Post on 02-Oct-2020
Leicester and Leicestershire
Integrated Transport Model (LLITM)
PR06: Forecasting Report
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 3 of 16
PR06: Forecasting
Report
Prepared by: Ian Stanness Checked by: Mark Dazeley
Senior Consultant Associate Director
Approved by: Mark Dazeley
Associate Director
Rev No. Comments Checked by Approved by Date
1 Final Issue MJD MJD 30/09/2011
AECOM House, 63-77 Victoria Street, St Albans, Hertfordshire, AL1 3ER
Telephone: 01727 535000 Website: http://www.aecom.com
Job No: 60096658 Reference: M001.900 Date Created: 6th May 2011
This document has been prepared by AECOM Limited for the sole use of our client (the “Client”) and in accordance
with generally accepted consultancy principles, the budget for fees and the terms of reference agreed between
AECOM Limited and the Client. Any information provided by third parties and referred to herein has not been
checked or verified by AECOM Limited, unless otherwise expressly stated in the document. No third party may rely
upon this document without the prior and express written agreement of AECOM Limited.
Contains Ordnance Survey data © Crown copyright and database right 2011
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 4 of 16
Table of Contents Section 1 – Overview .................................................................................................................................. 5
1.1 Introduction ........................................................................................................................................ 5 1.2 Report Structure ................................................................................................................................ 5
Section 2 – Forecasting Processes ............................................................................................................ 6 2.1 Introduction ........................................................................................................................................ 6 2.2 Land Use Model ................................................................................................................................ 7 2.3 Supply Models ................................................................................................................................... 7 2.4 Demand Model .................................................................................................................................. 7
Section 3 – Forecasting Assumptions and Results .................................................................................... 9 3.1 „Core‟ Scenario Assumptions ............................................................................................................ 9 3.2 Demonstration Testing Results ....................................................................................................... 16
List of Tables Table 3.1: Forecast Assumptions ............................................................................................................... 9 Table 3.2: Highway Network „Core‟ Scheme List ..................................................................................... 12 Table 3.3: Public Transport Network „Core‟ Scheme List ......................................................................... 15
List of Figures Figure 2.1: Overview of Data Flow within LLITM ........................................................................................ 6 Figure 2.2: Example Choice Model Structure ............................................................................................. 8
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 5 of 16
Section 1 – Overview
1.1 Introduction
This report discusses the forecasting assumptions and processes used in the Leicester and
Leicestershire Integrated Transport Model (LLITM). The LLITM suite consists of four components,
namely:
a highway supply model (LLITM-HW), developed in SATURN by Scott Wilson;
a public transport supply model (LLITM-PT), developed in CUBE Voyager by Scott Wilson;
a variable demand model (LLITM-DM), the subject of this document, built in EMME by AECOM;
and
a land-use model (LLITM-LUM), developed in bespoke DELTA software by David Simmonds
Consultancy.
In addition to these, there is an environmental assessment package (EASE), developed by AECOM in
various software packages, primarily using MS Excel, MS Access and MapInfo. This is predominantly a
reporting tool, and is therefore not considered as part of this report.
1.2 Report Structure
This report is one of a series that document the LLITM model suite, as of September 2011. The
documentation takes the form of formal project reports, supported by various technical notes. The
project reports are as listed below:
PR01: Data Collection Report
PR02: Highway Local Model Validation Report;
PR03: Public Transport Local Model Validation Report;
PR04: Land Use Model Development Report;
PR05: Demand Model Development Report
PR06: Forecasting Report (this document);
PR07: Demonstration Testing Report; and
PR08: LLITM User Guide.
A significant proportion of the information relating to the forecasting processes can be found in the other
project reports, primarily PR04 and PR05. In addition to this there is also information contained in a
number of related technical notes.
Following this introduction, this report contains the following chapters:
Section 2 discusses the forecasting process adopted in the main elements of the integrated
LLITM model.
Section 3 contains the assumptions used within the initial demonstration testing of the model in
terms of economic assumptions, and highway and public transport scheme assumptions.
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 6 of 16
Section 2 – Forecasting Processes
2.1 Introduction
This chapter outlines the forecast processes contained within LLITM, drawing on information contained
within ‘PR04: Land Use Model Development Report’ and ‘PR05: Demand Model Development Report’.
These two project reports contain detailed commentaries on the assumptions and process contained
within the land use and demand models respectively, with this report giving an outline of these
processes.
Figure 2.1 shows an outline of the flow of information and data within LLITM when forecasting. The
land use model produces planning data forecasts based on the results of the previous transport model
year. These planning assumptions, along with the highway and public transport network assumptions,
and various economic assumptions are used within the variable demand model to produce forecast
future year demand. This iteration between the land use and transport models continues at five-yearly
intervals to build up a set of forecasts.
Figure 2.1: Overview of Data Flow within LLITM
The following sections detail some of the process containing in the main elements of LLITM, and give
references to other reports and technical notes where applicable.
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 7 of 16
2.2 Land Use Model
As previously stated, ‘PR04: Land Use Model Development Report’ should be referred to for information
on the processes and methodologies applied in the land use model.
LLITM-LUM is essentially a model of land use and economic changes over time, working in one year
steps. Every fifth year, land use data is passed to the transport model, which is run and matrices of
generalised costs are passed back to LLITM-LUM. The main model processes contained within LLITM-
LUM consider the following:
accessibility;
employment;
migration;
household transition;
housing location;
car ownership;
development; and
residential quality.
Further information on each of these key processes can be found in PR04. The main outputs from
these process that are used in the model as a whole are the population, household, employment and
car ownership forecasts. It is these forecasts that drive the forecast trip ends, through the use of
CTripEnd, and ultimately drive the starting point for demand in a given future year.
More detail on the interfaces between the land use and demand models can be found in ‘TN25 -
Demand Model - Land Use Model Interfacing’. ‘TN33 - The LLITM Trip End Model’ also contains
information of the adjustments made to the standard CTripEnd model in order to accommodate the
more disaggregate data provided for the Leicestershire area from the land use model.
2.3 Supply Models
LLITM contains both highway and public transport assignment models. The validation report for each of
these elements can be found in ‘PR02: Highway Local Model Validation Report’ and ‘PR03: Public
Transport Local Model Validation Report’ respectively.
In the process of undertaking the LLITM demonstration tests, which were designed to test various
elements of the model and test their responses, a series of future year highway and public transport
networks were required. The schemes that were included in these tests are given in Table 3.2 and
Table 3.3, in Section 3.1, for the highway and public transport models respectively.
It should be noted that these schemes are those used defined for the LLITM demonstration tests, and
may not be the schemes used in another application of LLITM. Each application may choose to define
a different set of „core‟ network scheme assumptions.
2.4 Demand Model
LLITM contains a WebTAG compliant variable demand model, which is detailed in ‘PR05: Demand
Model Development Report’. The demand model also contains a parking model for trips attracted to
Leicester City and Loughborough town centre, and this component of the model is discussed in ‘TN17 -
Parking Model Specification’.
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 8 of 16
One of the main elements of the demand model is the choice structure defining how forecast demand is
derived. Figure 2.2 shows an example of this choice structure for a car-available, non-freight trip
purpose. The choices available to other segments of demand may differ from this. For example, no-car
available demand does not have the choice of „car‟ as a mode, and so chooses between public
transport and active mode.
Figure 2.2: Example Choice Model Structure
These choices are based on the composite costs at each choice level, which are derived from the costs
from the assignment models and the parking model, along with the economic parameters assumed in a
given forecast year. The economic assumptions used within the demonstration testing can be found in
Table 3.1.
These choices, for productions within Leicestershire, are based on sensitivity parameters derived from a
local household survey. The estimation of these parameters from the household survey, and the
resultant sensitivity parameters, can be found in ‘TN32 - Household Survey Parameter Estimation’. For
trips produced outside Leicestershire, standard WebTAG sensitivity parameters have been applied.
Motorised Mode Choice(car vs. public transport)
Trip Frequency
Parking Choice
PublicTransport Car
Time Period Choice
Trip Distribution
PNR
Trip Distribution
Public Transport Mode Choice(rail vs. bus)
On-street
Off-street
Rail
BusCUBE
VOYAGER
EMME
Time Period Choice
Active Mode Choice(motorised vs. active)
Time Period Choice
Walk+Cycle
Trip Distribution
Park-and-Ride
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 9 of 16
Section 3 – Forecasting Assumptions and Results
This section details the forecasting assumptions used in the demonstration testing undertaken on
LLITM, which is reported on in ‘PR07 - Demonstration Testing Report’. The assumptions detailed here
are therefore not fixed, and can change in a given application. It is for each application to decide
whether to adopt these assumptions or define alternative inputs.
3.1 ‘Core’ Scenario Assumptions
There are a number of assumptions which are required when running the integrated model in
forecasting mode. These include network inputs for highway and public transport, assumptions on the
supply and cost of parking in Leicester and Loughborough, economic assumptions on such items as
values of time and fuel costs, and planning policy assumptions for the land use model.
In accordance with WebTAG 3.15.5, information regarding potential future land use and transport
developments has been considered together with their likelihoods. The conclusions of this review are
set out in Table 3.1 which lists the assumptions used within the „core‟ scenario demonstration testing,
excluding the network assumptions for the highway and public transport models.
Table 3.1: Forecast Assumptions
Input Assumptions / Source
Economic growth (GDP growth,
value of time)
Information on changes in GDP and thus values of time are taken
from DfT advice (WebTAG 3.5.6D, March 2010).
Values of time are assumed to be constant across modes, time
periods, productions and attractions, and vary only by purpose,
income segment and length of trip. Highway values of time have
been used for business PT trips.
Public transport fares
Bus - 2008 to 2010 based on observed data; 2010 to 2015 1.5% per
annum; and 0.75% per annum thereafter.
Rail - 2008 to 2010 based on observed data (with regulated and
unregulated components based on published information evident at
end of last year). Growth from 2010 to 2011 is based on the
observed growth between 2008 and 2010, and has been assumed to
be 3.1%. Growth from 2011 to 2015 has been assumed to be 3% per
annum; and 1% thereafter.
Vehicle operating costs Changes in fuel prices, vehicle fuel efficiency, and non-fuel operating
costs have been taken from WebTAG 3.5.6C, March 2010.
Parking charges
Parking charges assumed to grow 2% per annum over inflation, in
approximate line with historic salary increases.
For new park-and-ride sites the changes have been taken from the
existing Meynell‟s Gorse park-and-ride site.
Parking capacities The zone capacities of private / non-residential parking (PNR)
increase in relation to the changes in employment within each zone.
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 10 of 16
Input Assumptions / Source
In terms of the new park-and-ride sites, where no specific information
is available, the same capacity as Meynell‟s Gorse in 2008, of 500
spaces, has been assumed. This applies to all new park-and-ride
sites except Birstall which has a known capacity of 1,000 spaces.
Aside from new park-and-ride sites and PNR, the only parking
capacity change from the base year is an increase in the capacity at
Meynell‟s Gorse of 500 spaces.
Land use: population and
employment forecasts
Population and employment growth across the East Midlands sub-
region have been constrained to TEMPRO forecasts. Detailed
information on planning policy (land allocated by development type)
has been collated from individual districts and used in LLITM-LUM.
Car ownership Car ownership is forecast within LLITM-LUM.
Car occupancy Changes in car occupancy over time have been taken from WebTAG
3.5.6C, March 2010
Trip rates
Assumed to be constant over time. Demand growth is applied at a
24-hour level, so „reference demand‟ time period proportions by
purpose are also assumed to be constant over time; modelled
proportions may of course vary due to time-period choice model.
Highway congestion changes (for
external buffer network).
Derived from average changes in congestion in the internal
simulation network, for two forecast years: 2021 and 2031; other
forecast years are interpolated based on these.
NTM was initially considered as a source, but these congestion
changes were found incompatibly low compared with other model
assumptions, unless considerable, and unlikely, infrastructure
improvement in external areas was assumed.
Analysis based on historic trends was found to result in congestion
increases that were incompatibly high with other model assumptions.
Active mode costs
No changes to active mode costs relating to specific infrastructure
(cycle lanes for example) have been included. There are calibrated
mode shifts included in the „core‟ scenario relating to „Smarter
Choices‟.
Smarter Choices
Based on investment levels into Smarter Choices measures of
£200,000 pa for both Leicester City and the remainder of
Leicestershire, target mode shifts have been derived from existing
research and demonstration towns (discussed in Appendix A of
‘PR07 - Demonstration Testing Report’). These targets come into
effect in 2016, and the calibration parameters are constant thereafter,
assuming that investment continues at the same rate.
The mode shifts calibrated in 2016 are:
Workplace travel plans:
o 5% reduction in commuting car drivers to Leicester City
o 6% reduction in commuting car drivers to Leicestershire market
towns
School travel plans:
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 11 of 16
Input Assumptions / Source
o 3% reduction in education car drivers to Leicester City
o 6% reduction in education car drivers to the rest of Leicestershire
Targeted marketing:
o 0.4% reduction in car drivers from Leicester City
o 0.1% reduction in car drivers from the rest of Leicestershire
There are also calibrated changes in car occupancy as a result of
these Smarter Choices measures for workplace and school travel
plans.
Detailed discussion on the derivation of these targets, the calibration
procedure, and the results of this calibration can be found in ‘PR07 -
Demonstration Testing Report’.
Freight growth
Freight growth is not forecast by the land-use model, so growth from
the 2009 version of NTM. This provides growth forecasts for vehicle-
kms for freight, with these growth rates being applied separately to
LGV and OGV base year matrices.
There are also a number of „committed‟ or „highly-likely‟ infrastructure schemes relating to the highway
and public transport networks that are included within these models in the „core‟ scenario. These
schemes are listed in Table 3.2 and Table 3.3 for highway and public transport respectively.
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 12 of 16
Table 3.2: Highway Network ‘Core’ Scheme List
Scheme Description Implementation
Woodville Woodland link road 2008
Blaby, Lutterworth Road - Speed reducing measures 2008
Barrow Upon Soar, Sileby Road - Local Safety Scheme 2008 / 2009
Meton Mowbray, Thorpe End and Norman Way - Junction improvements 2008 / 2009
Rothley, Fowke Street - Speed reducing measures 2008 / 2009
Rothley, Hallfields Lane - Traffic Calming Measures 2008 / 2009
Thurmaston, Barkby Thorpe Lane - Traffic Calming Measures 2008 / 2009
Birstall, Birstall Road - Traffic Calming 2008 / 2009
Syston, Barkby Road - Traffic Calming 2008 / 2009
B582, Oadby, Wigston Road / JW / Brabazon Road - Cycleway and traffic signal alterations 2009
Blaby, Grove Road - speed reducing measures 2009
South Wigston, South Leics College, LTP Contr to S278 2009
B582, South Wigston, Jn Blaby / Saffron Road 2010
A47, Kirby Lane, Signal junction imp - Cycleway and traffic signal alterations 2010
A5199, Wigston, Welford Road - revised speed limit, 2 VAS and signing and lining 2010
Wanlip, Rectory Road - Traffic Calming 2010 / 2011
Glen Parva , Little Glen Road - 3 chicanes, 1 VAS and moving of speed limit 2011
Sanvey Gate junction improvements (Waterside) 2011
Loughborough Eastern Gateway 2012
Loughborough Inner Relief Road 2013
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 13 of 16
Scheme Description Implementation
Melton Bypass 2014 / 2015
A60 / Barrow Road off site Safety Improvement Scheme - Barrow upon Sour Before 2015
Barrow Upon Soar, Cotes Road, Nottingham Road - Traffic Calming Before 2016
Lutterworth Road / Soar Valley Way junction improvements Before 2021
Castle Donington relief road By 2021
Strategic Link - M1 By 2021
Improvements to the Grange Road and Ellistown crossroads - linked to Coalville SUE By 2021
Link across M69 - linked to Enderby / Blaby SUE By 2021
Improve Hinckley Northern Perimeter Road - linked to Hinckley SUE By 2021
Junction 19 improvements By 2021
A5, A47 - Long Shoot New Junction By 2021
A563 Krefeld Way (Ashton Green - Infrastructure to support 3500 houses) Before 2026
Hathern Bypass Linked to Loughborough SUE By 2026
Improvements to B676 to A46, including Wymeswold Burton on the Wolds Bypass - linked to Loughborough SUE By 2026
A512 widening B591 to M1 J23, improvements to J23 and completion of dualling thereafter to either Sneels Nook Lane or
Epinal Way Junction - linked to Loughborough SUE By 2026
Garendon Park Link - linked to Loughborough SUE By 2026
Improvements to A511 Stephenson Way - Coalville - Linked to Coalville SUE By 2026
Improvements to A511 between Coalville and M1 J22, including Birch Tree, Bardon Chapel roundabouts - linked to
Coalville SUE By 2026
Loughborough Eastern Distributor Road - linked to Loughborough SUE By 2026
Bardon Road Link - linked to Coalville SUE By 2026
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 14 of 16
Scheme Description Implementation
Syston Eastern Link Road - linked to Thurmaston SUE By 2026
Improve link to A607 including ASDA roundabout linked to Thurmaston SUE By 2026
Countywide interchanges By 2026
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 15 of 16
Table 3.3: Public Transport Network ‘Core’ Scheme List
Scheme Description Locale Type Implementation
A47 Humberstone Road Quality Bus Corridor (Phase 1) - St Mathew's Way to Forest Road City Bus Completed 2008
Enderby park-and-ride City / County Bus Completed 2008
St. Nicholas Place park-and-ride City Bus Completed 2009
Birstall park-and-ride County Bus 2010 / 2011
Glen Parva A426 Bus Corridor County Bus 2011
Oadby A6 bus lane scheme County Bus 2009
A6 Oadby, Leicester Road - Added road width to accommodate bus lanes and touch crossing County Bus 2009
A6, Oadby, Leicester Road - Bus priority for lights and toucan provision County Bus 2011
A607 Melton Road Quality Bus Corridor including Melton Road / Troon Way junction
improvements City Bus Before 2016
A426 Aylestone Road / Lutterworth Road Quality Bus Corridor City Bus Before 2016
Birstall park-and-ride City Bus Before 2016
Leicester SCOOT with bus prioritisation City Bus Before 2016
Public transport link to Lubbesthorpe SUE County Bus By 2016
A50 Quality Bus Corridor - Sanvey Gate to County Boundary City Bus Before 2021
A47 Humberstone Road Quality Bus Corridor (Phase 2) - Forest Road to County boundary City Bus Before 2021
Glenfield park-and-ride City / County Bus Before 2021
Meynell's Gorse park-and-ride +500 capacity to 1050 City Bus Before 2026
A6 Oadby Racecourse park-and-ride City Bus Before 2026
A47 Nurseries site park-and-ride City Bus Before 2026
Leicester and Leicestershire
Integrated Transport Model (LLITM)
Page: 16 of 16
3.2 Demonstration Testing Results
‘PR07 - Demonstration Testing Report’ details the results of the forecasting undertaking using the
processes detailed in Section 2, and the assumptions contained in Section 3.1. This report contains,
amongst other information, the following:
The results of the land use forecasts from the land use model over time in the „core‟ scenario,
with comparisons against TEMPRO forecasts in terms of both planning data and car ownership
results.
The changes in travel demand by mode and purpose that this generates, and the effect of the
demand model on the „reference‟ demand derived from the planning data forecasts.
The impact that the changes in highway demand has on the performance of the highway
network in terms of some key network statistics such as vehicle-kms and average speeds.
Included in the demonstration testing was the response of the model to schemes, or „interventions‟, in
the highway, public transport and parking model assumptions. These schemes are a highway
infrastructure change, a new bus service and a change to the parking charge assumptions in a town.
These schemes were located in three discrete areas of the model so that they could be testing together
with minimal interaction between the three interventions. The details of these schemes, and the results
of these tests, can be found in ‘PR07 - Demonstration Testing Report’.