance that the model reproduces the observed volumes of traffic. The DMRB criteria for comparing the performance of the model traffic counts are reproduced in Table 5.4. Modelled flows are expected to be within a certain tolerance of the observed values and this goodness of fit is measured using the GEH statistic. Both the flow comparison and GEH criteria are assessed in this report.
M)+(O 0.5
2M)-(O
= GEH
NSTM - LMVR Highway Model Version 1.0 35
5.4.6 Either or both conditions are expected to be satisfied for 85% of the cases. Appendix No 5 details the calibration results for the AM Peak while Appendix No 6 details the calibration results for the PM Peak. The calibration counts are shown in purple on the diagrams while the validation counts are shown in red.
5.4.7 Table 5.5 shows the overall calibration results for the AM Peak and PM Peak. The Light Matrix and Heavy Matrix pcu volumes have been converted to vehicle volumes for the comparison with the DMRB flow criteria. Considering the nature and size of the model this level of calibration is thought to be robust and acceptable.
Table 5.5: Calibration results
AM PEAK
Flow Criteria Number of Observed
Counts
Number of Modelled Counts
Acceptability Guideline
Observed flow < 700 vph
Modelled flow within ±100 vph 247 222 90%
Observed flow 700 - 2,700 vph
Modelled flow within ±15% 142 129 91%
Observed flow > 2,700 vph
Modelled flow within ±400 vph 15 12 80%
TOTAL FLOWS 404 363 90% GEH for individual links < 5 404 348 86%
PM PEAK
Flow Criteria Number of Observed
Counts
Number of Modelled Counts
Acceptability Guideline
Observed flow < 700 vph
Modelled flow within ±100 vph 246 217 88%
Observed flow 700 - 2,700 vph
Modelled flow within ±15% 144 132 92%
Observed flow > 2,700 vph
Modelled flow within ±400 vph 14 11 79%
TOTAL FLOWS 404 360 89% GEH for individual links < 5 404 355 88%
36 NSTM - LMVR Highway Model Version 1.0
5.4.8 The calibration results for the AM Peak period meet the standards as stated in DMRB Volume 12a. Of the 404 counts analysed 348 counts, or 86%, have a GEH value under 5. 222 counts (90%) out of 247 have a flow of less than 700 while 129 counts, or 91%, out of 142 with a flow between 700 and 2,700 meet the standards. For counts with a flow of over 2,700 there are 12 counts (80%) out of 15 that meet the criteria.
5.4.9 The calibration results for the PM Peak period meet the standards as stated in DMRB Volume 12a. Of the 404 counts analysed, 355 counts or 88%, have a GEH value under 5. 217 counts (88%) out of the total 246 have a flow of less than 700, which meet the standards. 132 counts (92%), out of 144 have a flow of between 700 and 2,700, which meet the standards. For counts with a flow of over 2,700 there are 11 counts (79%) out of 14 that meet the criteria.
5.5 CALIBRATION CONCLUSIONS
5.5.1 The model calibration process was undertaken successfully and produced a high standard and quality of results for the modelled time periods. It has been shown that the prior trip matrices were improved by the use of matrix estimation techniques and that this process did not significantly alter the integrity of the prior trip matrices. Table 5.5 has shown that the criteria set out in Table 5.4 have been met for both the AM Peak and PM Peak. Considering the nature and size of the model this level of calibration is thought to be robust and acceptable.
NSTM - LMVR Highway Model Version 1.0 37
6 Validation - Highway Assignment Model
6.1 INTRODUCTION
6.1.1 A primary element of assessing the performance of a model is its ability to match an independent set of count data, which has not been used in the calibration process. For this purpose, a set of counts were identified that had not been used in the matrix estimation process.
6.1.2 A number of key validation checks were made between the modelled flows and observed flows:
Network Validation
Network validation, in terms of range checking and routing (See Section 3.4)
Checks to ensure that speed/flow calculations on network links and delay calculations at junctions were operating as expected (not covered in the report)
Matrix Validation
Comparison of modelled flows against independent observed flows which are those not used in the matrix building process (See Section 6.2, Appendix 7 and Appendix 8)
Comparison of observed and modelled journey time routes (See Section 6.2, Appendix 7 and Appendix 8)
6.1.3 The validation of the model has been assessed in a similar way to that of the calibration by the calculation of the GEH statistic. Acceptability guidelines on validation have been taken from DMRB Volume 12a and are as shown in Table 5.4. Validation has been carried out using modelled vehicle link flows extracted from an assignment and compared to observed vehicle link flows.
6.2 VALIDATION RESULTS
6.2.1 The validation of the network models has been conducted under three main headings:
Network validation (See Section 3.4)
Routeing validation (See Section 3.4)
Link flow validation
38 NSTM - LMVR Highway Model Version 1.0
6.2.2 A primary element of assessing the performance of a model is its ability to match an independent set of count data, which has not been used in the calibration matrix estimation process. For this purpose, a set of counts have been identified that had not been used in the matrix estimation process.
6.2.3 The validation of the model will be assessed in a similar way to that of the calibration by the calculation of the GEH statistic. Acceptability guidelines on validation have been taken from DMRB vol. 12a and are as shown in Table 5.4. Validation has been carried out using total modelled flows extracted from a stacked assignment and compared to observed light and heavy goods vehicle flows combined.
6.2.4 Model validation was undertaken against the following criteria:
Independent flow data
Journey time data
6.3 LINK FLOW VALIDATION
6.3.1 Tables detailing the calibration and validation results are included in Appendix No 5 for the AM Peak and Appendix No 6 for the PM Peak. The calibration counts are shown in purple on the diagrams while the validation counts are shown in red.
6.3.2 Table 6.1 shows the overall validation results for the AM Peak and PM Peak. The Light Matrix and Heavy Matrix pcu volumes have been converted to vehicle volumes for the comparison with the DMRB flow criteria. Considering the nature and size of the model this level of validation is thought to be robust and acceptable.
NSTM - LMVR Highway Model Version 1.0 39
Table 6.1: Validation results
AM PEAK
Flow Criteria Number of Observed
Counts
Number of Modelled Counts
Acceptability Guideline
Observed flow < 700 vph
Modelled flow within ±100 vph 55 42 76%
Observed flow 700 - 2,700 vph
Modelled flow within ±15% 41 28 68%
Observed flow > 2,700 vph
Modelled flow within ±400 vph 3 1 33%
TOTAL FLOWS 99 71 72% GEH for individual links < 5 99 69 70%
PM PEAK
Flow Criteria Number of Observed
Counts
Number of Modelled Counts
Acceptability Guideline
Observed flow < 700 vph
Modelled flow within ±100 vph 55 43 76%
Observed flow 700 - 2,700 vph
Modelled flow within ±15% 40 31 78%
Observed flow > 2,700 vph
Modelled flow within ±400 vph 4 2 50%
TOTAL FLOWS 99 76 77% GEH for individual links < 5 99 70 71%
6.3.3 The results show that 72% of counts meet the flow validation requirements, and 70% meet the GEH requirements in the AM peak. In the PM peak, 77% of counts meet the flow validation requirements and 71% meet the GEH requirements. Although these results are good for a strategic model of this size, they do not meet the requirements set out in the DMRB. This is a problem acknowledged by the DMRB in Volume 12, section 11.1.7, which states:
“In theory, the adjustment of a model subsequent to a validation exercise to take account of independent observations will improve the model calibration but the model is no longer independently validated. In practice, if a model has performed adequately at validation, then the independent data may be absorbed to achieve an improved calibration. This should, however, be a considered decision taken after validation has been completed (and reported as in section 11.4). Under these circumstances an addendum to the calibration report should be prepared to document the fitting to the enlarged data base and to catalogue the effects of the recalibration”
40 NSTM - LMVR Highway Model Version 1.0
6.3.4 In order to improve the validation of the model, all counts were therefore incorporated into a single set for matrix estimation, as reported in Section 4.9.5. The same prior matrix was used to create the final matrix. The results of the combined link flow validation dataset are shown below.
Table 6.2: Combined Dataset Link Validation
AM PEAK
Flow Criteria Number of Observed
Counts
Number of Modelled Counts
Acceptability Guideline
Observed flow < 700 vph
Modelled flow within ±100 vph 302 273 90%
Observed flow 700 - 2,700 vph
Modelled flow within ±15% 183 163 89%
Observed flow > 2,700 vph
Modelled flow within ±400 vph 18 16 89%
TOTAL FLOWS 503 452 90% GEH for individual links < 5 503 439 87%
PM PEAK
Flow Criteria Number of Observed
Counts
Number of Modelled Counts
Acceptability Guideline
Observed flow < 700 vph
Modelled flow within ±100 vph 301 270 90%
Observed flow 700 - 2,700 vph
Modelled flow within ±15% 184 173 94%
Observed flow > 2,700 vph
Modelled flow within ±400 vph 18 15 83%
TOTAL FLOWS 503 458 91% GEH for individual links < 5 503 450 89%
6.3.5 The above results show that combining the datasets results in successful matching to the DMRB criteria. 90% of links in the AM peak model meet the link flow validation criteria, and 87% meet the GEH criteria. In the PM peak, 91% of counts meet the flow criteria and 89% meet the GEH criteria. The model should therefore be considered an accurate representation of base conditions. Full details of the updated link validation are given in Appendix No. 7 and Appendix No. 8.
6.4 JOURNEY TIME VALIDATION
6.4.1 Journey time surveys were undertaken in both directions for 21 routes to assist with the validation of the transport model. These are shown on Figure 6.1. The surveys were conducted using the “moving observer” method, where a set route was driven with a GPS receiver set to record the vehicle’s position every three seconds.
NSTM - LMVR Highway Model Version 1.0 41
Each vehicle was driven at the speed of general traffic flow while also observing speed limits in order to ensure journey times were representative of typical traffic conditions.
6.4.2 The journey time surveys were carried out in such a way to try and ensure that at least six observations in each direction during each time period were taken. The surveys were carried out between 4th-18th November for the 07:00-10:00 and 16:00-19:00 time periods.
6.4.3 Table 6.3 shows the level of accuracy observed for each of the routes, both in the AM Peak and PM Peak modelled periods. Table 6.3: Observed Journey Time Statistics – AM Peak and PM Peak
Route & Time Period Surveyed
Number of Observations
Observed Mean Journey Time
(Seconds)
Standard Deviation (seconds)
Accuracy (95% Confidence
Level) Route 4 - AM EB 6 640 116 19.02
Route 4 - AM WB 6 536 35 6.85
Route 4 - PM EB 6 631 80 13.31
Route 4 - PM WB 5 536 27 6.25
Route 7 - AM CW 6 952 24 2.65
Route 7 - AM ACW 6 992 80 8.46
Route 7 - PM CW 5 964 56 7.21
Route 7 - PM ACW 5 1002 75 9.29
Route 10 - AM EB 6 903 114 13.25
Route 10 - AM WB 5 860 6 0.87
Route 10 - PM EB 6 867 50 6.05
Route 10 - PM WB 6 903 74 8.60
Route 12 - AM CW 7 788 20 2.35
Route 12 - AM ACW 6 763 30 4.13
Route 12 - PM CW 6 800 39 5.12
Route 12 - PM ACW 6 784 51 6.83
Route 16 - AM NB 6 666 25 3.94
Route 16 - AM SB 6 725 59 8.54
Route 16 - PM NB 6 710 27 3.99
Route 16 - PM SB 6 768 35 4.78
Route 17 - AM EB 15 286 20 3.87
Route 17 - AM WB 16 291 19 3.48
Route 17 - PM EB 16 283 14 2.64
Route 17 - PM WB 13 289 25 5.23
Route 18 - AM EB 18 110 8 3.62
42 NSTM - LMVR Highway Model Version 1.0
Route & Time Period Surveyed
Number of Observations
Observed Mean Journey Time
(Seconds)
Standard Deviation (seconds)
Accuracy (95% Confidence
Level)
Route 18 - AM WB 18 118 10 4.21
Route 18 - PM EB 18 114 8 3.49
Route 18 - PM WB 18 119 7 2.93
Route 23 - AM EB 6 1410 178 13.25
Route 23 - AM WB 6 1169 104 9.34
Route 23 - PM EB 6 1420 92 6.80
Route 23 - PM WB 6 1309 48 3.85
Route 24 - AM NB 6 775 142 19.23
Route 24 - AM SB 6 738 152 21.61
Route 24 - PM NB 6 767 59 8.07
Route 24 - PM SB 6 743 127 17.94
Route 26 - AM NB 20 426 101 11.10
Route 26 - AM SB 20 470 118 11.75
Route 26 - PM NB 20 417 57 6.40
Route 26 - PM SB 20 362 26 3.36
Route 27 - AM NB 12 274 14 3.25
Route 27 - AM SB 12 304 41 8.57
Route 27 - PM NB 11 287 8 1.87
Route 27 - PM SB 9 294 23 6.01
Route 29 - AM CW 7 701 33 4.35
Route 29 - AM ACW 7 686 35 4.72
Route 29 - PM CW 7 711 14 1.82
Route 29 - PM ACW 7 703 37 4.87
Route 30 - AM EB 6 764 124 17.03
Route 30 - AM WB 5 760 59 9.64
Route 30 - PM EB 6 883 221 26.27
Route 30 - PM WB 5 774 48 7.70
Route 31 - AM NEB 6 865 65 7.89
Route 31 - AM SWB 6 834 27 3.40
Route 31 - PM NEB 5 869 39 5.57
Route 31 - PM SWB 5 896 46 6.37
Route 33 - AM EB 8 700 90 10.75
Route 33 - AM WB 7 664 27 3.76
Route 33 - PM EB 6 672 16 2.50
Route 33 - PM WB 5 757 59 9.68
NSTM - LMVR Highway Model Version 1.0 43
Route & Time Period Surveyed
Number of Observations
Observed Mean Journey Time
(Seconds)
Standard Deviation (seconds)
Accuracy (95% Confidence
Level)
Route 34 - AM NEB 6 987 231 24.56
Route 34 - AM SWB 6 979 199 21.33
Route 34 - PM NEB 5 1266 196 19.22
Route 34 - PM SWB 5 1174 95 10.05
Route 35 - AM NEB 7 1946 253 12.02
Route 35 - AM SWB 6 1977 84 4.46
Route 35 - PM NEB 6 2196 190 9.08
Route 35 - PM SWB 6 2063 178 9.05
Route 36 - AM CW 8 1269 304 20.03
Route 36 - AM ACW 8 1128 158 11.71
Route 36 - PM CW 7 1442 224 14.37
Route 36 - PM ACW 5 1273 95 9.27
Route 37 - AM CW 7 777 65 7.74
Route 37 - AM ACW 7 573 51 8.23
Route 37 - PM CW 8 705 43 5.10
Route 37 - PM ACW 8 609 103 14.14
Route 38 - AM CW 6 1047 110 11.03
Route 38 - AM ACW 6 879 209 24.95
Route 38 - PM CW 6 1026 68 6.96
Route 38 - PM ACW 6 858 149 18.22
Route 39 - AM NB 6 1718 212 12.95
Route 39 - AM SB 6 1759 255 15.21
Route 39 - PM NB 6 1867 295 16.58
Route 39 - PM SB 6 1954 226 12.14
6.4.4 DMRB Volume 13 recommends that the sampled journey times should have an accuracy of 10%. This means that, at the 95% confidence level, the mean observed journey time is within ±10% of the true mean journey time. Of the 84 directional routes surveyed (across all time periods), 56 routes (67%) were within the criteria, 14 (17%) were just outside of the criteria (with an accuracy of between 10-15%) while 14 routes (17%) had accuracies of greater than 15%.
6.4.5 The 2008 observed journey times have been compared against the 2008 modelled journey times and the results have been included as Appendix No 9 and Appendix No 10 for AM and PM peak respectively in graphical form. The DMRB Volume 12, Section 2, 4.4.43 suggests that total modelled journey time should be within 15% (or 1 minute, whichever is greater) of the observed average journey time.
44 NSTM - LMVR Highway Model Version 1.0
Table 6.4 and Table 6.5 compare the accuracy of the modelled and observed journey times in the AM Peak and PM Peak respectively. Table 6.4: Journey Time Summary Results – AM Peak
6.4.6 The journey time results for the AM Peak period show that 39 out of the total 42 counts (93%) of all the journey time runs are within 15% (or 1 minute) of the observed time. The graphs for the AM Peak show that any queues and therefore delays which would have been observed have been reflected within the modelled journey time runs.
6.4.7 The journey time results for the PM Peak period show that 36 out of 42 (or 86%) of all the journey time runs are within 15% (or 1 minute) of the observed time. The graphs for the PM Peak all show that any queues and therefore delays which would have been observed have been reflected within the modelled journey time runs.
6.4.8 Although the above results indicate a good fit between observed and modelled journey times, it should be noted many of the observed journey times show a wide degree of variation. Consequently, it is likely that some of the modelled journey times that do not meet the validation requirements are actually a good representation of typical journey times. However, further journey time surveys would be required on those routes with a high variance.
NSTM - LMVR Highway Model Version 1.0 47
6.5 SCREENLINES
6.5.1 Figure 6.2 shows the screenlines that have been used in the validation exercise. DMRB validation standards require most screenlines to show total modelled flows within ±5% of observed flows, and the screenline GEH should be less than 4. Table 6.6 shows the screenline totals for the AM peak. Table 6.6: AM Peak Screenline Validation
Screenline Direction Observed Modelled GEH
Town Centre In 9979 8938 10.707
Out 6890 6298 7.284 Outer
Northampton In 9671 9327 3.531
Out 8319 7423 10.108
North-South EB 6837 6599 2.91 WB 7815 7304 5.872
East-West NB 7234 6827 4.845 SB 7853 7925 0.815
Corby In 3982 3755 3.658
Out 4134 3841 4.64
Kettering In 5788 5684 1.375
Out 5037 4822 3.064
Wellingborough In 4966 4809 2.246
Out 3704 3484 3.662
6.5.2 The above results show that all but 6 screenlines in the AM peak fulfil the GEH validation criteria, indicating that all of the key strategic movements are generally successfully replicated by the NSTM base model. The remaining screenlines all have a total flow within 5% of observed, indicating that they are still a close match.
6.5.3 Table 6.7 shows the screenline validation results for the PM peak period. Table 6.7: PM Peak Screenline Validation
Screenline Direction Observed Modelled GEH
Town Centre In 8299 7969 3.661
Out 9495 8804 7.223 Outer
Northampton In 7828 7491 3.85
Out 9800 9137 6.81
North-South EB 7426 7150 3.239 WB 7076 6688 4.676
East-West NB 7786 7524 2.99 SB 7658 7324 3.851
Corby In 4112 3965 2.311
Out 3907 3704 3.29
Kettering In 5883 5598 3.76
Out 5057 5088 0.425
Wellingborough In 3910 3949 0.612
Out 4996 5169 2.424
48 NSTM - LMVR Highway Model Version 1.0
6.5.4 In the PM peak, all but 3 screenlines meet the GEH criteria. The outbound Town Centre screenline and the outbound Outer Northampton show significant deviation, with a GEH of 7.2 and 6.8 respectively. However, this is still within 8% of observed flows, indicating that overall traffic flows are well-represented.
6.5.5 Overall, the NSTM validates well for screenlines capturing all of the significant strategic movements within the study area.
7.1.1 A Highway Model has been developed using SATURN as a foundation for forecasting the effects and impacts of the proposed developments in and around Northamptonshire. The use of the SATURN modelling software with its detailed junction delay capabilities enables the accurate representation of the road network/junctions. This allows as accurate representation of the flows, speeds and delays on the road network as possible. This information has been used within the Demand Model and the Public Transport Model to allow their calibration and validation. The modelled modes included in the Highway assignment model include:
Lights Matrix (Car/LGV)
Heavy Matrix (HGV)
7.1.2 The models have been developed in accordance with DMRB and WebTAG standards and represent a weekday AM Peak (08:00-09:00) and PM Peak (17:00-18:00) hour in a typical neutral month in 2008.
7.1.3 Data collected for the Highway Model development includes comprehensive ATC surveys to allow the factoring of RSI trip distribution patterns to a 2008 Base Year, 2001 census data for journey to work trip distribution patterns for all modes, Manual Classified Link Counts and journey time surveys. This data has been used to develop the 2008 Base Year Highway Model in accordance with DMRB and WebTAG standards.
7.1.4 The Demand Model uses travel costs taken from the Highway Model, which is deemed more accurate in assignment than the Demand Model. These costs are used to generate the demand model matrices, elements of which are passed into the Highway Model. In turn new costs are calculated and so on until model validation is achieved.
7.2 MODEL VALIDATION / CALIBRATION
7.2.1 The robustness of the observed Highway Model as a forecasting tool has been measured by comparing link flows and journey times against observations. The comparisons were benchmarked against DMRB calibration and validation standards and have been found to be a suitable base for supplying information for the Demand Model.
7.3 MODEL APPROVAL & FIT FOR PURPOSE
The technical information presented in this report demonstrates that the NSTM is an accurate representation of base year traffic flows. As such, it is considered fit for the purposes of forecasting and scheme assessment.
50 NSTM - LMVR Highway Model Version 1.0
APPENDIX NO 1: Zone Correspondence Table
Sector Zone No. Sector Name District
18 1 Brackley South Northants18 2 Brackley South Northants18 3 Brackley South Northants18 4 Brackley South Northants18 5 Brackley South Northants18 6 Brackley South Northants18 7 Brackley South Northants18 8 Brackley South Northants18 9 Brackley South Northants18 10 Brackley South Northants18 11 Brackley South Northants18 12 Brackley South Northants18 13 Brackley South Northants18 14 Brackley South Northants18 15 Brackley South Northants18 16 Brackley South Northants18 17 Brackley South Northants18 18 Brackley South Northants18 19 Brackley South Northants18 20 Brackley Daventry18 21 Brackley Daventry18 22 Brackley Daventry15 23 Daventry Daventry15 24 Daventry Daventry15 25 Daventry Daventry15 26 Daventry Daventry15 27 Daventry Daventry15 28 Daventry Daventry15 29 Daventry Daventry15 30 Daventry Daventry15 31 Daventry Daventry15 32 Daventry Daventry15 33 Daventry Daventry15 34 Daventry Daventry15 35 Daventry Daventry15 36 Daventry Daventry15 37 Daventry Daventry15 38 Daventry Daventry15 39 Daventry Daventry15 40 Daventry Daventry15 41 Daventry Daventry15 42 Daventry Daventry
9 81 Corby Corby9 82 Corby East Northants9 83 Corby East Northants10 84 Oundle East Northants10 85 Oundle East Northants10 86 Oundle East Northants10 87 Oundle East Northants10 88 Oundle East Northants10 89 Oundle East Northants10 90 Oundle East Northants10 91 Oundle East Northants10 92 Oundle East Northants10 93 Oundle East Northants10 94 Oundle East Northants
189 Calibration 8 North of Sudborough EB Jun-05 ATC 497160 282190 24101 25161 250 488 12.388 No No190 Calibration 8 North of Sudborough WB Jun-05 ATC 497160 282190 25161 24101 388 271 6.412 No No
191 Calibration 1 Pytchley Lane, South of A14 NB Jun-05 ATC 478920 282990 10253 10711 143 128 1.283 Yes Yes
192 Calibration 1 Pytchley Lane, South of A14 SB Jun-05 ATC 478920 282990 10711 10253 105 104 0.093 Yes Yes
193 Calibration 2 Isham Rd, South of A14 EB Jun-05 ATC 485996 274743 10254 12708 158 160 0.122 Yes Yes194 Calibration 2 Isham Rd, South of A14 WB Jun-05 ATC 485996 274743 12708 10254 157 113 3.814 Yes Yes195 Calibration 4 Kettering Rd, South of A14 NB Jun-05 ATC 489697 275605 12058 12724 459 408 2.438 Yes Yes196 Calibration 4 Kettering Rd, South of A14 SB Jun-05 ATC 489697 275605 12724 12058 471 429 1.988 Yes Yes
197 Calibration 7 Irthlingborough Rd, Southeast of Finedon NB Jun-05 ATC 493860 271370 10416 12123 790 766 0.859 Yes Yes
NSTM Link Flows
ID Calibration / Validation Site No. Site Location Dir Date Data Type X Y A-Node B-Node C-Node
28015 – NORTHAMPTON SOUTH SUE (MAPLE FARM COLLINGTREE DEVELOPMENT)
NSTM TRANSPORT MODELLING SCOPING - DO MINIMUM AND DO SOMETHING
TESTS - ISSUE 4 – June 18 th, 2013
1 Introduction 1.1 This Scoping Note has been prepared to set out the scope of the proposed runs of
the Northamptonshire Strategic Transport Model (NSTM) to support transport technical work for the forthcoming outline planning application for the Northampton South SUE Development (NSSUE – to be progressed as the Maple Farm Development, Collingtree). It will be updated as further information is sought / made available to contain a current summary of the work instruction.
1.2 Maple Farm Collingtree, the NSSUE, is one of the sites identified within the West
Northamptonshire Pre Submission Joint Core Strategy for residential development of 1,000 units, and forms the strategic urban extension for South Northampton. It is located on land to the north of the M1, west of the A45, east of Towcester Road and the rail line, and south of Rowtree Road. It is being promoted by Bovis Homes Ltd.
1.3 Peter Brett Associates LLP has been appointed by Bovis Homes Ltd to resolve all transport and access matters, to enable an outline planning application to be submitted in Spring 2013 supported with all necessary transport work.
1.4 Peter Brett Associates has undertaken initial discussions with Northamptonshire
County Council (NCC), the Highways Agency and MGWSP Northamptonshire Highways to work together to set the context for this work; the Notes of the relevant meeting held on Friday 25th January 2013 are attached in Appendix 1.
1.5 Further clarity was sought by Northamptonshire Highways by email on April 8th. This email, and Peter Brett Associates’ response is included in Appendix 1. This resulted in Issue 3.
1.6 Following discussions with Northamptonshire Highways seeking further clarity to the on-site infrastructure and trip rate generation in May / June, a further version of this Note has been prepared. This is referred to as Issue 4. Relevant correspondence is also included in Appendix 1.
1.7 It is acknowledged that further iterations of this instruction may be required following the discussions with NCC and MGWSP Northamptonshire Highways. Similarly, further iterations of the model are likely - with differing levels of infrastructure and development to respond to issues identified in the initial testing.
1.8 Issue 1 of this Note contained details of the NSSUE and the associated highway
infrastructure within this document along with the proposed work stages. Issue 1 of the Scoping Note was issued to: i) seek agreement from NCC / MGWSP Northamptonshire Highways to the
proposed scope of works;
ii) inform NCC / MGWSP Northamptonshire Highways of the likely future work commitments;
iii) to seek an initial fee proposal from NCC and MGWSP Northamptonshire Highways for this work;
iv) to clarify work timescales; and v) to obtain client confirmation to this process.
1.9 Issue 2 of the Scoping Note contained: i) the updated development masterplan;
ii) a minor amendment to the the land-use budget to reflect the available
information;
iii) further clarity on the development junction and access arrangements; iv) the output from the junction capacity assessments for the site accesses and
local junctions; v) further clarity on how the TEMPRO growth factors are to be amended and
applied. 1.10 Issue 3 of the Scoping Note contained:
i) In Appendix 7 (now 8), details of the vehicle trip generation from this
development to be included in the NSTM NSSUE Development option tests (para 5.3 (x));
ii) inclusion of a request for 2012 Base traffic flows (para 5.3 (xv)); iii) inclusion of a request that alternative Future Year Mitigated Do Something
traffic flows be allowed for at a later date (para 5.3 (xv)); iv) identifying that the future Mitigated Do Something work would include
modelling of the mode shift from single occupancy vehicles at a later date (para 5.3 (xii)).
1.11 Issue 4 of the Scoping Note contains:
i) a Node - Link diagram (Appendix 3);
ii) revised Rowtree Road Roundabout Site Access plan (para 3.3(i)), a further
design evolution;
iii) revised Windingbrook Lane Site Access Priority Junction plan (para 3.3(ii)); iv) internal link road network design speed, amended to reflect NCC’s comments
(para 3.3(iii)); v) clarity of the Phase 1 link road network (para 3.4); vi) clarity of the Phase 1 modelling sought, and renumbering of Section 5 to
provide easier reference;
vii) revised Appendix 8, a spreadsheet containing the details of the vehicle trip generation from this development to be included in the NSTM NSSUE Development option tests (para 5.3 (x));
viii) definition of the area for which model information is sought in Appendix 9.
2 NSSUE Development details 2.1 The details of the NSSUE Phase 1 and Full Development are summarised in Table 1.
2.2 The Working Draft Development Framework Plan prepared by David Lock Associates
(reference BHL/005 D) is enclosed as Appendix 2, but will be the subject of evolution and amendment as the masterplan develops.
Deleted: s
Deleted: a copy of Transport Technical Note 2 – containing
Land-use Size (m2) / Units Phase 1 – complete in 2021 – forming Zone 1 Housing – Zone 1
400 units
Development – complete in 2026
Housing Zone 1 Zone 2
1,000 units 400 units 600 units
Local Centre – All Zone 2 B1(a) Employment – 3,000m 2 Local Centre Retail – 1,350m 2 Community facilities – 550m 2
Primary School – All Zone 2 2 Form Entry - 420 pupils 2.3 The main points of vehicular access to the Development will be formed from Rowtree
Road and Windingbrook Lane.
2.4 Bovis Homes Ltd is seeking to complete this work to enable the planning application to be submitted in Spring 2013.
3 Modelling of the NSSUE within the current iteration of the NSTM model
3.1 The NSTM currently includes the following for the NSSUE quanta in Zone 668, the Core Strategy Site “Northampton South”: i) 2016 – 400 dwellings, 0 Employees, 0 School places, 0 Retail;
The 2026 Zone 668 Development quanta reflects the current development proposals, albeit the phasing of the development implementation needs to be amended to support any application.
3.2 Zone 668 has two connections to the highway network, as shown below in the extract from the NSTM:
i) to the north - to Rowtree Road, at the junction of Penvale Road; and
ii) to the south – to Ash Lane, the Collingtree / Milton Malsor road.
3.3 The NSTM NSSUE primary link road arrangement for the Full Development is to be
as follows – see Appendix 3 for an indicative Node – Link diagram: i) to provide a worst case for any junction capacity assessment at this location,
a new four-arm roundabout, replacing the existing three arm Rowtree Road / Lichfield Road priority junction. This is shown on PBA drawing 28015/002 contained in Appendix 4;
ii) a new priority junction on Windingbrook Lane shown on PBA drawing 28015/001 contained in Appendix 4;
iii) the alignment of the internal primary link road network is to be modelled as
shown on the masterplan in Appendix 2. For the purposes of the modelling this road is assumed to be 6.5m, and designed to 30mph;
iv) the Zone 1 and Zone 2 residential development zone loaders are to be
connected to the primary link road / secondary link road junctions shown on the masterplan;
v) the existing vehicular link to the south towards Ash Lane in Collingtree village shown in the model is to be deleted.
3.4 The NSTM NSSUE primary link road arrangement for the Phase 1 Development (the
area identified as “1” to the south-east of the site shown on the plan in Appendix 2, consisting of 400 units) is to be as follows:
i) Phase 1 will be accessed only by the Windingbrook Lane priority junction
shown on PBA drawing 28015/001 contained in Appendix 4 - the Rowtree Rd Site Access Rbt will not be provided as part of Phase 1;
ii) the alignment of the internal primary link road network is to be modelled as
shown on the masterplan in Appendix 2. The internal site road for Phase 1 will be provided from the Windingbrook Lane access to the completion of the internal loop – it will not continue past the public footpath until Phase 2. For the purposes of the modelling this road is assumed to be 6.5m, and designed to 30mph;
iii) the Phase 1 residential development zone loaders are to be connected to the
primary link road / secondary link road junctions shown on the masterplan;
Deleted: 1/001
Deleted: 3
Deleted: four arm roundabout
Deleted: (the two new arms being the site access roads – the two roads being implemented to provide security of access to the new development). This is
iv) the existing vehicular link to the south towards Ash Lane in Collingtree village shown in the model is to be deleted.
3.5 The junction figure plans, contained in Appendix 4, are work-in-progress, and
represent the best available information at this stage. These have been submitted for the purposes of the modelling only.
3.6 Output from the JUNCTIONS8 ARCADY junction capacity assessments for these two junctions are contained in Appendix 5.
4 Work Stages undertaken
4.1 At the Transport Scoping Meeting with the Highways Agency and Northamptonshire County Council, Peter Brett Associates agreed to prepare Technical Note 1 to inform the discussions over the extent of the of the Transport Assessment Study Area. This Note is contained in Appendix 6.
4.2 The assignment of the trips within Technical Note 1, hence the Study Area, was informed by a Select Zone Analysis of the NSSUE zone from the Northamptonshire Strategic Transport SATURN model for both the AM and PM peaks, as provided by NCC / MGWSP Northamptonshire Highways.
4.3 Technical Note 1 summarised existing traffic count data sources, and informed the need and extent of further traffic count data. All of this traffic count data will be made available to NCC and their consultants for the purposes of the NSTM modelling.
5 Future work to be undertaken
5.1 To enable the application of the NSTM to support transport assessment work and planning applications for individual developments within the County, the following work stages have been identified.
5.2 We would be grateful for costs from NCC and MGWSP for the following model work. We require this work to be undertaken as soon as possible.
5.3 In the interests of expediency, further details of some of the work elements may need to follow at a later stage.
5.4 Base Model 2012
i) local validation of the Base Model by NCC / MGWSP Northamptonshire Highways using the traffic count data obtained or commissioned by the client. Any necessary amendments to the Base Model to be undertaken by NCC / MGWSP Northamptonshire Highways;
ii) provision of a written report of the validation exercise by NCC / MGWSP Northamptonshire Highways to Northamptonshire County Council, the Highways Agency and Peter Brett Associates LLP, and confirmation that the validation is acceptable;
iii) output from the ARCADY and PICADY junction capacity assessments for local
junctions has been requested by MGWSP Northamptonshire Highways. This is included in Appendix 7;
5.5 2026 Future Year Do Minimum Model
i) in the 2026 Do Minimum test, the following numbers of houses allocated to all the seven Northampton Related Development Area SUE are to be deducted from the TEMPRO total of 50,193 for 2026:
Development: Quanta Northampton South SUE 1,000 Northampton Upton Park SUE 979 Northampton Kings Heath SUE – Dallington Grange 3,000 Northampton North – SUE 2,000 Northampton West – SUE 1,500 Northampton South of Brackmills SUE 1,000 Northampton – North of Whitehills SUE 1,000 Remaining total across West Northamptonshire 39,714
ii) no further allowance should be made for the Northampton South SUE in the Do
Minimum assessment;
iii) vehicle flows associated with the six other SUE households (a total of 9,479 households - ie, not the NSSUE) within the Northampton Related Development Area are to be loaded onto the network at the approximate location of these developments;
iv) unless Northamptonshire Highways has better information relating to the location
of the remaining development, the remaining general growth in household numbers for Northampton is to be applied across the town, the development associated with SNDC, DDC etc is to be focussed at these areas. The total households in the West Northamptonshire Area in the Do Minimum test is to be set at 49,193.
v) confirmation by NCC / MGWSP Northamptonshire Highways that the modelled
surrounding transport network infrastructure and transport enhancements by all parties - the HA, NCC and other developers - reflects the current position, and if necessary, updating the model;
5.6 2026 Future Year Do Something Model
i) NCC / MGWSP Northamptonshire Highways to amend the model network to
reflect the NSSUE proposed infrastructure. This includes dividing the NSSUE zone into 2 zones to reflect the proposed phasing with the necessary zone connectors – see Section 3;
ii) Peter Brett Associates has reviewed the external vehicle trip generation of the
Full Development NSSUE development. The output from the spreadsheet is contained in Appendix 8. The person trip assessment is informed by national trip generation and mode share data, and reflects the development location, development land use areas and mix, and proposed transport linkages and provision. This was submitted for agreement with NCC and MGWSP Northamptonshire Highways to Peter Brett Associates’ analysis of the NSSUE Full Development vehicle trip generation movements;
iii) substitution by NCC / MGWSP Northamptonshire Highways of the Full
Development vehicle flows contained within the NSTM model for the vehicle trip generation;
5.7 2026 Future Year Mitigated Do Something
i) agreement between NCC, MGWSP Northamptonshire Highways and Peter Brett Associates to the necessary infrastructure mitigation measures. For the purposes of establishing NCC / MGWSP Northamptonshire Highways’ fees, it is to be assumed that this would include mode shift from the number of Base vehicle trips, as well as enhancements to be coded at eight junctions;
ii) re-run of the model by NCC / MGWSP Northamptonshire Highways incorporating
these changes.
Deleted: Phase 1 and
Deleted: , prepared with reference to a person trip assessment and appropriate mode shift analysis for the development
i) as per the 2026 Do Minimum test, the numbers of houses allocated to the seven Northampton Related Development Area SUE in the 2021 Do Minimum test are to be deducted from the TEMPRO total;
ii) no further allowance should be made for the Northampton South SUE in the Do Minimum assessment;
iii) vehicle flows associated with the six other SUE households within the Northampton Related Development Area are to be loaded onto the network at the approximate location of these developments;
iv) unless Northamptonshire Highways has better information relating to the location
of the remaining development, the remaining general growth in household numbers for Northampton is to be applied across the town, the development associated with SNDC, DDC etc is to be focussed at these areas;
v) confirmation by NCC / MGWSP Northamptonshire Highways that the modelled
surrounding transport network infrastructure and transport enhancements by all parties - the HA, NCC and other developers - reflects the current position, and if necessary, updating the model;
5.9 2021 Future Year Do Something Model
i) NCC / MGWSP Northamptonshire Highways to amend the model network to
reflect the NSSUE proposed infrastructure. This includes modelling the NSSUE Phase 1 to reflect the proposed phasing with the necessary zone connectors – see Section 3;
ii) Peter Brett Associates has reviewed the external vehicle trip generation of the
Phase 1 NSSUE development. This assessment is included in Appendix 8. The person trip assessment is informed by national trip generation and mode share data, and reflects the development location, development land use areas and mix, and proposed transport linkages and provision. This was submitted for agreement with NCC and MGWSP Northamptonshire Highways to Peter Brett Associates’ analysis of the NSSUE Phase 1 vehicle trip generation movements;
iii) substitution by NCC / MGWSP Northamptonshire Highways of the Phase 1
vehicle flows contained within the NSTM model for the vehicle trip generation;
5.10 Data to be provided i) as well as the standard Forecasting Modelling reports, Peter Brett Associates’
initial requirements for information to be provided by the NSTM are as follows -we may need to specify further work once we understand the overall initial impacts;
ii) provision of initial model output information by NCC / MGWSP Northamptonshire
Highways across the area shown in Appendix 9 for:
• 2012 Base
• 2021 Do Minimum; • 2021 Do Something - Phase 1; • 2026 Do Minimum; and • 2026 Do Something - Full Development
iii) for all option runs, PDF copies of the following plots, showing i) figures and (ii)
figures and bandwidths:
• actual AM / PM link flows for this area;
• AM / PM v/c plots for this area;
• AM / PM queue plots for this area;
• AM / PM delay plots for this area;
• AM / PM plot of the select zone analysis origin and destination flows for the NSSUE Development through the same area, assuming Actual flows;
• full Junction Turning Movements (AM / PM) (identifying the origin / destination of all movements through up to 15 junctions) including:
- Rowtree Road Site Access; - Windingbrook Lane Site Access; - Wooldale Road / Berry Lane Roundabout; - Bridge Meadow Way / A45 southbound Off Slip / southbound On
5.11 After a review of the results of the initial model tests, further tests are likely to commissioned at a later stage, but will be discussed and agreed with the highway authorities in advance.
6 Programme
6.1 A draft programme is requested from NCC and MGWSP Northamptonshire Highways
to enable the client to comment and agree to the proposed programme.
APPENDIX 1 – NOTES OF SCOPING MEETING – FRIDAY 25TH JANUARY 2013 AND SUBSEQUENT LIAISON
APPENDIX 2 – DRAFT DEVELOPMENT FRAMEWORK PLAN APPENDIX 3 - NODE – LINK DIAGRAM APPENDIX 4 – PLANS OF THE SITE ACCESS JUNCTIONS APPENDIX 5 – ARCADY / PICADY OUTPUTS FOR THE SITE ACCESS JUNCTIONS APPENDIX 6 – PBA TRANSPORT TECHNICAL NOTE 1 - NSTM ASSIGNMENT AND TRAFFIC SURVEY STRATEGY APPENDIX 7 – ARCADY / PICADY OUTPUTS FOR THE LOCAL JUNCTIONS APPENDIX 8 – EXTRACTS FROM THE VEHICLE TRIP CALCULATION SPREADSHEET APPENDIX 9 – MODEL AREA FROM WHICH INFORMATION IS SOUGHT
28015/NTN/RSH/SND/JPH June 18th, 2013
Deleted: 3
Deleted: 4
Deleted: 5
Deleted: 6
Deleted: 7
Deleted: TRANSPORT TECHNICAL NOTE 2 – PERSON TRIP ASSESSMENT
APPENDIX 1 – NOTES OF SCOPING MEETING - FRIDAY 25TH JANUARY 2013 AND SUBSEQUENT LIAISON Meeting Title: Maple Farm Development, Collingtree (the Northampton South SUE), Northampton Pre-Application Transport Scoping Consultation Meeting Attendees: Rob Sim Jones - Northamptonshire County Council (NCC) Lou Mason Walsh - MG-WSP Northamptonshire Highways (MG-WSP) Jonathan Price - Highways Agency (HA) Simon Davis - Peter Brett Associates (PBA) John Hopkins - Peter Brett Associates Date of Meeting: Friday 25th January, 2013 - 15.00 – 16.00 Location: Riverside House, Bedford Road, Northampton Job Number: 28015
Item Subject
1 1.1 1.2
Introduction. This meeting was arranged to discuss the scoping for the transport assessment work necessary to support the emerging Northamptonshire South SUE (Maple Farm, Collingtree) development through the planning process, leading to a planning application. MG-WSP requested the Site be referred to as the “NSSUE” for clarity.
2 2.1 2.2 2.3
Details of the Maple Farm Collingtree Development Maple Farm, a residential development of 1,000 units, forms the strategic urban extension for South Northampton. It is located on land to the north of the M1, west of the A45, east of Towcester Road and the rail line, and south of Rowtree Road. It is being promoted by Bovis Homes Ltd. Post Meeting Note – the masterplan quantum referred to during the meeting has been amended following a review of the masterplan to that within the West Northants Pre Submission Joint Core Strategy.To avoid subsequent confusion, no references are made to the previous proposals discussed at this meeting. Bovis Homes Ltd‘s preference is to submit a planning application for the end of March 2013. NCC cautioned that it would be unlikely that sufficient progress would have been made with the necessary supporting NSTM and transport assessment work to achieve this deadline. A first phase of development is being proposed. A further assessment, based on an earlier timescale, will also be submitted.
3 3.1 3.2 3.3
HA comments - A45 proposals and the Northampton Growth Management Strategy. The HA has developed, jointly with the other authorities in the area, the Northampton Growth Management Strategy, to provide the vehicle for ringmaster contributions to be accepted from the SUE to fund the necessary trunk road highway infrastructure to mitigate all strategic development. This NGMS work was supported by DIAMOND modelling of Northampton, incorporating the West Northants Pre-Submission Joint Core Strategy document development quanta. The NSSUE was modelled as a 1,000 unit development. A series of junction and link enhancements and measures were considered based on this modelling, and a per capita contribution will be sought. HA confirmed that once the appropriate contributions were paid for the NGMS that development could proceed without hinderance from them. All authorities that sign up to the NGMS Memorandum of Understanding can hold contributions as ring-master
(including Northampton Borough Council), the County Council would deliver the works. Should alternative development quanta be sought by a SUE promoter, further runs of the DIAMOND modelling would be required to reassess the impact, and confirm that increased contributions are not required. With the currently proposed development quanta (1,000 units), this is not applicable.
4. 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11
NCC comments NCC agreed that the Northamptonshire Strategic Transport Model (NSTM) is an appropriate tool to assess the development vehicle impact on local links, and to confirm cumulative impacts in the context of committed development. MG-WSP commented that their intial review suggested that the assessment study area would need to include Towcester Road up to the Gas Street Roundabout, possibly around St James and Bedford Road. Before the NSTM can be used to consider the NSSUE, the model would need to be subjected to a local model validation exercise, this process supported with more recent traffic count data. MG-WSP has already provided PBA with select zone analysis output for the NSSUE from the 2026 NSTM model. PBA is to submit a review of this select zone analysis provided by MG-WSP to confirm the extent of the Study Area and the traffic survey requirements, and to ascertain what counts are already held by NCC. NCC will not accept traffic count data collected during half term. PBA to contact NCC prior to commissioning surveys to confirm the programme for future road works in this area.MG-WSP to liaise with their modelling team to ascertain whether ATC or MCC at junctions are required. With respect to the NSSUE trip generation, PBA would need to submit and agree the development external vehicle trip quantum with NCC. These agreed vehicle trips for the development would be fixed in the model, then assigned from the NSSUE zone in the NSTM. These vehicle trips would be established by a person trip analysis prepared by Peter Brett Associates, and would include consideration of internalisation, and appropriate mode share and mode shift for the location, transport infrastructure and development quanta. The NSSUE zone is currently one zone. PBA to discuss with MG-WSP the need to split this into smaller zones better to respond to the proposed land uses. NCC is developing a strategy for Towcester Road and the Southern Inner Ring Road informed with a further, VISSIM, model of the town centre (hence why further survey work may be available in this area). This Study includes consideration of various options to deal with the Gas Street Roundabout and Plough Gyratory. This work is to enable ringmaster contributions to be accepted from the SUE to fund the necessary trunk road highway infrastructure along these corridors in a town-wide manner. The VISSIM model also incorporated the West Northants Pre-Submission Joint Core Strategy document development quanta. MG-WSP confirmed that they would be able to start the model validation process upon receipt of the traffic count data. MG-WSP advised that the 2031 option test is being worked on now, and will be available shortly. Following agreement to the traffic count surveys, further Notes will be issued by PBA to the joint highway authorities to scope the transport assessment work for submission and agreement.
Email correspondence - Northamptonshire Highways and Peter Brett Associates From: John Hopkins Sent: 29 April 2013 18:06 To: 'Mason-Walsh, Lou' ([email protected]) Cc: Ron Henry; Simon Davis; Robert Sim-Jones ([email protected]); '[email protected]' ([email protected]); '[email protected]' ([email protected]) Subject: 28015 Maple Farm, Collingtree : Northampton South SUE Modelling Note Lou, Many thanks for the email regarding the NSTM Scoping Note. For the ease of reading, our responses are embedded within the text of this email in Bold Red font. Within this response, to enable the Do Minimum modelling to proceed we have provided the necessary information in the attached Scoping Note Issue 2, as well as some of the Do Something information. A further Scoping Note Issue 3 will be issued in the next few days with all the information required for the Do Something modelling. As requested, we have provided copies of the Site Access plans for the purposes of modelling – these will be the subject of further review and discussions. Should you need any further assistance, please do not hesitate to contact me. Regards. John Hopkins For and on behalf of Peter Brett Associates LLP From: Mason-Walsh, Lou [mailto:[email protected]] Sent: 08 April 2013 12:36 To: Ron Henry Cc: John Hopkins; Simon Davis; Sim-Jones, Rob; Price, Jonathan ([email protected]); Graham Stevenson; [email protected] Subject: Northampton South SUE Modelling Note Ron, I have reviewed Issue 1 of the modelling note dated the 20th Feb 2013 and have the following comments to make; Section 2 NSSUE Development details In order to test the proposed development in the NSTM we will need details of the land use proposed, the phasing of the proposed development and the trip rates that are to be used in the assessment. The indicative details of the land uses and phasing we re contained in Table 1 of the NSTM Scoping Note. Further clarity has been provided on the Local Centre in Issue 2 of the Scoping Note (attached), as summarised below: Land-use Size (m2) / Units Phase 1 – complete in 2021 Housing
400 units
Full Development – complete in 2026 Housing
1,000 units
Local Centre B1(a) Employment – 3,000m 2 Local Centre Retail – 1,350m 2 Community facilities – 550m 2
Primary School 2 Form Entry - 420 pupils
Deleted: 11 Prospect Court, Courteenhall Road, Blisworth, Northampton, NN7 3DG¶t 01604 878305¶f 01604 878333¶m 07876 576290¶e [email protected]¶w www.peterbrett.com ¶ ¶Roger Tym & Partners and Baker Associates are now part of Peter Brett Associates LLP.¶
The revised Scoping Note Issue 2 contains more details of the land-use proposals, and the development phasing. The Person Trip Analysis, which leads to the Development vehicle trip generation, will be included in the next Issue 3 of the Scoping Note, to be issued within the next few days. Section 3: Modelling of the NSSUE within the current iteration of the NSTM model The modelling of the NSSUE in the current version of the NSTM is in line with the Emerging Core Spatial Strategy for West Northamptonshire and does not include any internal road structure but simply indicates which link the development is expected to access the local road network. To ensure that the modelling reflects the emerging mas terplan, Peter Brett Associates has confirmed within Issue 2 of the Scoping Note where the proposed Development is to be connected to the surrounding highway network – the NSSUE Development link already in the NSTM to the south towards Ash Lane / Collingtree Village is to be removed. Para 3.3: In order to undertake the required testing of the proposed development the current assumptions will be removed and replaced with the site specific trip rates, it will also be necessary to code the proposed internal road network into the NSTM, this should include the assumed link type and junction arrangements. Details of the proposed Development lin k and indicative junction type arrangements are included in the attached Issue 2 of the Scoping Note. Details of the NSSUE vehicle trip generation are contained in Issue 3 of the Scoping Note, shortly to be issued. To aid the modelling of the internal road network we will need to see the junction assessment models so that these can be agreed and the relevant information such as sat flows extracted and input into the NSTM. The Site Access junction assessment ARCADY output file s (obtained from the JUNCTIONS8 suite) are contained in Issue 2 of the Scoping Note. Section 5: Further Work to be undertaken Base Model The revalidation exercise of the NSTM will mean that the base year for the model will be 2012. Noted, thanks. The local model validation exercise includes not just the inclusion of the new count data but also checking of the local road network to ensure that any changes to the local road network since the previous base validation are included. This process will be informed by the individual junction models to be used in the preparation of the TA for the above scheme, these should be submitted as soon as possible to enable the models to be agreed. As requested, the ARCADY and PICADY assessment files for the adjacent junctions are included in Issue 2 of the Scoping Note. Future Year Do Minimum Model The NSTM is constrained to TEMPRO. Noted, thanks. We comment: With respect to the number of households modelled within this area:
i) we agree, the total Do Something number of households within the model should be constrained to that in TEMPRO;
ii) at present, TEMPRO 6.2 assumes growth in the West Northamptonshire area (South Northants, Northampton Borough and Daventry Districts) of 50,193 households between 2001 and 2026. This is essentially the same as the housing projections that support the Pre-Submission West Northamptonshire Joint Core Strategy which assumes 50,150 completions to 2026 (TEMPRO assumes a further 43 households). Based on the information set out above, it is assumed that modelled growth in this area is constrained to the TEMPRO figures of 50,193 set out above;
iii) further clarity regarding the number of homes proposed over the period to 2026 in the Pre-Submission West Northamptonshire Joint Core Strategy is set out in West Northamptonshire JPU’s ‘Housing Technical Paper Update’ dated July 2012. This paper confirms that the growth of 50,150 households
iv) on this basis, we conclude that the TEMPRO housing number trajectory already incorporates the NSSUE households within the TEMPRO growth for this area;
v) as such, the 1,000 dwellings assumed within TEMPRO for the NSSUE should be excluded from the Do Minimum testing. Any households at the NSSUE Development would form the basis of the Do Something testing.
The requested work items are as follows:
i) in the 2026 Do Minimum test, the following numbers of houses allocated to all the seven Northampton Related Development Area SUE are to be deducted from the TEMPRO total of 50,193 for 2026: Development: Quanta Northampton South SUE 1,000 Northampton Upton Park SUE 979 Northampton Kings Heath Dallington Grange - SUE 3,000 Northampton North - SUE 2,000 Northampton West – SUE 1,500 Northampton South of Brackmills SUE 1,000 Northampton – North of Whitehills SUE 1,000 Remaining total across West Northamptonshire 39,714
ii) no further allowance should be made for the Northampton South SUE in the Do Minimum assessment;
iii) vehicle flows associated with the six other SUE households (a total of 9,479 households - ie, not the NSSUE) within the Northampton Related Development Area are to be loaded onto the network at the approximate location of these developments;
iv) unless Northamptonshire Highways has better information relating to the location of the remaining development, the remaining general growth in household numbers for Northampton is to be applied across the town, the development associated with SNDC, DDC etc is to be focussed at these areas. The total number of households in the West Northamptonshire Area in the Do Minimum test is to be set at 49,193.
This is defined in the Scoping Note Issue 2, attached. The forecasts will be rebuilt as a result of the local model validation exercise and will take account of the latest information for West Northants, including the phasing of developments. Noted, thanks. We would be grateful for confirmation o f the assumed transport network enhancements to be delivered by all parties – HA, NCC and other developers – as requested in paragraph (viii). Future Year Do Something Model The growth rates for the future year Do Something Model will be constrained to TEMPRO with the trips associated with the proposed development added in on top. We comment:
i) as outlined above, we consider that the NSSUE househol ds are already considered within TEMPRO;
ii) as such, the Do Minimum test should reflect 49,193 uni ts (ie “TEMPRO minus
1,000 units for NSSUE”); iii) the trip generation defined by Peter Brett Associates for the additional 1,000
units within NSSUE is to added to the Do Something model – see below for methodology.
The phasing of the proposed development will need to be confirmed to ensure that the future year accurately reflects the desired development form. Confirmation of the phasing is con tained in Table 1 of the attached Scoping Note. The trip rates for the proposed development should be vehicle trip rates with the same level of detail as previously agreed for the East of Kettering development. To respond in full to Nor thamptonshire Highways’ request, we will shortly be issuing a further Technical Note detailing the Person Trip Analysis with source data which leads to the Development vehicle trip generation by plot in Issue 3 of the Scoping Note. All previous details of the Site in the NSTM will be removed and replaced with site specific information. Noted, thanks – we request that this includes the prop osed site links as well as the development quanta as stated in Issue 2 of the Scoping Note. Future Year Mitigated Do-Something The level of retesting as a result of mitigation required will be determined once the results of the initial tests are complete. The costs and programme for this element will be agreed prior to any work being undertaken. Noted, thanks. Data to be provided The revised modelling note needs to set out the details of the area for which data is required and a list of junctions for which specific turning movements is required. A list is included in the attached Issue 2 of the Scoping Note. Whilst we have endeavoured to accommodate Northamptonshire Highways to the best of our ability, this is an iterative pro cess and the list may require some amendment during this work. Section 6: Programme As previously stated the programme for the post validation elements will be agreed once the local model validation exercise is complete and the HA have confirmed that they are content with the results of the revalidated model. Acknowledging that there will be the need for li aison and agreement with third parties, it would be most advantageous to our client to have a “Most Informed” estimate of the programme, reflecting Northamptonshire Highways current commitments, and accepting that this programme may need to be revised at a later date should third party approvals etc affect the programme.
Regards
Lou
Lou Mason-Walsh BA (Hons) MCIHT Senior Transport Planner Northamptonshire Highways
Rob, Following a discussion last week regarding progress on the NSTM modelling with Lou Mason Walsh, I write to gain NCC’s approval to the NSSUE network to be modelled within the NSTM. We propose that the NSTM NSSUE primary link road arrangement be as follows:
Full Development i) the alignment of the Full Development internal primary link road network is to be modelled
as shown on the attached masterplan. For the purposes of the modelling this road is assumed to be 6.5m, and designed to 20mph;
ii) to provide a worst case for any junction capacity assessment at this location, a new four-
arm Rowtree Road Site Access Roundabout is being considered, replacing the existing three arm Rowtree Road / Lichfield Road priority junction. This is shown on the attached PBA drawing 28015/021/001;
iii) a new priority junction access arrangement off Windingbrook Lane. This is shown on the
attached PBA drawing 28015/021/002; iv) the existing vehicular link in the NSTM to the south towards Ash Lane in Collingtree village
currently shown in the model is to be deleted; v) the Phase 1 and Phase 2 residential development zone loaders are to be connected to the
primary link road / secondary link road junctions at the locations shown as internal loop roads on the masterplan;
Phase 1 Development vi) Phase 1 Development is the area identified as “1” to the south-east of the site shown
vii) Phase 1 will be accessed by the Windingbrook Lane priority junction and Windingbrook
Lane only, the Rowtree Rd Site Access Rbt will not be provided as part of Phase 1; viii)the internal site road for Phase 1 will be provided from the Windingbrook Lane access to
the completion of the internal loop – it will not continue passed the public footpath until Phase 2.
These two junction figure plans are work-in-progress, and represent the best available information at this stage. These have been circulated to those involved in the agreement to the modelling, for the purposes of informing this modelling only. Whilst the detail of the junction designs still needs to be agreed, we suggest that these plans are sufficient and acceptable for the purposes of modelling. I trust that this is sufficient, but if you should need any further data please do not hesitate to contact me. Regards, John Hopkins For and on behalf of Peter Brett Associates LLP
From: Sim-Jones, Rob [mailto:[email protected]] Sent: 11 June 2013 09:16
Dear John, Thank you for the information. I am generally content with the information provided on the basis that it is to progress with the modelling work only at this stage. We will clearly comment in detail on the access proposals, impacts on the Public Rights of Way, walking and cycling links, and public transport provision etc., as the scheme progresses. The only comment I have at this stage, is that the 6.5m primary road through the site will have a design speed of 30mph, and not the 20mph proposed, as it will be the bus route and in reality it will be very difficult to control speeds below 30mph. As such, the junction and forward visibility splays will be based on a y-distance of 43m. The primary road network should therefore by modelled with a 30mph design speed. I trust this assists. Kind regards, Rob Rob Sim-Jones Principal Engineer – Development Management Northamptonshire Highways Riverside House, Riverside Way, Bedford Road, Northampton, NN1 5NX DDI: +44(0)1604 364338 Web: www.mgwsp.co.uk;
Subject: NSTM assignment and traffic survey strategy
1. Introduction
1.1 This Technical Note has been prepared by Peter Brett Associates LLP to support the
forthcoming Northampton South SUE (NSSUE - Maple Farm Development, Collingtree) application for outline planning consent.
1.2 At the recent Transport Scoping Meeting with the Highways Agency and Northamptonshire County Council, Peter Brett Associates agreed to prepare this Note to inform the discussions of the Transport Assessment Study area, hence the requirement to obtain existing and commission further traffic count data.
1.3 The County Council has provided Select Zone Analysis SATURN model of the NSSUE zone from the Northamptonshire Transport Strategic Model for both the AM and PM peaks. The assignment of these trips summarised below informs this assessment.
1.4 The SATURN model shows a connection to the south, towards Collingtree. This link
does not form part of the current proposals, flows to this direction have been reassigned, and are assumed to pass along either the A43 (west) or A45 (south). Along both corridors surveys and analysis is being proposed.
2. Summary of NSTM Results 2.1 The NSSUE has been assessed in the NSTM with the Joint Core Strategy proposal for
1,000 dwellings. This shows that the NSTM assumes that there are total NSSUE Development two-way flows of:
i) 282 vehicles in the AM Peak; and
ii) 288 vehicles in the PM Peak.
2.2 For the purposes of establishing the initial Study area, it is assumed that junctions and links with more than 20 two-way trips in either peak generated by the NSSUE should be considered as:
i) the model shows a trip rate of around 0.3 trips per dwelling - it is likely that further work would establish a trip rate around 50% higher (ie, a reported modelled flow of 20 trips would increase to 30 two-way trips);
ii) whilst it is acknowledged that the Highways Agency’s “A Protocol for Dealing
with Planning Applications” applies to trunk roads, it provides a reasonable threshold requirement for link and junction assessments where there are more than 30 No. two-way trips generated by the Development.
2.3 The reassignment of trips as a consequence of the removal of the Collingtree link are
included in this table as the “+ xx” figures.The following assumptions have been made for these trips that have been assigned to the highway network:
i) movements modelled as entering / leaving the Development between the south-east through Collingtree to the A45 would reassign to Rowtree Road (East) and the A45;
ii) A43 northbound movements modelled as passing through Blisworth (along the
old A43) would be reassigned along A43, Danes Camp Way, Towcester Road, and Rowtree Road;
iii) A43 southbound movements modelled as passing through Blisworth (along the
old A43) would be reassigned along Penvale Road, Mere Way, Danes Camp Way to the A43;
iv) similar routes to (ii) and (iii) for trips to Gayton and Rothersthorpe passing
Milton Malsor. 2.4 The flows are summarised by link in Table 2.1, Figure 1, the Link Reference Plan is
attached at the end of this Technical Note which shows the links highlighted in orange. As contributions to the Northampton Growth management Strategy will respond to Development trips on the Highways Agency’s trunk road network, flows along the trunk road have not been considered. Table 2.1 – Summary of NSSUE Development link flows (from NSTM)
Ref. Link AM Peak PM Peak AM % PM % 01 Rowtree Road (West)
31 + 9 = 40
38 + 12 = 50
14.2% 17.4%
02 Rowtree Road (East)
146 + 16 = 162
133 + 13 = 146
57.4% 50.7%
03 London Road entrance onto A45
48 + 2 = 50
29 + 1 30
17.7% 10.4%
04 Bridge Way Meadow / Wooldale Road Link
51 20 18.1% 6.9%
05 Penvale Road
49 + 22 = 71
64 + 12 = 76
25.2% 26.4%
06 Mere Way (East of Penvale Road)
13 31 4.6% 10.8%
07 Mere Way (West of Penvale Road)
33 + 22 = 55
25 + 12 = 37
19.5% 12.8%
08 Danes Camp Way (East of A5076 Upton Way)
15 + 31 = 46
8 + 24 = 32
16.3% 11.1%
09 Towcester Road (North of Ring Road)
30 37 10.6% 12.8%
10 Towcester Road (South of St Peter’s Way)
20 28 7.1% 9.7%
11 St Peter’s Way (West) 10 8 3.5% 2.8% 12 Horseshoe Street 7 17 2.5% 5.9% 13 St Peter’s Way (East) 2 1 0.7% 0.3% 14 A43 South of Danes Camp 6 + 16
= 22 1 + 11 = 12
6.7% 4.2%
15 Banbury Lane 2 + 15 = 17
3 + 13 = 16
6.0% 6.6%
16
Towcester Road (South of Mere Way / Danes Camp Way / Towcester Rd Rbt to Clannell Road)
13 + 9 = 22
21 + 7 = 28
7.8% 9.7%
17 Wooldale Road (east of A45 East Rbt)
21 23 7.5% 8.0%
18 A43 Lumbertubs Way (South of Wellingborough Rd to Standens Barn Road)
2.5 This identifies that the impact of the flow towards the town centre along Towcester
Road is in excess of 20% - hence significant - to the St Peter’s Way / Towcester Road (Gas Street) Roundabout. As the flow then divides at this junction, the impacts on the feeder arms to this junction reduce towards being not significant at this point.
2.6 With respect to the flow in excess of 20 two-way trips along Link 18 Lumbertubs Way, the total Development trips on the A43 reduces and dissipates gradually at each side road junction, so that the Development impact would reduce towards being not significant prior to any at-grade junction to the north along the A43.
2.7 The proposed Study Area is shown on Figure 1.
3. Junction surveys
3.1 Of the above links and junctions, NCC has already confirmed that they hold the
following data within the Study Area, shown on Figure 2 (highlighted in red): Table 3.1 – Existing and appropriate Traffic Count Survey within the Study Area: Survey Ref Location Type (MCC /
ATC) Date
NDC2456-M01 Queen Eleanor Rbt
MCC 06/11/2012
NDC2456-M02 Mereway / Wootton Hall Park
MCC 06/11/2012
NDC2456-M03 Mereway / Penvale Road
MCC 06/11/2012
NDC2456-M04 Towcester Rd / Mereway / Tesco entrance / Tesco Exit / Towcester Road / Danes Camp Way
MCC 06/11/2012
NDC0379-M17 Towcester Rd / St Leonards Rd / Main Rd, Northampton.
MCC 08/07/2009
165 Danes Camp Way ATC June 2012
3.2 We propose that this existing data would need to be augmented with a data collection
exercise at the following locations shown on Figure 2 below (highlighted in green): 1. Wooldale Road / Berry Lane Roundabout; 2. Bridge Meadow Way / Berry Lane / A45 southbound Off Slip / southbound On
Slip Roundabout; 3. Wooldale Road / London Road / Rowtree Road / A45 northbound Off Slip
Roundabout; 4. Rowtree Road / Windingbrook Lane / Butts Road Roundabout; 5. Rowtree Road / Penvale Road Mini Roundabout; 6. Danes Camp Way / Upton Way / Upton Valley Way (East) Traffic Signal
Roundabout; 7. St James Mill Road East / Towcester Road Roundabout;
8. Towcester Road / Horseshoe Street / St Peter’s Way Traffic Signal Roundabout;