Handbook to Assess the Impacts of Constrained Parking at Airports (2010)

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.

77 The data needed and approaches for measuring whether the strategies implemented to resolve constrained parking conditions achieved the desired outcome on parking and ground access travel behavior are described in this chapter. Evaluating the outcomes of strategies implemented allows an airport operator to judge whether a strategy or multiple strategies were successful within the context of the goals and objectives for the parking program. Furthermore, devel- opment of an understanding of parking and ground access travel behavior can provide an airport operator insight to facilitate the future formulation of strategies, as discussed in Chapters 6 and 7. The primary focus of this chapter is on ways to meas- ure the effectiveness of strategies implemented to resolve constraints on an ongoing basis (strategies to respond to ongoing constraints are discussed in Chapter 5). Short-term strategies (also discussed in Chapter 5), on the other hand, are operational solutions whose success can often be judged during the constrained situation and adjusted accordingly, and data are not typically collected to analyze the effective- ness of such strategies. In some cases, however, such as the use of temporary overflow parking, an airport operator may track information on the overflow parking operation, such as the number of days it is in use and the number of automo- biles parked. The typical data sources described in this chapter are useful for understanding airline passenger parking and travel pat- terns, and are similarly useful for understanding the changes in parking and travel patterns that have occurred following implementation of the strategies to address constrained park- ing. The information in this chapter on measuring the effects of parking strategies provides guidance on how the data iden- tified in this section can be used independently or in their entirety to quantify the results of strategies implemented to reduce parking constraints. Data Sources At a basic level, an airport operator may want to measure how parking behavior changed after strategies were imple- mented. Expanding this evaluation to consider the relation- ship between changes in parking behavior and the use of other ground access modes provides a more holistic view of the effects of strategy implementation. For example, shifts between ground access modes result in changes in levels of vehicle traf- fic on airport and regional roadways, and changes in vehicle traffic effect levels of mobile source emissions generated. Shifts between different parking products and from parking to other modes also affect parking revenues. Understanding these relationships and how a strategy implemented to influ- ence parking behavior under constrained parking conditions may influence behavioral changes beyond the parking system itself is relevant to the evaluation of a strategy or multiple strate- gies. Therefore, an airport operator should consider collecting and maintaining data on the use of other ground access modes that airline passengers use to access and depart the airport. To measure changes in parking and ground access activity, data are necessary from periods both before and after a strat- egy is implemented. Before-and-after data can be collected in one of the following two ways: • Through an ongoing data collection program or • By planning in advance of strategy implementation to col- lect the necessary before data, and by collecting data after the strategy has been implemented. An ongoing data collection program provides the airport operator and others with the flexibility to evaluate the outcomes of strategies that may be implemented within a short timeframe when there otherwise may not have been sufficient time for data collection prior to strategy implementation. However, it should also be recognized that the airport’s existing parking C H A P T E R 8 Evaluating the Effectiveness of Strategies

78 revenue control systems, ongoing survey programs, and other data sources that are part of the airport operator’s ongoing data collection program may not collect, calculate, and readily report the data that may be most effective in measuring the success of a strategy. As a result, the potential limitations of these two approaches to data collection should be considered. Data should be evaluated, as discussed in the next section of this chapter, in consideration of other factors that could influ- ence parking and travel patterns at an airport to understand whether any changes are a direct result of implementing a strat- egy or a result of other factors that may have influenced airline passenger behavior. Collection of information to understand other influences on parking and travel patterns should be con- sidered. What information is relevant to an analysis depends on the unique environment and conditions at a specific air- port. Examples of relevant information include the following: • Changes in local or national economic conditions; • Introduction of low-cost airline service on airline passenger activity (e.g., annual enplanements) and parking activity (e.g., transactions, revenue, space demand); and • Changes in pricing or supply of other ground access modes, including privately operated off-airport parking and HOV modes. Typical data sources for measuring changes in parking activ- ity and other modes, as well as how the data are obtained, include the following: • Parking revenue control system, • Supplemental parking data, • Airline O&D passenger survey data, • Vehicle activity and vehicle occupancy counts, and • Enplaned O&D passenger activity. Parking Revenue Control System The parking revenue control system is the most important data source an airport operator can use for measuring changes in parking activity following the implementation of strategies to address constrained parking. A variety of information can be collected from a parking revenue control system, depend- ing on the sophistication of the system and the reports gener- ated. Sometimes data that are easy to capture and retain if programmed into the parking revenue control system in advance can be costly, time-consuming, or even impossible to retrieve if requested after system programming is complete. Therefore, it is important for members of an airport opera- tor’s staff that have an interest in parking performance from a variety of perspectives to collaborate on determining which data are necessary for developing an understanding of overall changes in parking volume, changes in parking duration, changes in use between facilities, changes in revenue, and other metrics. The data should be provided in formats that lend themselves to data manipulation or meaningful review. Data obtained from the parking revenue control system can be analyzed and postprocessed to provide a range of statistics in a variety of reporting formats. The reports can range from very detailed ticket-level information to high-level monthly and annual statements of parking transactions and revenues. However, for purposes of assessing constrained parking activ- ity, at a minimum, the following basic data should be captured and reported. • Vehicle exits—Vehicle exits or parking transactions by facility provide the airport operator and others with the ability to determine overall changes in parking activity and changes by facility. It may be important to review exit data by hour of the day, day of the week, and month of the year. Parking transaction data may be used to generate bench- marks, such as transactions per space, revenue per transac- tion, and parking transactions per O&D passenger. • Vehicle length-of-stay distribution—A distribution of vehicle exits by length of stay by facility provides an air- port operator with an understanding of potential changes in the duration of facility use following implementation of different strategies, such as parking rate changes, alloca- tion changes, or the introduction of new products. Length- of-stay data should be collected in increments that are meaningful to the specific airport situation. Generally, it is better to collect the information in smaller increments that later can be aggregated during analysis. The length of stay may correspond to the rate structure, or another increment if the rate structure does not provide enough information to understand trip durations. One suggested distribution would be to collect information on stays of 24 h or less in hourly increments, and stays of more than 24 h in daily increments. If special pricing is in effect dur- ing the first hour, such as providing the first 30 min free of charge, the first hour should be segmented into two 30-min increments. • Parking revenue—Parking revenue by facility by length of stay, month, and year provides the ability to generate benchmarks, such as revenue per space, revenue per trans- action, and parking revenue per O&D passenger. Although parking revenue will be generated primarily by resident O&D passengers, this benchmark is somewhat useful in comparing performance between years for similar periods and for assessing the implication of parking rate changes. A comparison of parking revenues generated by the num- ber of exits by length of stay is important for understanding patterns that affect revenue.

79 • Facility occupancy—A count of vehicles in a facility at a specific time or times during the day or evening can help parking staff determine if a facility will become constrained. For example, maximum daily parking occupancy and min- imum daily (typically overnight) occupancy by facility are helpful metrics in assessing overall parking demands by day of the week and seasonally throughout the year. These data also are useful for comparing similar periods before and after strategy implementation. Newer parking revenue control systems are typically able to provide this informa- tion, while older ones can not. Supplemental Parking Data If an airport operator can not collect parking occupancy data from its parking revenue control system, it is recommended that overnight parking occupancy be collected manually by facility. Even when the parking revenue control system provides occupancy data, the collection of overnight counts, including a vehicle license plate inventory, may be useful to validate the data collected from the parking revenue control system (e.g., park- ing management audits). Overnight license plate invento- ries also serve as a backup for estimating length of stay when customers lose their parking tickets and when no automated license plate recognition systems are in use. The license plate inventory data also are useful for helping customers locate their cars when they can not remember where they parked. Airline O&D Passenger Survey Data Airline O&D passenger survey data are important for under- standing how airline passenger ground access travel patterns may have changed in response to strategies implemented to resolve constrained airport parking. Data on private automo- biles that were parked can be collected from the parking rev- enue control system; however, data on the use of other modes, such as private automobiles picking up and dropping off air- line passengers at the curbside, typically are not captured. Data obtained from an airline O&D passenger survey pro- vide a wealth of information on airline passenger character- istics and travel behavior, including use of airline passenger access modes to and from the airport, travel party size, place of residence, place of trip origin, and trip purpose. Data from the O&D survey that are relevant to under- standing the demographics of the parking customer, as well as changes following implementation of strategies to resolve constrained parking, are presented in this section and followed by a discussion of data collection methodology. • Trip purpose—Trip purpose is an important factor in how airline passengers make mode choices. Trip purpose data enable the analyst to determine the proportion of parkers traveling for business and for pleasure. In general, business travelers are more time-sensitive and less price-sensitive than nonbusiness travelers, and their business trips are typ- ically subsidized by their employers. It can be expected that a smaller proportion of parkers traveling for business pur- poses than nonbusiness travelers will divert from terminal area parking when parking rates are increased. • Resident status—Resident status is a determinant of mode choice, since residents have a more extensive knowledge of specific modes available, and only residents may park a pri- vate automobile for the duration of a trip. The importance of understanding customer behavior by resident status and trip purpose was described in Chapter 1. Since most of the long-term parking supply and a large amount of the short- term parking supply is used by resident airline passengers, it is important to estimate the number of resident O&D passengers using the airport. It is equally important to esti- mate the number of nonresident O&D passengers that gen- erated private vehicle trips by greeters and well-wishers, as some of those people use short-term parking. Before- and-after data, including numbers of residents, will pro- vide information on whether changes in parking activity may have occurred as a result of changes in the proportion of residents and nonresidents. • Access mode—Access mode share provides the analyst with information on all travel modes used by airline pas- sengers to access and depart the airport. This information, along with resident and nonresident status and trip pur- pose, allows the analyst to understand mode preferences by resident business, resident nonbusiness, nonresident busi- ness, and nonresident nonbusiness travelers. It is valuable to evaluate this information before and after strategies have been implemented to understand how customers may have shifted modes. Looking at overall mode share as well as mode share by market segment allows the analyst to under- stand if overall mode shifts have occurred as a result of shifts in the proportion of airline passengers by market seg- ment, strategies implemented to influence behavior, or potentially for other reasons. The number of passengers using each mode can be estimated by applying the mode share for the time period of the survey to the volume of O&D passengers for a comparable time period, such as average daily O&D passengers. • Private automobile disposition—Survey data should capture the proportion of passengers using private auto- mobiles that have been picked up and dropped off by greeters and well-wishers versus those that parked their private automobiles for the duration of their trips. The pickup and drop-off customers can be further segmented into short-term parkers and those who used the curbside only. Long-term parkers can be segmented by parking facility used.

80 • Airline passenger travel party size—Travel party size data enable the analyst to estimate the number of single-party vehicle trips generated by airline passengers to and from the airport on a daily basis. For a long-term parker, the number of vehicle trips generated would be calculated by dividing the number of passengers that accessed the airport by the “private automobile parked for the duration of the trip” mode by travel party size. For an airline passenger dropped off and picked up by a single-party vehicle, the number of vehicle trips generated would be calculated by dividing the number of passengers dropped off by private vehicle by travel party size and then by multiplying the resulting number by two to account for the vehicle trips departing from and returning to the airport that are not transporting airline passengers. It can be assumed that the data from the enplaning survey can be doubled to arrive at the number of daily trips. This methodology (discussed later in this chapter in the section on measuring the effects of parking strategies) also allows the analyst to estimate the number of rental car trips, taxicab trips, and single-party limousine trips. • Length of stay—Length of stay information allows the ana- lyst to determine trip durations for different customers by customer segment, mode, private automobile disposition, and parking facility. Data from the parking revenue control system provide length-of-stay data, but do not reveal length of stay by customer segments or by airline passenger travel party size. Surveying enplaning passengers is more efficient than sur- veying deplaning passengers because enplaning passengers have time in the terminal and in hold rooms prior to board- ing their flights whereas deplaning O&D passengers do not. An enplaning passenger survey is typically administered beyond security. It is often assumed that an airline passen- ger’s mode of airport egress is the same as the mode of access, which would almost always be the case for the airline passen- ger who parks an automobile at the airport before departing on the airline trip. This simplifying assumption significantly reduces the effort and costs associated with administering air- line passenger surveys. An airport operator may collect O&D survey data period- ically, such as every 2 to 5 years, to monitor airline passen- ger demographics and travel behavior. Airport operators also may collect such data for a specific purpose, such as a master plan. If an O&D survey is conducted at regular inter- vals, it can be used to measure before-and-after behavior subsequent to implementation of a strategy, with the recog- nition that other changes may have occurred during the time period between surveys that also may have affected travel behavior. It is important for questions and answer choices to be similar each time the survey is administered so that the results from survey to survey are comparable. It is also important to carefully consider the timing of survey implementation, such as collecting data at a “typical” travel time, when business and nonbusiness travel is not unusually high or low. Airline O&D Passenger Survey Conducted by an Airport Operator (Los Angeles World Airports) Los Angeles World Airports, the operator of Los Angeles International Airport (LAX), administers an O&D survey of departing airline passengers every 4 to 5 years. The most recent survey was con- ducted in 2006. The sampling occurred during August and October to capture peak and “typical” travel periods. Survey responses were collected from 18,458 O&D airline passengers. (22) Airline O&D Passenger Survey Conducted by an MPO (Metropolitan Washington Council of Governments) The Metropolitan Washington Council of Governments, the MPO for the greater Washington, D.C. area, conducted an O&D survey of departing airline passengers at Ronald Reagan Washington National Airport (DCA), Washington Dulles International Airport (IAD), and Baltimore/Washington International Thurgood Marshall Airport (BWI) during a 2-week period in October 2007. Approxi- mately 19,100 usable responses were obtained, with 24%, 39%, and 37% of responses, respectively, from airline passengers at DCA, IAD, and BWI. The survey was jointly funded by the Metropolitan Washington Airports Author- ity and the Maryland Aviation Administration. It is the eighth survey that has been conducted since 1981. (8)

81 For more information on conducting an O&D survey, please refer to ACRP Report 26: Guidebook for Airport-User Survey Methodologies, which is the guidebook developed in support of ACRP Project 03-04. (21) Vehicle Activity and Vehicle Occupancy Counts For airports with congested roadway and curbside systems, the airport operator may want to understand if parking strategies have caused a mode shift that resulted in a diversion of vehicle traffic from the curbside (i.e., passenger pickup and drop-off) to the parking facilities or vice versa. Although some airport operations personnel may have an intuitive feel for this information, supporting data are derived either through an O&D passenger survey, as described in the previ- ous section, or estimated from a variety of data sources. The latter approach will require the distillation of data from dif- ferent sources and time periods, and supplemental data from manual or automated traffic counts. Vehicle count data by mode and data on the number of occupants by mode are nec- essary for estimating airline passenger mode share. Although some airports use automatic vehicle identifica- tion (AVI) systems to collect detailed activity data associated with commercial vehicle traffic using the airport curbsides, these systems track only those vehicles that are mounted with a radio-frequency-based transponder. However, these AVI data, in combination with roadway traffic counts of all vehi- cles accessing the airport, have been used to obtain estimates of private automobiles and other vehicles that are not tracked by the AVI system. Vehicle Traffic Generated by Mode Automatic traffic recorders (ATRs) can be temporarily installed on the airport roadway system, including the termi- nal curbsides, parking access roadways, recirculation road- ways, and in the vicinity of cell phone lots to collect vehicle counts for time periods before and after implementation of a strategy. Permanent count stations using imbedded loop detectors or other technology also have been installed by some airport operators interested in obtaining an ongoing database of traffic activity. These traffic counting systems can be config- ured to estimate vehicle classification based on the number of axles per vehicle, but are limited in their ability to identify spe- cific modes. Other means can be used to collect mode-specific distribution to supplement the vehicle counts as follows: • An AVI system tracks trips by commercial vehicles that have been mounted with a radio-frequency-based transponder (e.g., toll tag) and can provide the commercial vehicle dis- tribution by mode for all vehicles affixed with transpon- ders. Commercial vehicle trips can be subtracted from the total traffic counts obtained using ATRs or permanent imbedded loops to provide the share of traffic by private automobiles and other vehicles not tracked by the AVI sys- tem, which includes maintenance vehicles, rental cars, and “nontenant” commercial vehicles. Traffic counters and AVI readers should be located in the same areas to ensure that the data are comparable. • Numbers and types of commercial vehicles can be estimated through other tracking mechanisms, such as taxicab and limousine dispatch counts and HOV schedules, if an AVI system is not available. • Manual vehicle classifications by mode can be obtained. Per- sonnel with handheld computer devices or clipboards would be stationed at different points on the roadway and terminal curbsides to count vehicles by a variety of classifications. These methods to capture mode distribution do not allow for the classification of rental cars and private automobiles. Activity at a terminal curbside that appears to be generated by private automobile will typically include (1) rental car cus- tomers dropping off members of their travel party at the curb- side prior to returning the rental car, (2) long-term parkers and short-term parkers dropping off members of their travel party prior to parking the automobile, (3) recirculating vehicles that make multiple passes of the curbside in advance of meeting passengers for pickup, and (4) bypassing vehicles accessing an adjacent terminal that drive past the terminal while traveling to or from the adjacent terminal. An O&D survey is more effec- tive at estimating the share of parking customers versus those that are picking up and dropping off passengers, and estimat- ing the share of rental cars versus private automobiles. Finally, nonpassenger vehicles at the terminal curbsides, such as enforcement personnel, employees, airport mainte- nance, and delivery vehicles, will be captured in the non-AVI automatic traffic counts and will need to be separated from the vehicle traffic related to airline passengers. Only some of this vehicle traffic can be identified through a manual classification count. Vehicle Occupancy Counts Vehicle counts may provide a distribution of vehicles by vehicle type or mode, but vehicle occupancy is necessary to estimate a passenger mode-share distribution. If not estimated through an O&D survey, manual counts can be conducted before and after implementation of a strategy. Manual vehicle occupancy counts can be conducted through observations— personnel with handheld computer devices or clipboards stationed at different points on the roadway and terminal curbsides, counting occupants or numbers of passenger entries and exits per vehicle by vehicle classification. This count should be conducted as part of the vehicle classification count. Com- mercial vehicle operators could be recruited to keep a log of the numbers of passengers per vehicle during the same time period.

82 Enplaned O&D Passenger Activity Enplaned O&D passenger activity for the periods before and after the strategy was implemented, preferably on a monthly basis, should be obtained and used to normalize changes in parking activity and changes in the use of other ground access modes. Measuring Effects of Parking Strategies Using the data sources identified in the previous section, this section describes ways to measure changes in parking and ground access behavior following the implementation of strategies to address constrained public parking. Often an air- port operator will adopt more than one strategy to address parking constraints, so the resulting changes in behavior will be difficult to attribute to an individual strategy; therefore, it may be sufficient to quantify whether a combination of strategies achieved the desired changes. The airport environment is complex and, as discussed in the previous section, parking and ground access activity are dependent on many factors. Changes in activity following the implementation of strategies to address constrained parking may indicate their effectiveness or ineffectiveness when, in real- ity, other factors also influenced the outcome, thereby temper- ing the effects of the strategies. If the airport operator wishes to understand the cause and effect of strategies rather than just overall changes in activity, information on other factors that may have influenced the rates of change in activity should be considered, such as local or national economic conditions, introduction of low-cost airline service, changes in pricing, or supply of other modes, including privately operated parking and HOV modes, and other factors. The strategies described in Chapter 5 for addressing ongo- ing parking constraints will have varying levels of effectiveness in balancing demand with supply when applied at a given air- port, depending on the circumstance and how the strategy is formulated and applied. The primary goal and measure of the success of a strategy or strategies is the resolution of the park- ing constraint. However, it is likely that the strategies will result in other outcomes that affect airport operations and that may not be acceptable to the airport operator, or that may be inconsistent with the goals and objectives for the ground access program or the airport. In other words, in solving the constrained parking problem, another problem may be created. Changes in activity and related impacts that are a logical conse- quence of implementing strategies to address ongoing parking constraints include the following impact categories: • Public parking activity, • Financial performance, • Vehicle traffic volume, • Emissions generated, • Mode-share distribution, and • Customer service. Potential impacts specific to each strategy were described in Chapter 5 in the discussion of considerations. Approaches for measuring the changes in activity by impact category are presented in the remainder of this chapter. The airport oper- ator should consider how changes in each category relate to the goals and objectives for the parking program, and with the airport’s guiding principles, to determine the acceptabil- ity of the strategies implemented. It can be expected that some strategies may have a more dramatic short-term impact as customers are introduced to the change, and a lesser impact as customers become accus- tomed to the change. It is recommended that activity data be collected and compared for a minimum of 1 year prior to the change and 1 year following the change to understand short- term impacts and how behavior reaches equilibrium over time. By continuing to compare activity beyond the first year after the change, the airport operator can monitor the ongoing effectiveness of the strategy. Similar time periods should be selected in the before-and-after comparisons. Increments of time that are meaningful for comparison will depend on the constrained parking situation, but monthly, quarterly, and annual increments may be useful to evaluate. For example, if a reallocation of parking and the introduction of a new park- ing product were introduced at an airport in June 2005, it would be prudent to compare monthly parking activity from July 2004 through June 2005 to parking activity from July 2005 through June 2006, as well as changes in airline passenger activ- ity during those periods. Data on changes in parking activity for July 2006 through June 2007, normalized for changes in air- line passenger activity, would provide insight into the ongoing effectiveness of the strategies. Financial data should be col- lected for the same time periods to measure financial impacts. For vehicle traffic impacts, less data may be available, but the time periods compared should be similar for the before-and- after periods to ensure valid comparisons. Public Parking Activity The primary goal of strategies to address constrained pub- lic parking is to reduce or resolve the parking constraints. Analysis of parking activity data will reveal whether changes resulted following the implementation of a particular strategy. This analysis is accomplished through a comparison of park- ing activity before and after the implementation of strategies for similar periods for the total public parking system and by facility or product, while paying particular attention to periods that were typically constrained.

83 The following benchmarks are useful in this comparison, and all should be evaluated in the context of changes in air- line passenger activity and other factors that may influence passenger parking behavior beyond the specific strategy or strategies an airport operator has implemented. • Vehicle exits—A change in the numbers of vehicles exit- ing the parking facilities, in total, in specific facilities, or between facilities, indicates that the strategy implemented may have influenced passenger choice of parking facilities, or the decision to use parking facilities. • Average vehicle length of stay—A change in the average vehicle length of stay within a facility or complementary changes among facilities indicates that the strategy imple- mented may have influenced passenger choice of parking facilities. For example, a strategy may be implemented to discourage long-term parkers’ use of a facility designated for short-term parking. A decrease in the average length of stay in that facility would indicate that the use of the facil- ity by long-term parkers was influenced by the strategy (or potentially by other factors). Furthermore, similar analysis of average length of stay in other parking facilities may indicate which facility the long-term parkers were diverted to, or if they are no longer using public parking. • Vehicle length-of-stay distribution—Similar to the aver- age length of stay analysis discussed above, a more detailed analysis of the length of stay distribution may provide more insight into parkers’ behavioral changes within a facility and between facilities. • Facility occupancy—A comparison of percent occupancy of parking facilities during overnight and peak periods before and after implementation of a strategy provides an indication of parkers’ behavioral changes within a facility and between facilities. This information also can supple- ment and help support conclusions drawn from length-of- stay data. • Amount and duration of overflow parking used—A com- parison of the number of times overflow parking was used, the duration of its use, and number of parkers using the overflow parking facility before and after implementation of the strategy provides an indication of whether the strat- egy minimized (or eliminated) the times that an airport parking system is in overflow condition. • Level and duration of additional resources deployed—A comparison of resources deployed (e.g., extra staff, tempo- rary shuttle services) to manage parking constraints from before and after the implementation of a strategy provides an indication of whether the strategy minimized (or elimi- nated) the amount of time that an airport experiences con- strained parking conditions. • Number of parking facility closures—A comparison of the number of times or the durations of parking facility closures before and after the implementation of a strategy provides an indication of whether the strategy minimized (or eliminated) the times that a parking facility reaches capacity. Financial Performance Changes to gross parking revenue and net parking revenue will affect the airport operating environment. Public parking revenue is a significant source of nonaeronautical revenue at most airports because it often generates revenue in excess of cost. Portions of the net revenue from public parking may be used to cover other airport expenses, lower airline rates, or fac- tor into calculations for airport financing for capital programs. A common problem for the airport operator is to resolve constrained parking while minimizing the financial impact of implementing the strategy to relieve constrained parking. Scenarios that could result in an unacceptable loss of revenue from unanticipated consequences may include the following: • Loss of public parking customers to privately operated off- airport parking facilities; • Shift of public parking customers from higher priced pub- lic parking to lower priced public parking; • Shift from public parking to other access modes, such as curbside pickup and drop-off by private automobile, taxi- cab, and transit; and • Construction of a new public parking facility (e.g., a park- ing garage) resulting in substantially more capital costs or operating costs that are not offset by a corresponding increase in parking revenue. Overall changes in gross parking revenue and net parking revenue will be the most important benchmarks for the air- port operator, and should be evaluated in conjunction with changes in parking activity, as described above, and other fac- tors that may have influenced changes in behavior. Changes in revenue for the period before and after the implementation of strategies should be compared to changes in numbers of airline passengers and changes in rates or available parking supply at a certain rate level. Additional financial metrics to consider and evaluate include the following: • Gross revenue per space or gross revenue per exit or transaction—Comparison of changes in revenue gener- ated per parking space or per exit provides an understand- ing of changes on a unit basis for total parking activity and by facility. • Net revenue per space or net revenue per exit or trans- action—Comparison of total net revenues of capital and operating costs to adopt the strategies and manage con- strained parking prior to strategy implementation can be

84 used to understand the financial effects of a strategy on a unit basis for total parking activity and by facility. • Gross revenue by facility or parking product—Revenue by facility or parking product should be compared to changes in parking activity and O&D passenger activity to under- stand the relationship between changes in activity and rev- enue. For example, in a particular public parking facility, parking exits may have decreased 2%, and revenue may have increased 4%, while O&D passenger activity increased 1%. These results would be compared to similar changes in other public parking facilities, and for the total parking supply, to understand the influence of the strategy on financial performance and if the financial and activity changes are acceptable. If the strategy implemented is a rate change, the data also can be used to develop parking elasticities. • Gross revenue by facility or parking product by duration— Revenue by duration is useful when determining if a new parking rate schedule was successful in changing parking behavior for defined durations. For example, long-term parkers at an airport may be using short-term facilities, spaces for people who desire to park for less than an hour. To remedy this, the airport operator may have increased the rates for parking in the facility for more than 6 hours to a significantly higher rate than the daily rate in the long- term parking facility. By analyzing parking revenues by duration, the airport operator can determine if the strat- egy implemented to discourage long-term parkers using the short-term facility has been effective and if the strat- egy has resulted in a change in the revenue streams for the facility. If applied across all parking facilities, the analysis of revenue by facility by duration can be used to under- stand the financial impact of the strategy for all parking products. For each length-of-stay category by facility, a comparison of the contribution of each category to rev- enue and to activity allows an understanding of which users have the greatest effects on financial performance and facility use. • Net revenue by facility or parking product—A compari- son of revenues net of capital and operating costs to adopt the strategies and to manage constrained parking prior to and following strategy implementation sheds further light on the revenue analysis discussed above, and allows the air- port operator to evaluate the overall financial effect of the strategy compared to changes in parking activity. If unacceptable changes in revenue can be isolated to users exhibiting certain characteristics or to a specific park- ing product, this may provide the information the airport operator needs to modify the strategy to lessen the revenue impacts, or the airport operator may decide to reverse the strategy and try a different approach to address the parking constraints. Vehicle Traffic Volume Airline passenger airport access and egress trips affect the flow of vehicle traffic on the airport roadway system and ter- minal curbside, the roadway system surrounding the airport, and the regional roadway system. A relationship exists between constrained public parking and airline passenger shifts to other modes, particularly other single-party modes that offer the customer a door-to-door experience between the trip ori- gin or destination in the region and the airport. Shifts between public parking and other modes affect vehicle traffic volumes. Strategies to resolve constrained public parking could improve vehicle traffic flows on the airport and in the region if trip vol- umes were reduced. Conversely, vehicle traffic would worsen if strategies increase trip volumes. Airline passengers shifting from parking for the duration of their trips to being picked up and dropped off by single-party modes contribute to an increase in trip volumes. Airline pas- sengers shifting from single-party modes to high-occupancy modes, such as buses and shared-ride vans, contribute to a decrease in trip volumes. Single-party modes that the long-term parker may con- sider for airport access in a constrained parking environment typically include the following pickup and drop-off options: • Private automobile, • Taxicab, and • Single-party limousine. Nonresident airline passengers may consider shifting between pickup and drop-off by private automobile, taxicab, single-party limousine, and rental car. Table 27 describes the characteristics of enplaning passen- gers’ airport access trips by mode, including the number of vehicle trips generated to and from the airport for the trip. For example, the long-term parker that shifts to being picked up and dropped off by private automobile for future airline trips is generating twice the number of vehicle trips than were gen- erated when the vehicle was driven to the airport and parked for the duration of the trip. For the enplaning passenger party that is being dropped off by private automobile, two one-way vehicle trips are generated (i.e., the driver drops off the pas- sengers at the terminal and departs the airport without the air- line passengers). Another round trip must be made to pick up the passenger party upon its return. The number of vehicle trips generated by a long-term parker that shifts to the taxicab or single-party limousine mode also will increase, because not all taxicabs and single-party limousines transport passengers both to and from the airport during the round trip. Using the logic from Table 27, Table 28 shows the general changes in vehicle traffic volumes due to shifts between modes. If trip volumes increase and vehicle traffic becomes con- gested (or more congested) following the implementation of

Airline Passenger Vehicle Occupancy 1 Vehicle Trips Generated Mode Travel Parties Served Trip to Airport Trip from Airport Per Enplaning Travel Party Per Enplaning Airline Passenger Curbside Drop-off ? Private Automobile Drop-Off (Curbside Only) Single Travel Party Size 0 2 2 ÷ Party Size Yes Private Automobile Drop-off (Short-Term Parker) Single Travel Party Size 0 2 2 ÷ Party Size Potentially 2 Long-Term Parker (Terminal Area) Single Travel Party Size Not Applicable 3 1 1 ÷ Party Size Unlikely Long-Term Parker (Remote, On- or Off-Airport) Single Travel Party Size Not Applicable 3 1 1 ÷ Party Size Potentiall y 2, 4 Rental Car Single Travel Party Size Not Applicable 1 1 ÷ Party Size Potentiall y 2, 4 Taxicab Single Travel Party Size 0 or Party Size of New Party 5 1 or 2 6 (1 or 2) 6 ÷ Party Size Yes Single-Party Limousine Single Travel Party Size 0 or Party Size of New Party 5 1 or 2 6 (1 or 2) 6 ÷ Party Size Yes Shared-Ride Van Multiple Total Airline Passengers Served 0 or Total Airline Passengers Serve d 5 (1 or 2) 6 ÷ Number of Travel Parties Served (1 or 2) 6 ÷ Airline Passengers Yes Bus (Transit or Privately Operated) Multiple Total Airline Passengers Served Total Airline Passengers Served 1 ÷ Number of Travel Parties Served 1 ÷ Airline Passengers Yes Subwa y or Light Rail to Terminal Multiple Total Airline Passengers Served Total Airline Passengers Served 0 0 No Notes: 1 Excludes vehicle occupants who are not enplaning passengers (e.g., well-wishers). 2 Drivers ma y drop off som e or al l of the passenger party me mb ers at the curbside before proceeding to parking. 3 Vehicle does not depart from the airport until the enplaning travel party returns from airline travel. At that time, occupancy is equal to travel party size for the deplaning trip. 4 Consider trips associated with shuttle drop-off of passengers at the curbside. Shuttle trips are designed to serve multiple t ravel parties. 5 Mode may transport new deplaning passengers from the airport. The proportion of trips for taxicabs, limousines, and shared-ride vans that transport airline passengers on both access and egress trips versus those that transport airline passengers on one of the trips (access or egress) should be considered. Empty trips are less likely for shared-ride vans than for taxicabs and single-party limousines because of the business model of shared-ride van operators. 6 Two trips if the vehicle departs the airport without airline passengers and one trip if the vehicle returns to a staging area prior to returning to the curbside to pick up a deplaned passenger for an outbound trip. Source: Ricondo & Associates, Inc., and DMR Consulting, November 2, 2009. Table 27. Typical vehicle traffic impacts of access modes transporting enplaning passengers to an airport.

86 strategies, airport operations personnel will have a general feel for the magnitude of the impacts. Some changes may be more pronounced in the short-term and become less notice- able after customers become accustomed to the strategies. If trip volumes decrease and vehicle traffic flow improves, the airport operator may still wish to understand the reasons for the shift. At many airports, the airport operator is account- able to the surrounding communities for the impacts gener- ated by airport activity, and information on the changes in impacts would be useful. Benchmarks that can be compared over similar timeframes before and after the strategies have been implemented to assess the impacts of the strategies on vehicle traffic include the following: • Number of total parking exits—A change in total vehicle parking exits without a comparable change in total airline passengers (after accounting for seasonality) indicates that passengers have likely shifted to or from other modes, or the average occupancy per vehicle (party size) has changed. For example, if the party size has not changed and the number of airline passengers has not changed, but total vehicle park- ing exits have increased, customers have shifted to use of the parking facilities from curbside pickup and drop-off by private automobile and from other modes. Conversely, if vehicle exits have decreased without changes in party size and passenger numbers, customers have shifted to curbside pickup and drop-off by private automobile or other modes. In analyzing these data, it is important to recognize that mode shifts and resulting parking characteristics change seasonally; therefore, care should be taken to normalize the data to isolate the reasons for the change. • Number of exits by parking facility—A shift in vehicle parking exits from one parking facility to another also will have vehicle traffic implications. For example, customers shifting from terminal area parking to remote parking may result in an increase in the number of parking customers dropping off members of their travel party at the terminal curbside before driving to the remote lot. This results in additional vehicle traffic at the terminal curbside. • Private automobile activity on the airport roadway sys- tem, at terminal curbsides, and in cell phone lot—An increase in private automobile activity at the terminal curb- sides indicates an increase in the pick up and drop-off mode, or an increase in drivers dropping off travel party members prior to parking their automobiles for the duration of the airline trips. This analysis should be considered in con- junction with vehicle exit and length-of-stay distribution data from the parking revenue control system. • Changes in taxicab and limousine activity—Changes in taxicab and limousine activity provide additional insight into modes to which parking customers may have diverted. • Airline passenger mode choice distribution—A compar- ison of mode share by the four primary passenger market segments—resident business, resident nonbusiness, non- resident business, and nonresident nonbusiness—will Current Mode Previous Mode Private Automobile (Pick Up and Drop- Off) Private Automobile (Long-Term Parker) Rental Car Taxicab Single-Party Limousine Shared-Ride Va n Bus, Light Rail, Subway Private Automobile (Pickup and Drop - Off) Decrease Decreas e 1 Decreas e Neutral or Decrease Decrease Decrease Private Automobile (Long-Term Parker) Increase – Increase Increase Decrease Decrease Rental Car – – – – – – Taxicab Increase Decrease – – – – Single-Party Limousine Neutral or Increase Decrease – – – – Shared-Ride Van Increase Increase – – – – Bus, Light Rail, Subway Increase Increase – – – – Notes: – Mode shift unlikely to occur as a result of constrained parking or implem entation of strategies to address constrained parking. 1 Shift prim arily applies to nonresidents. Source: Ricondo & Associates, Inc., and DMR Consulting, Novem ber 2009. Table 28. Directional changes in vehicle trip generation from mode shifts resulting from constrained airport parking.

87 provide additional insight into mode choice. Mode share for all O&D passengers can be used to estimate access trips to and egress trips from the airport. To gain an under- standing of the nature of the changes in overall mode share, the analyst must review the changes in the propor- tion of resident and nonresident airline passengers and business and nonbusiness airline passengers to determine if overall changes in mode share were caused by shifts in the proportion of user groups or changes in mode prefer- ence among user groups. Vehicle trips for single-party modes can be estimated by applying the mode-share data to a count of O&D passengers, such as average daily O&D airline passengers. Trips generated by other modes, such as shared-ride vans and scheduled HOV modes, can be obtained from data sources maintained by the airport operator, such as the AVI system, trip dispatch logs, or published schedules. However, the number of scheduled HOV trips is unlikely to change as a result of parking strate- gies implemented, unless it was part of the strategy. The num- ber of shared-ride vans may change as a result of shifts to or from the private automobile. Changes in vehicle trips by passengers being picked up and dropped off by private automobile, taxicab, and single-party limousine will represent a similar change in vehicle trips at the terminal curbside, with the exception of recirculating trips. Vehicle trips can be estimated for each of the single-party passenger modes using Equation 1. The data for the calcula- tion often are obtained from O&D passenger surveys. Since the time interval between administration of O&D surveys will often be 3 or 4 years, any changes in parking and ground access patterns will likely be influenced by multiple factors during the elapsed time. Therefore, changes in mode share may be caused by factors in addition to, or other than, the strategies imple- mented to address constrained parking. Conversely, if mode shares did not change, it may be because of factors in addi- tion to, or other than, strategies implemented to address constrained parking. The calculation of vehicle trips generated by single-party modes at an airport is shown in an example presented in Table 29. where VXZ = Total estimated vehicle trips generated by mode X for period Z. YZ = Enplaning O&D passengers for period Z (for exam- ple, average day, peak day average month, annual). MX = Mode-share percentage for mode X. PX = Average party size for mode X. TX = Average vehicle trips per enplaning passenger trip. Includes the trip to the airport plus average empty trip rate. For long-term parkers, the empty trip rate is 0. For drop-off by private automobile, the average empty trip rate is 1, as all such automobiles depart from the airport without members of the enplaning passenger party. For taxicabs and limousines, the rates will typically be between 0 and 1. Airline passen- ger survey data, as well as dispatch information and information about vehicle regulations, will assist the airport operator in determining the empty trip rate. Explanation: YZ  MX = Estimated enplaning O&D pas- sengers using mode X for time period Z; ((YZ  MX) ÷ PX) = Estimated number of automo- biles carrying enplaning O&D V Y M P T Equation 1)XZ Z X X X= ( ) ÷( )( )   2 ( Single-Party Mode (X) Mode-Share Percentage (M) Average Party Size (P) Vehicle Trips per Enplaning Passenger Trip (T) Calculation ((((Y Z * M X ) ÷ P X ) * T X ) * 2) Total Estimated Average Daily Vehicle Trips (V) Private Automobile (Pick Up and Drop-Off) 32% 2.1 a 2.0 (((17,400 * .32) ÷ 2.1) * 2) * 2 10,606 Private Automobile (Long-Term Parker) 19% 1.4 b 1.0 (((17,400 * .19) ÷ 1.4) * 1) * 2 4,723 Rental Car 21% 1.8 a 1.0 (((17,400 * .21) ÷ 1.8) * 1) * 2 4,060 Taxicab 11% 1.3 a 1.5 (((17,400 * .11) ÷ 1.3) * 1.5) * 2 4,417 Single-Party Limousine 6% 1. 6 a 1.8 (((17,400 * .06) ÷ 1.6) * 1.8) * 2 2,349 Notes: (Y Z )—17,400 average annual daily enplaning airline passengers for Year Z. a Average party size for this mode does not include the driver of the vehicle. b Average party size for this mode includes the driver of the vehicle. Source: DMR Consulting, Novem ber 2009. Table 29. Example calculation of vehicle trips for single-party passenger modes.

passengers using mode X for time period Z; and (((YZ MX) ÷ PX) TX) = Estimated number of vehicle trips generated by enplaning O&D passengers using mode X for time period Z, including empty trips. Multiply the above by 2 to estimate the number of vehicle trips generated by enplaning and deplaning O&D passengers, assuming enplaning and deplaning trip behavior is generally the same. VMT is another measure of the effects of airport trips generated by airline passengers. Vehicle miles traveled for the periods before and after the parking strategies were implemented can be calculated based on the data from O&D surveys on local trip origins and the total trips estimated following the example calculation shown in Table 29 for an enplaning passenger. A private automobile drop-off trip from a point of origin 20 mi from the airport generates 40 VMT in the region and a drop-off trip from a point of origin 35 mi from the airport generates 70 VMT. To calculate the change in VMT from shifts in the mode-share distribution following implementation of parking strategies to relieve parking con- straints, a simple assumption would be to apply the average distances traveled from the O&D airline passenger survey by mode to the changes in the numbers of private automobiles that are parked for the duration of the airline passenger’s trip and used for picking up and dropping off passengers, as well as taxicabs and single-party limousines. For private automo- biles picking up and dropping off passengers, and the other on-demand modes, the average number of empty trips must be factored into the calculation (as shown in Equation 1 and Table 29). Changes in airline passenger activity should be factored into the analysis to normalize the results and help ensure that the results of the strategy are not skewed based on fluctuations in the numbers of airline passengers. Emissions Generated The change in emissions generated on the airport, in the vicinity of the airport, and in the region by automobiles trans- porting airline passengers to and from the airport as a result of strategies implemented to address constrained airport park- ing may be a consideration for the airport operator when determining the acceptability of a strategy. The VMT can be used along with information about pol- lutants emitted by vehicles in the airport’s catchment area to determine the changes in environmental impacts. A data source for estimating vehicle emissions is the Emissions and Dispersion Modeling System (EDMS), a model approved for use by the FAA and EPA for estimating the air quality impacts of airport emissions sources, including ground access vehi- cles. (23) In EDMS, vehicle emissions factors are estimated from the latest version of the EPA’s mobile source emissions model. Mode-Share Distribution The airport operator may be concerned about how strate- gies implemented to address constrained parking affect the mode-share distribution of airline passengers, particularly if the use of HOV modes is encouraged. Information from an O&D airline passenger survey from before and after the park- ing strategies have been implemented is the best source for keeping track of airline passenger mode-share distribution, and should be evaluated in total and for the four airline pas- senger market segments, as described earlier in the section on “Vehicle Traffic Volume.” Customer Service The customer service implications of strategies imple- mented to address constrained parking will be of concern to the airport operator. If the primary intent of implementing the strategies is accomplished (improving or resolving constrained parking), it would seem that more customers would be pleased or fewer would be displeased. Strategies aimed at resolving the problem for some users may create inconveniences for others. This inconvenience will be evident from changes in parking activity, increased vehicle traffic, or decreased revenue. Changes in the customer service category are more difficult to measure than changes in other categories. One indication of a change that affects customer service is if the airport operator receives more complaints following the implementation of strategies. At some airports, customer satisfaction surveys are administered on a regular basis. If questions are included related to the availability of parking, pricing, and other fac- tors, as well as questions on other access modes, this informa- tion should be compared for similar periods before and after strategies are implemented. 88