FILENAME: C:\USERS\EHEINE\APPDATA\LOCAL\MICROSOFT\WINDOWS\TEMPORARY INTERNET FILES\CONTENT.OUTLOOK\TN1T99MT\TECH MEMO - FORECASTS - DRAFT - 061113.DOCX TECHNICAL MEMORANDUM Lake County Transit Development Plan Update ‒ Preliminary Fixed-Route Ridership Forecasting Date: June 11, 2013 Project #: 11073.0 To: Mike Woods, Lake-Sumter Metropolitan Planning Organization From: Kelly Blume, Kittelson & Associates, Inc. cc: Caroline Ferris, TranSystems OVERVIEW Preliminary fixed-route ridership forecasts for existing LakeXpress routes and potential new routes were developed using the sketch-level process described in this technical memorandum. The sketch- level process relies on elasticities, model-generated origin-destination (O-D) flows, existing and future population densities, and research that relates household densities to the amount of transit service that a given household density can support. Elasticities assess changes to existing routes. O-D flows and transit-supportive densities assess the need for new routes and identify the level of transit service to be offered by the new routes. Transit-supportive densities also link ridership to land use. The resulting forecasts relayed in this memorandum are considered to be preliminary because other project activities (e.g., public outreach) are anticipated to shape the identification of specific needs. BACKGROUND CONCEPTS Elasticity Elasticity is a concept originating in economics that relates a 1% change in a parameter such as transit travel time to a corresponding change in ridership. Elasticities for transit service coverage (expressed as bus miles) and bus frequency (expressed as buses per hour) were obtained from Exhibit 3-19 of TCRP Report 118. These elasticities are +0.8 and +0.4, respectively. TCRP Report 118 also provides the following formula for applying elasticities to produce estimates of ridership change: ( ) ( ) ( ) ( ) Where E = elasticity R 1 = base ridership R 2 = estimated future ridership X 1 = quantity of base attribute (travel time or frequency)
29
Embed
Lake County TDP Preliminary Fixed-Route Ridership Forecasting · Lake County Transit Development Plan Update ‒ Preliminary Fixed-Route Ridership Forecasting Date: June 11, ... (heavy
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
FILENAME: C:\USERS\EHEINE\APPDATA\LOCAL\MICROSOFT\WINDOWS\TEMPORARY INTERNET FILES\CONTENT.OUTLOOK\TN1T99MT\TECH MEMO - FORECASTS - DRAFT -
061113.DOCX
TECHNICAL MEMORANDUM Lake County Transit Development Plan Update ‒ Preliminary Fixed-Route Ridership Forecasting
Date: June 11, 2013 Project #: 11073.0
To: Mike Woods, Lake-Sumter Metropolitan Planning Organization
From: Kelly Blume, Kittelson & Associates, Inc.
cc: Caroline Ferris, TranSystems
OVERVIEW
Preliminary fixed-route ridership forecasts for existing LakeXpress routes and potential new routes
were developed using the sketch-level process described in this technical memorandum. The sketch-
level process relies on elasticities, model-generated origin-destination (O-D) flows, existing and future
population densities, and research that relates household densities to the amount of transit service
that a given household density can support. Elasticities assess changes to existing routes. O-D flows
and transit-supportive densities assess the need for new routes and identify the level of transit
service to be offered by the new routes. Transit-supportive densities also link ridership to land use.
The resulting forecasts relayed in this memorandum are considered to be preliminary because other
project activities (e.g., public outreach) are anticipated to shape the identification of specific needs.
BACKGROUND CONCEPTS
Elasticity
Elasticity is a concept originating in economics that relates a 1% change in a parameter such as transit
travel time to a corresponding change in ridership. Elasticities for transit service coverage (expressed
as bus miles) and bus frequency (expressed as buses per hour) were obtained from Exhibit 3-19 of
TCRP Report 118. These elasticities are +0.8 and +0.4, respectively. TCRP Report 118 also provides
the following formula for applying elasticities to produce estimates of ridership change:
( ) ( ) ( ) ( )
Where E = elasticity
R1 = base ridership
R2 = estimated future ridership
X1 = quantity of base attribute (travel time or frequency)
Lake County Transit Development Plan Update ‒ Preliminary Fixed-Route Ridership Forecasting Project #: 11073.0 June 11, 2013 Page 2
Kittelson & Associates, Inc. Orlando, Florida
X2 = quantity of future attribute (travel time or frequency)
Using this formula, it is possible to estimate the change in ridership that corresponds to a change in
bus miles (possibly resulting from extension of an existing route and/or an increase in service span) or
to a change in bus frequency.
Transit-Supportive Densities
Several sources of information exist that relate density (typically household or employment density)
to the level of transit service that it can support. The formative research on this subject is that of
Pushkarev and Zupan (1987). Table 1 summarizes their findings.
Table 1 Transit-Supportive Densities
Mode Service
Minimum Necessary Residential Density (Dwelling
Units per Net Acre)1 Remarks
Dial-a-bus2
Many origins to many destinations
6 Only if labor costs are not more than twice those of taxis.
Dial-a-bus2
Fixed destination or subscription service
3.5-5 Lower figure if labor costs are twice those of taxis; higher if thrice those of taxis.
Local bus
“Minimum” 1/2-mile route spacing 20 buses per day ~60-minute headways
4.5
Average. Varies as a function of downtown size and distance from residential areas to downtown.
“Intermediate” 1/2-mile route spacing 40 buses per day ~30-minute headways
7
“Frequent” 1/2-mile route spacing 120 buses per day ~10-minute headways
15
Express bus, accessed on foot
5 buses during 2-hour peak period
15 (average density over 2 square mile tributary area)
From 10-15 miles away to largest downtowns only.
Express bus, accessed by auto
5-10 buses during 2-hour peak period
3 (average density over 20 square mile tributary area)
From 10-20 miles away to downtowns larger than 20 million square feet of non-residential floor space.
Light rail3
5-minute headways or better during peak hour
9 (average density for a corridor of 25-100 square miles)
To downtowns of 20-50 million square feet of nonresidential floor space.
Rapid transit (heavy rail)
5-minute headways or better during peak hour
12 (average density for a corridor of 100-150 square miles)
To downtowns larger than 50 million square feet of nonresidential floor space.
Commuter rail 20 trains per day 1-2 Only to largest downtowns, if rail line exists.
1 Multiply by 0.67 for gross acres. SOURCE: Pushkarev and Zupan (1987)
2 Demand-responsive service available to the general public
3 Full-featured bus rapid transit (BRT) also
Lake County Transit Development Plan Update ‒ Preliminary Fixed-Route Ridership Forecasting Project #: 11073.0 June 11, 2013 Page 3
Kittelson & Associates, Inc. Orlando, Florida
APPROACH TO RIDERSHIP FORECASTING
Existing LakeXpress Routes
The steps used to develop preliminary 2023 ridership forecasts for the existing routes are as follows:
1. Obtain 2012 ridership data for the existing LakeXpress routes.
2. From the 2012 ridership data, determine average daily riders for each route.
3. Estimate ridership changes for each existing route using elasticities under three preliminary
scenarios:
a. 10% increase in bus miles
b. 20% increase in bus miles
c. 50% reduction in headway
4. Apply a growth factor to represent 2023 conditions.
Potential New Routes
The steps used to develop preliminary 2023 ridership forecasts for potential new routes are as
follows:
1. Obtain 2005 and 2035 O-D flows between population centers in Lake and Sumter Counties
and between Lake and Sumter Counties and adjacent counties from the Central Florida
Regional Planning Model (CFRPM). Interpolate to obtain 2012 and 2023 O-D flows.
2. Determine existing transit mode split in corridors that are currently served by transit.
a. Identify O-D flows in corridors that are currently served by transit.
b. Using 2012 ridership data, calculate the transit mode split for each of the corridors
currently served by transit.
3. Use data about transit-supportive future densities to identify the appropriate level of transit
service (if any) to be provided in corridors that currently served by fixed-route transit or are
not currently served by transit.
a. Consider densities within 1/4 mile of the potential transit route.
b. Consider that the pattern of transit-supportive densities may suggest express or
limited-stop bus service instead of local bus service.
4. Estimate potential 2023 daily ridership in the corridors identified in Step 3.
a. If an existing route could support increases in service based on level of transit-
supportive densities in the corridor, apply elasticities and a growth factor .
b. Identify O-D flows in corridors that are not currently served by transit.
c. Apply a transit mode split, calculated in Step 2, from a comparable corridor that is
currently served by transit.
d. Select the transit mode split that best represents the potential service in the new
corridors. For example, a new route that connects multiple cities might use the transit
mode split determined for Route 1 (i.e., 6%) while a new circulator might use 2 or 3%
(representative of Routes 2 and 3).
Lake County Transit Development Plan Update ‒ Preliminary Fixed-Route Ridership Forecasting Project #: 11073.0 June 11, 2013 Page 4
Kittelson & Associates, Inc. Orlando, Florida
e. Elasticities could be used to refine the results of this step. If the new route that
connects multiple cities operates at half-hour headways, for example, application of
the 6% mode split to the corridor may underestimate transit trip potential.
RESULTS
Elasticities and Existing Routes
Table 2 summarizes the results of applying elasticity factors to existing ridership under three
preliminary scenarios. Existing transit ridership data can be found in Attachment 1. The detailed
calculations for the application of the elasticity factors can be found in Attachment 2. The
calculations can be readily updated to reflect alternative improvement scenarios for the existing
transit routes, as alternatives are developed and refined to reflect input from other tasks.
Table 2 Changes in Average Daily Ridership Based on Elasticities
Route Existing August 2012
(Peak Month) 10% Increase in Revenue Miles
20% Increase in Revenue Miles
50% Reduction in Headway
1 696 751 805 910
2 249 269 288 326
3 171 184 198 223
4 310 334 358 405
Total 1,426 1,538 (+112) 1,649 (+223) 1,864 (+438)
O-D Flows
Attachment 3 contains detailed O-D flow tables for 2005 and 2035. Table 3 below reports the total
2023 O-D flows. The 2023 O-D flows were interpolated from the 2005 and 2035 tables. Table 3
includes home-based work, home-based shopping, home-based social/recreational, home-based
other, and non-home-based trips.
Table 4 shows approximate transit mode splits in corridors currently served by transit, which were
calculated by comparing 2011/2012 ridership data to O-D flows (interpolated from the 2005 and 2035
O-D data) in the same service areas. This is an approximation, but it provides a general idea of how
often transit is currently used with respect to total travel demand in a given service area. When these
mode splits are applied in comparable service areas, they provide estimates of ridership potential in
the new service areas.
The transit mode splits also can be used to estimate ridership associated with a given O-D forecast
year. Table 5 shows growth factors for converting 2012 O-D flows to 2023 flows. Application of
growth factors in this manner assumes that transit ridership grows in proportion to total travel
demand. This assumption does not take into consideration factors such as increased congestion
(which may or may not provide an incentive for drivers to switch to transit), intensified transportation
Lake County Transit Development Plan Update ‒ Preliminary Fixed-Route Ridership Forecasting Project #: 11073.0 June 11, 2013 Page 5
Kittelson & Associates, Inc. Orlando, Florida
demand management (TDM) programs, fare increases, etc. It assumes that current levels of transit
*Based on interpolation between 2005 and 2035 O-D flows.
Lake County Transit Development Plan Update ‒ Preliminary Fixed-Route Ridership Forecasting Project #: 11073.0 June 11, 2013 Page 10
Kittelson & Associates, Inc. Orlando, Florida
Table 6 Preliminary 2023 Service Plan for New and Increased Transit Service in Lake County
Route Description of 2023
Service Assumptions and Estimation Process
2023 Daily Ridership Forecast
1 (existing)
Maintain 2012 headway and service span.
Subject to efficiency and effectiveness analysis conducted under a separate task. If route is continued in its current form, apply Table 5 growth factor (averaged
for the route's service area) to existing ridership.
7,675 x growth factor of 1.25 = 9,600 (rounded)
2 (existing)
Maintain 2012 headway and service span.
Subject to efficiency and effectiveness analysis conducted under a separate task. If route is continued in its current form, apply Table 5 growth factor (averaged
for the route's service area) to existing ridership.
2,613 x growth factor of 1.33 = 3,500 (rounded)
3 (existing)
Maintain 2012 headway and service span.
Subject to efficiency and effectiveness analysis conducted under a separate task. If route is continued in its current form, apply Table 5 growth factor (averaged
for the route's service area) to existing ridership.
2,089 x growth factor of 1.18 = 2,500 (rounded)
4 (existing)
Maintain 2012 headway and service span.
Subject to efficiency and effectiveness analysis conducted under a separate task. If route is continued in its current form, apply Table 5 growth factor (averaged
for the route's service area) to existing ridership.
1,703 x growth factor of 1.26 = 2,100 (rounded)
1 (enhanced)
Increase amount of service currently provided (e.g.,
reduce headways and/or increase service span).
Long-term potential indicated by transit-supportive densities analysis. Apply elasticities to estimate ridership
increase associated with a specific change in headway. Assume 50% reduction in headways initially. Increase
ridership per Table 2. Then apply Table 5 growth factor (averaged for the route's service area) to the increase.
7,675 + 910 = 8,585 (Year 2012)
8,585 x growth factor of 1.25 =
10,700 (Year 2023 rounded)
2 (enhanced)
Increase amount of service currently provided (e.g.,
reduce headways and/or increase service span).
Long-term potential indicated by transit-supportive densities analysis. Apply elasticities to estimate ridership
increase associated with a specific change in headway. Assume 50% reduction in headways initially. Increase
ridership per Table 2. Then apply Table 5 growth factor (averaged for the route's service area) to the increase.
2,613 + 326 = 2,939 (Year 2012)
2,939 x growth factor of 1.33 =
3,900 (Year 2023 rounded)
3 (enhanced)
Increase amount of service currently provided (e.g.,
reduce headways and/or increase service span).
Long-term potential indicated by transit-supportive densities analysis. Apply elasticities to estimate ridership
increase associated with a specific change in headway. Assume 50% reduction in headways initially. Increase
ridership per Table 2. Then apply Table 5 growth factor (averaged for the route's service area) to the increase.
2,089 + 223 = 2,312 (Year 2012)
2,312 x growth factor of 1.18 = 2,700 (Year 23
rounded)
4 (enhanced)
Increase amount of service currently provided (e.g.,
reduce headways and/or increase service span).
Desirable to provide headways of 60 minutes or better as a QOS policy. Apply elasticities to estimate ridership increase associated with a specific change in headway. Assume 50% reduction in headways initially. Increase
ridership per Table 2. Then apply Table 5 growth factor (averaged for the route's service area) to the increase.
1,703 + 405 = 2,108 (Year 2012)
2,108 x growth factor of 1.26 =
2,700 (Year 2023 rounded)
5 (new)
Provide limited-stop service at 60-minute headways between central Lake County,
Clermont, and
Potential indicated by transit-supportive densities analysis. Apply 5% mode split (modeled after existing Route 4) to 2023 O-D flow in corridor (i.e., Groveland-
Clermont-Tavares). As existing Route 4 mode split is not reflective of level of transit service to be established in
26,584 x 5% = 1,329 (Year 2023
base)
1,329 + 409 =
Lake County Transit Development Plan Update ‒ Preliminary Fixed-Route Ridership Forecasting Project #: 11073.0 June 11, 2013 Page 11
Kittelson & Associates, Inc. Orlando, Florida
Groveland. corridor (i.e., 60-minute headways), apply elasticities to account for the headway reduction.
1,700 (enhanced, rounded)
6 (new)
In partnership with Sumter County local
governments, provide limited-stop service at 60-minute headways between central Lake County,
The Villages, and Wildwood
Potential indicated by transit-supportive densities analysis. Apply 5% mode split (modeled after existing
Route 4) to 2023 O-D flow in corridor (i.e., Leesburg-The Villages-Wildwood). As existing Route 4 mode split is not
reflective of level of transit service to be established in corridor (i.e., 60-minute headways), apply elasticities to
account for the headway reduction.
62,478 x 5% = 3,124 (Year 2023
base)
3,124 + 961 = 4,085 (enhanced,
rounded)
7 (new)
In partnership with LYNX, extend
commuter bus service west of
Clermont to Groveland
Extension of LYNX 204 Clermont Xpress. To be determined
The information in Table 6 suggests the following:
Headway improvements may offer more return on investment for Route 4 than the other existing routes.
A service that crosses the Lake/Sumter County line has the potential to transport a daily ridership volume more than that currently transported by Route 2. Services that cross county lines create opportunities for partnerships to offset the costs of improved regional transit connectivity.
A staged approach to implementing transit service improvements is feasible. Routes can be established primarily to create connections, and the QOS for each route can be improved as justified over time.
NEXT STEPS
Ridership forecasts can be used to develop revenue forecasts and assess the need for additional
person capacity (in the form of more buses and/or larger buses) in a given corridor. Revenue
forecasts and person capacity improvements feed into cost estimation, which is the topic of a
subsequent task.
ATTACHMENT 1: RIDERSHIP DATA
LakeXpress Route 1 Oct 2011 Nov 2011 Dec 2011 Jan 2012 Feb 2012 Mar 2012 Apr 2012 May 2012 June 2012 July 2012 Aug 2012 Sep 2012