Fixed-Route Transit Ridership Forecasting and Service Planning Methods (2006)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

39 FIXED-ROUTE TRANSIT RIDERSHIP FORECASTING AND SERVICE PLANNING METHODS The sum of the number of responses does not equal the total number of respondents, because many questions allowed multi- ple answers. The number of responses is listed first, followed by the percentage of total respondents for that question. Responses to open-ended questions have been summarized into categories. A number next to an answer in the “Other” cate- gory indicates that more than one transit agency listed this response; if there is no number, the response was mentioned once. GENERAL RIDERSHIP FORECASTING 1. Do you forecast ridership for: Minor adjustments to a route segment 12 33% Scheduling changes 11 31% Route changes affecting less than 25% of a route 16 44% Route changes affecting 25% or more of a route 24 67% New routes 31 86% New mode/new type of service 24 67% The next fiscal year 22 61% The next five or ten years, or other long-term forecast 23 64% Other (please describe): 5 14% Other includes fare changes (3), contingency service reductions, new or extended fixed guideway, New Starts projects, park-and-ride lots, economic/demographic shifts, rolling stock/facility/capacity needs. 2. Is there a threshold in terms of the scale of service change that triggers a ridership forecast? If so, what is the threshold? Formal 13 41% Informal 8 25% None 11 34% Total responding 32 100% Thresholds include: Greater than 10% change in platform miles/hours Any change that significantly (by 10% or more) affects the resources required to deliver service on a route 25% change in miles/hours, less for Title VI analysis 25% change in ridership (required by city ordinance) 25% of route unless number of riders affected is moderate Action required by Board Added cost Any fare change Anything other than minor schedule adjustments Change in resource allocation Changes in route alignment and number of trips Significant enough to be part of service change program When transit agency requests 3. Do you have more than one method of forecasting ridership, depending on the scope of the change? Yes 23 66% No 12 34% APPENDIX B Summary of Survey Results

40 4. How are these forecasts distributed and used? Internally 33 97% To Board members 23 68% To the MPO 10 29% To elected official 9 26% Other (please specify): 10 29% Others include FTA (3), interest groups/stakeholders/general public (4), other agency/external departments (3), consultants. 5. Is there a dedicated person or group responsible for ridership forecasts, or do planners or other personnel estimate future ridership among their other duties? Dedicated person/group 8 24% Not dedicated—part of general duties 13 38% Depends on scale/extent of ridership forecast 13 38% 6. Which department or agency has the lead for preparing ridership forecasts? Transit operations planning department 9 25% Transit operations department 3 9% Transit planning department 22 65% Transit budget/finance department 8 24% MPO 6 18% Other (please specify): 4 12% Other includes strategic planning and policy, transit research section (marketing department), business development, consultant (for BRT), service evaluation section, systems analysis section. 7. Do you consider the following as inputs in your methodology? If a factor is involved for some types of changes but not others, please indicate. Existing system ridership 28 80% Existing route or route segment ridership 31 89% Ridership on similar routes 30 86% Origin/destination information 24 69% Demographic factors within the service area 27 77% Land use within the affected service area 25 71% Economic trends within the service area 21 60% Other (please specify): 10 29% Other includes auto ownership, travel time (2), congestion level (2), distance from major activity centers, fare/pricing infor- mation (2), modal competition, service frequency, transfer activity, access mode (2), egress mode, market research surveys, new employment/retail development, trip generators in affected area. 7A. Type of change or forecast for existing system ridership. Annual budget/forecast 5 Long-range plan 3 Four-step travel model 2 Fare changes 2 For system ridership 2 5-year plan 2 Major new service, including fixed guideway extensions 1 Span of service 1 Change in scheduled service level 1 All 1 Any route or service change 1

41 Elasticities 1 Model validation 1 7B. Type of change or forecast for existing route/route segment ridership. Change in route 3 Span of service 2 Change in scheduled service level 2 All 2 Any route or service change 2 Elasticities 2 Major new service, including fixed guideway extensions 1 Four-step travel model 1 Timed transfer 1 New route or corridor 1 Model validation 1 Significant service changes 1 7C. Type of change or forecast for ridership on similar routes. New route or corridor 7 Change in route 4 Span of service 2 Major new service, including fixed guideway extensions 1 Annual budget/forecast 1 All 1 Any route or service change 1 5-year plan 1 7D. Type of change or forecast for origin/destination information. Major new service, including fixed guideway extensions 3 Four-step travel model 3 If available 2 Long-range plan 2 Change in route 2 Timed transfer 1 New route or corridor 1 Service to new areas 1 Used to plan, not forecast 1 7E. Type of change or forecast for demographic factors. Change in route 3 Major new service, including fixed guideway extensions 2 New route or corridor 2 Service to new areas 2 Four-step travel model 1 Span of service 1 All 1 Long-range plan 1 Used to plan, not forecast 1 Title VI analysis 1 Large-scale projects that may require environ. analysis 1 Mode choice in model 1 5-year plan 1 Significant service changes 1

42 7F. Type of change or forecast for land use. Change in route 4 New route or corridor 3 Major new service, including fixed guideway extensions 2 Four-step travel model 2 Long-range plan 2 Activity centers 1 All 1 For system ridership 1 Used to plan, not forecast 1 For planning area level forecasts 1 At gross level (residential vs. commercial vs. industrial) 1 Significant service changes 1 7G. Type of change or forecast for economic trends. Major new service, including fixed guideway extensions 2 Four-step travel model 2 Annual budget/forecast 2 Long-range plan 2 New route or corridor 1 All 1 Emerging markets 1 Change in route 1 5-year plan 1 7H. Type of change or forecast for other. Four-step travel model 2 8. What techniques are included in your methodology? If a technique is involved for some types of changes but not others, please indicate. Econometric model 7 20% Four-step travel demand forecasting model 18 51% Regression analysis 7 20% Service elasticities 22 63% Rules of thumb or similar route analysis 28 80% Professional judgment 29 83% Other (please specify): 7 20% Other includes trend analysis (3), GIS/similar routes, internal model based on actual experience, trip generation rates from ITE. 8A. Type of change or forecast for econometric model. Major new service, including fixed guideway extensions 1 New route or corridor 1 Fare changes 1 8B. Type of change or forecast for four-step travel demand model. Major new service, including fixed guideway extensions 8 Four-step travel model 5 Annual budget/forecast 1 For system ridership 1

Large-scale projects that may require environ. analysis 1 Capital projects 1 Modified travel model, not standard approach 1 Long-range plan 1 8C. Type of change or forecast for regression analysis For system ridership 2 Change in route 2 New route or corridor 1 Any route of service change 1 Annual budget/forecast 1 Long-range plan 1 5-year plan 1 8D. Type of change or forecast for service elasticities Any route or service change 9 Fare changes 6 Headway/schedule changes 3 For system ridership 1 8E. Type of change or forecast for rules of thumb/similar routes Any route or service change 5 Change in route 4 Headway/schedule changes 3 Span of service 2 New route or corridor 2 Major new service, including fixed guideway extensions 1 Long-range plan 1 Significant service change 1 8F. Type of change or forecast for professional judgment Any route or service change 5 New route or corridor 2 All 2 Change in route 2 Span of service 1 Headway/schedule changes 1 Check on all forecasts 1 Significant service change 1 8G. Type of change or forecast for other New route or corridor 1 Annual budget/forecast 1 Fare changes 1 Long-range plan 1 For system ridership 1 9. What data sources do you use in developing ridership forecasts? Ridership data from APCs 14 40% Ridership data from recent ridechecks 28 80% Ridership data from the farebox 30 86% Origin/destination data from on-board surveys 22 63% 43

44 Origin/destination data from models 14 40% Census demographic data 23 66% CTPP demographic data 9 26% Existing land use 25 71% Forecast land use 19 54% Economic trends 10 29% Economic forecasts 11 31% Other (please specify): 11 31% Other includes household surveys (3), driver trip counts, special event surveys, on-board ridership surveys, distance walked, frequency of use, point checks for volume at peak load points/CBD cordon points, regional model, FTA Summitt software for BRT, service characteristics (2), fare levels, traffic condition, type of fare paid (if available). 10. Are you satisfied with the reliability of the input data? If not, why not? Satisfied 14 41% Partially satisfied 15 44% Not satisfied 5 15% Total responding 34 100% Issues with data reliability: Ridership 13 65% Origin/destination data 5 25% Demographic data 5 25% General 5 25% Total responding 20 100% 11. How is origin/destination data included in your forecasting methodology? Major part 10 29% Considered, but not a major part 15 43% Not considered 8 23% Depends on time frame/level of analysis 2 6% 12. Do you forecast linked or unlinked ridership? Linked ridership 0 0% Unlinked ridership 24 71% Both linked and unlinked ridership 10 29% 13. Has technology affected your forecasting methodology? If so, how? Yes 22 63% No 13 37% Specific technology: APC 10 56% Farebox upgrade/AFC 5 28% Upgrade/new use for travel model 4 22% GIS 4 22% Improved PCs/software 3 17% AVL/GPS 2 11% Data integration software 1 6% Effect of technology: Data reliability/accuracy 7 30% More data 7 30% Improved analytical tools 7 30%

Greater detail in data 6 26% Data integration from different sources 4 17% Makes origin/destination estimation possible 3 13% Faster analysis time 3 13% Better reporting 2 9% 14. If your system operates more than one mode, do you use different methods to develop forecasts for each mode? Yes 9 45% No 11 55% 15. Do you use different methods for long-range and short-range forecasts? Yes 22 71% No 9 29% 16. Is there an optimal amount of data for your forecasting and planning process? Do you have that amount of data available? Optimal Amount of Data Is Optimal Amount of Data Available? Yes 23 85% Yes 8 26% Sometimes 2 7% Sometimes 6 19% No 2 7% No 17 55% 17. How long does it take to prepare a ridership forecast? What is involved in terms of resources/staff? Time and Effort Required—Short-range Time and Effort Required—Long-range Less than one day 8 32% One day or less 7 47% One to three days 12 48% One to three months 3 20% Two weeks or longer 5 20% Four months or longer 5 33% 18. Are you satisfied with the current ridership forecasting methods? Yes 11 31% Partially 12 34% No 12 34% 18A. If not, what would you like to see improved? Availability and/or accuracy of input data at the appropriate scale 22 81% Less time-intensive methodology 11 41% Inclusion of more predictive variables 11 41% Simplification of the procedures 8 30% Accuracy of the results 16 59% Flexibility to address a wider variety of situations 11 41% Other (please specify): 7 26% Other includes actually having a forecasting process; automated methodology for short-range to minimize steps needed; FTA should allow non-rail to use alternative specific constants; improved accuracy for local bus routes; incorporate TCRP Report 95 into guidelines; more commitment to input data in region and by other operators; more effective methodology— fare and short-term good, long-term and New Starts poor. 19. How do you assess the reliability and value of the methodology? Comparison of actual and projected ridership 31 94% Board understanding and approval 9 27% Other (please specify): 2 6% Other includes meet expectations for growth; pre/post-implementation review plus professional judgment. 45

46 20. Please describe how your agency would forecast ridership for the following scenarios: A. A half-mile rerouting of an existing route to serve a new shopping center Similar conditions/area 13 36% Similar routes/changes 11 31% Current route ridership 9 25% Consideration of through ridership 8 22% Trip generation rate 6 17% Professional judgment 5 14% Would not analyze 5 14% Transfer data/connected routes/ridership shift 3 8% Productivity 3 8% Socioeconomic/demographic data 3 8% Local mode share 2 6% Current/planned development 2 6% Elasticities 2 6% Regression/sketch planning mode 2 6% No. employees 1 3% Population/population density/no. households 1 3% Consultant 1 3% Four-step travel model 1 3% Assume minimum performance standard 1 3% Service level changes 1 3% Travel time 1 3% Understanding it takes time to reach ridership forecast 1 3% Evaluate trip generators/land use within 1/4 mile 1 3% Rider input (public comments/on-board survey) 1 3% Change made to provide access 1 3% B. Extension of an existing route for one mile to serve a new residential development Similar routes/changes 12 33% Similar conditions/area 11 31% Socioeconomic/demographic data 7 19% Productivity 6 17% Trip generation rate 5 14% Assume minimum performance standard 5 14% Would not analyze 5 14% Population/population density/no. households 5 14% Professional judgment 5 14% Local mode share 3 8% Current ridership 2 6% Consideration of through ridership 2 6% Current/planned development 2 6% Regression/sketch planning model 2 6% Four-step travel model 2 6% Evaluate trip generators/land use within 1/4 mile 2 6% Elasticities 1 3% Consultant 1 3% Understanding it takes time to reach ridership forecast 1 3% Rider input (public comments/on-board survey) 1 3% Change made to provide access 1 3% Origin/destination data 1 3% C. Change in headway from 12 to 10 minutes during peak hours Elasticities 12 33% Productivity 10 28% Would not analyze 8 22%

Professional judgment 4 11% Similar routes/changes 4 11% Assume minimum performance standard 3 8% Current ridership 3 8% Socioeconomic/demographic data 2 6% Four-step travel model 2 6% Similar conditions/area 1 3% Evaluate trip generators/land use within 1/4 mile 1 3% Consultant 1 3% Service level changes 1 3% Travel time 1 3% D. Implementation of a new crosstown route to enhance service area coverage and provide more direct connections Similar routes/changes 15 42% Transfer data/connected routes/ridership shifts 8 22% Socioeconomic/demographic data 6 17% Productivity 5 14% Would not analyze 5 14% Four-step travel model 4 11% Similar conditions/area 4 11% Origin/destination data 4 11% Evaluate trip generators/land use within 1/4 mile 4 11% Assume minimum performance standard 4 11% Elasticities 3 8% Professional judgment 2 6% Consultant 2 6% Travel time 2 6% Trip generation rate 2 6% Population/population density/no. households 2 6% Local mode share 2 6% Regression/sketch planning model 2 6% No. employees 2 6% Current ridership 1 3% Service level changes 1 3% Consideration of through ridership 1 3% Current/planned development 1 3% Understanding it takes time to reach ridership forecast 1 3% Roadway congestion 1 3% E. Implementation of a new mode such as BRT Four-step travel model 17 47% Would not analyze/would not implement 7 19% Consultant 6 17% Travel time 4 11% Elasticities 4 11% Origin/destination data 3 8% Similar routes/changes 2 6% Assume minimum performance standard 2 6% Experience in other cities 2 6% Econometric model/new or refined model 2 6% Transfer data/connected routes/ridership shifts 1 3% Productivity 1 3% Evaluate trip generators/land use within 1/4 mile 1 3% Professional judgment 1 3% Service level changes 1 3% 47

48 Roadway congestion 1 3% FTA new starts methodology 1 3% F. Prediction of next year’s ridership as part of the budget process Trend line 21 58% Service level changes 13 36% Fare changes 5 14% Professional judgment 4 11% Would not analyze 4 11% Economic/employment/sales tax revenue changes 3 8% Demographic trends 3 8% Elasticities 3 8% Four-step travel model 2 6% Current/planned development 2 6% Understanding it takes time to reach ridership forecast 2 6% Regression/sketch planning model 2 6% Consultant 1 3% Econometric model/new or refined model 1 3% Done elsewhere in agency 1 3% G. A 10-year ridership forecast as part of a long-range plan Four-step travel model 15 43% Trend line 12 34% Service level changes 8 23% Would not analyze 5 14% Demographic trends 3 9% Fare changes 2 6% Professional judgment 2 6% Economic/employment/sales tax revenue changes 2 6% Consultant 2 6% Elasticities 1 3% Current/planned development 1 3% Understanding it takes time to reach ridership forecast 1 3% Econometric model/new or refined model 1 3% Done elsewhere in agency 1 3% Origin/destination data 1 3% Similar routes/changes 1 3% Productivity 1 3% Population/population density/no. households 1 3% No. employees 1 3% 21. If you could change one aspect of your ridership forecasting methodology, what would you change? Input data 11 44% Methodology 10 40% Improved linkages (GIS, regional indicators) 2 8% Staff expertise/understanding 2 8% Written guidelines 1 4% Allow alternate specific constants 1 4% Based on industry standards and best practices 1 4% 22. Please describe any “lessons learned” that would benefit other transit agencies that are considering changes to their rider- ship forecasting methods. Caution regarding results 7 37% Simplify the approach 4 21%

49 Caution regarding data and application 4 21% Communication and partnering 2 11% Develop local factors 2 11% Simplify the model 2 11% Smaller vs. larger agencies 1 5% Neither overly simple nor overly complex approaches work 1 5% GIS as data integration tool simplifies data management 1 5% Transferability 1 5% Take the time to develop ridership forecasts 1 5% Interpretation/presentation as important as results 1 5% Admit when forecasts are wrong—it’s the best teacher 1 5% 23. Is there another transit system that you suggest we contact for this synthesis project? Yes 10 28% No 26 72% Total responding 36 100% Total agencies named 14 Of named agencies, included in survey 10 71% 24. Would you be willing to participate further as a case study, involving a telephone interview going into further detail on your forecasting methodology, if selected by the TCRP panel for this project? Yes 25 69% No 11 31%