Passenger Value of Time, Benefit-Cost Analysis and Airport Capital Investment Decisions, Volume 1: Guidebook for Valuing User Time Savings in Airport Capital Investment Decision Analysis (2015)

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.

3 FIVE-STEP PROCESS Step 1: Screen Project for Applicability It takes extra time and effort to adjust benefit-cost studies to incorporate different values of time savings for different segments of the air passenger trip, and that extra effort is not always necessary. So the first step is to apply a project “screener” to identify whether the extra effort is necessary that allows the analyst to identify likely passenger travel time savings. In this way, the analyst can save time by determining immediately whether or not it is worthwhile to use the different values of time for different segments of the total trip described in this guidebook for a given project. The screening process focuses on passenger travel time savings. Most airport improvement projects are actually packages or combinations of elements. Some are designed to reduce time delays for air passengers, while others are designed to reduce operating cost, enhance safety or amenities, or otherwise improve operational efficiency or throughput. As a result, many projects will involve improvements that do not necessarily save time for air passengers. The Project Screener is shown on the next page in Table 4. The screener provides a list of project types that may be elements of a broader project package. The list of project types is intended to be illustrative and not exhaustive. The screener also lists classes of potential actions and shows whether or not the project is likely to save time for passengers. The analyst should identify the types of project elements and associated actions that are applicable, and then consider whether or not that action appears to warrant analysis of passenger time savings. The analyst should focus on the aspects of the project that are likely to generate relatively large travel time savings. 3 Page 15

Table 4: Project Screener Project Type Action Type Effect on Time Delay Yes (potentially) No Airport (non-terminal) - Airside Air Traffic Control (usually not an airport responsibility) Upgrade Incresases airport capacity, which may reduce aircraft delays Aircraft Ground Control (usually not an airport responsibility) Capacity Save time in waiting to take off Runway New, expanded or enhanced Reduce aircraft delay via higher throughput rates/use of larger aircraft, more (longer distance) direct flights Maintain, repave X Enhance (safety areas, lighting drainage, grading) X Taxiways Expand or improve Reduce aircraft delay by enabling faster aircraft exitfrom runways or reducing taxiway congestion Maintain, repave X Apron Area, Taxilanes, and Aircraft Gate Positions Expand area Reduce aircraft ground delay waiting for a gate to become available Maintain, repave X Hangers, Tie-Downs Add number X Maintenance Facility Expand X Cargo Complex Handling Capacity (aircraft, tonnage) X Airport (non-terminal) – Groundside Access Road to Airport Add lanes, increase travel speed Reduce congestion delay, save travel time People Mover Access to Airport Terminal (from Transit, Rental Car or Parking Facilities) Construct; or add frequency, increase speed Reduce waiting delay, save travel time Parking Lot/ Garage Capacity, travel time, driver information systems Reduce in-vehicle search time for parking space or out-of-vehicle walking time from garage to terminal Central Bus or Train Transfer Facility to/from Airport Terminal Capacity, travel time Reduce wait time, walk time Airport Circulation Improvments for Taxis Add capacity Reduce wait time, walk time Drop-off & Pickup Areas by Terminal Curbfront Add Capacity Reduce wait time, walk time Page 16

Project Type Action Type Effect on Time Delay Yes (potentially) No Terminal – Airside Aircraft Gates Number, aircraft size Reduce delays to arriving aircraft waiting for a gate to become available or rub-off delays to departing aircraft Seat Capacity at Gates Expand X* Walkway to Gates Provide moving walkway or people mover Reduce time getting to gate, but potentially offset increased wait at gate Moving Walkway or People mover to Gate Capacity, frequency, travel time Terminal - Landside Passenger Check-in Add positions Savings due to faster check-in Passenger Screening (TSA) Add lanes and other capacity enhancements Reduce passenger wait time Baggage Handling Improve inbound baggage facilities Save wait time to pick up baggage Baggage Claim Expand claim devices Save wait time to pick up baggage Food Court, Shops Expand or enhance X* * The perceived value of time spent in the gate hold area, food court or other terminal concession areas may be influenced by comfort and enjoyment levels, but the derivation of the values given in the Guidebook did not attempt to measure these effects. Step 2: Map Project to Time Categories The second step helps the analyst determine the time categories affected by the project. Time categories are portions (or segments) of an air passenger’s trip that have different values of time. This guidebook recognizes the ten time categories shown in Figure 2. Figure 2: Time Categories for Aviation Trips Ground Access and Terminal - Departure Ground Access Terminal Access Check-In/Security Reach Gate Area At Gate Airside Flight Time Flight Delay Terminal and Ground Egress - Arrival To Bag Claim Baggage Claim Ground Egress These categories are defined as follows: • Airport Ground Access Time: The travel time from a passenger’s origin to a parking location, drop-off location, or transit stop/station at the airport. Page 17

• Terminal Access Time: The time to reach the airport terminal from the parking location, drop-off location, or transit stop/station. • Check-in and Security Time: The time to check-in and check baggage, reach security screening, wait in line and clear security. Not all passengers check-in or check bags at the airport. There is no reason to believe that the disutility associated with waiting at check-in would be different from the disutility of waiting at security screening. • Time to Reach the Gate Area: The time spent getting to the gate area after clearing security. • Time at Gate: The time after reaching the gate area until boarding commences, including any time spent in airport concessions after clearing security. • Airport-to-Airport Flight Time: The total scheduled airport-to-airport travel time, time spent for taxi, takeoff and landing, in-flight time, and time spent at connecting airports. • Flight Delay Time: Any increase in the above airport-to-airport flight time. • Time to Baggage Claim or Terminal Exit: The time passengers spend walking from the arrival gate to the baggage claim area or terminal exit. The ACRP 03-19 research did not directly address the time to reach the baggage claim area or terminal exit, but it is reasonable to believe that the disutility associated with this is the same as that involved in getting to the gate area after clearing security on departure. • Wait Time at Baggage Claim and Exit Terminal: The time passengers spend in the baggage claim area waiting for baggage to arrive and subsequently exiting the terminal. The ACRP 03-19 research did not directly address wait time at baggage claim or exiting the terminal, but it is reasonable to believe that the disutility associated with waiting at baggage claim and then exiting the terminal is the same as that at check-in and security screening. • Airport Ground Egress Time: The travel time from the airport to the passenger’s destination. The ACRP 03-19 research did not directly address egress time, but it was assumed that the value of egress time is roughly equal to that for ground access. The analyst starts by identifying the relevant project elements from Step 1. These are then mapped to the ten time categories using the matrix shown in Table 5. A project may involve elements of multiple project types. In such a case, the analyst should note the time categories for each of the relevant project types. In addition, the analyst should take into account the caveats noted in the introduction. For example, a runway project may reduce delays and provide more direct flights. These lead to time savings at the gate or in-flight times, but the magnitude of the impact can vary Page 18

depending on the buffer times included in airline schedules and travel time margins allowed by passengers. Table 5: Time Impact Mapping Matrix Project Type (Elements) Ground Access Terminal Access Check-In and Security Reach Gate At Gate Flight Time Flight Delay To Bag Claim or Exit Baggage Claim Ground Egress AIRSIDE Air Traffic Control X X Runway X X X Taxiways X X Apron Area, Taxilanes and Aircraft Gate Positions X X GROUNDSIDE Access Road to Airport X X People Mover Access to Airport Terminal (from Transit, Rental Car or Parking Facilities) X X Parking Lot/Garage X Central Bus or Train Transfer Facility to/from Airport Terminal* X X Airport Circulation Improvments for Taxis X X Drop-off & Pickup Areas by Terminal Curbfront X TERMINAL LANDSIDE (DEPARTURES) Passenger Check-in X Passenger Screening (TSA) X People Mover to Gate X Aircraft Gates X X TERMINAL LANDSIDE (ARRIVALS) People Mover from Gate X Baggage Handling X International Arrival Facilities x** x x * If remote passenger processing available, it be treated in the same way as at the airport. **Queueing and processing in customs and immigration is assumed to have the same value of time savings as for security screening. Page 19

Step 3: Calculate Travel Delay The third step calculates the size of the travel delay with and without the project (or the net time savings). All time savings come from two fundamental drivers: (1) a reduction in the size of queues that form because of insufficient capacity at some part of the airport, or (2) a technological improvement that reduces the time required for travelers to complete some step in the process of getting through the airport to or from their flight or an improvement in the functioning of some transport element that allows people to move faster between key points. • Most commonly, an airport project is intended to enhance capacity to meet anticipated future demand. In many of these cases, there is acceptable queue delay today and the project is intended to insure that delays remain acceptable in the future.5 However, to show the time benefit, it is necessary to identify the potential source of queues being addressed. This makes it possible to compare anticipated future conditions with the project against a base case in which current capacity conditions prevail, so that the avoided delay can be calculated. • Time savings that do not involve capacity bottlenecks typically result from some improvement in the service frequency or speed of moving walkways, people movers, or other technologies involved in moving people from parking or other ground access facilities to the terminal, or from the check-in area to gates, or from gates to remotely parked aircraft, or from the application of some technology that saves travelers time, such as parking guidance systems that direct drivers to the closest available space. For projects that involve relief or avoidance of current or potential future capacity constraints, it is useful to consider that the effect of such constraints is more severe at peak times. In addition, a higher proportion of business passengers (who have higher values for their delay) are likely to travel at peak times.6 Not all airports have information about the trip purpose split (i.e., business versus leisure) of airport passengers, let alone how passenger characteristics vary between peak and off- peak times. Thus, this guidebook does not require that information to calculate the passenger travel time benefits. However, having these data available may result in higher benefit values, because using daily average passenger values to estimate travel time benefits underestimates the severity of passenger queues that occur at peak times, and using a composite value of time (that averages both business and leisure travel) will fail to account for the effect of greater business use during peak periods. 5 Even in situations with adequate capacity for existing traffic levels, some queues may form in peak periods. 6 Exceptions to this general rule may include airports that are located at major vacation destinations, including those serving major cruise ship ports. Page 20

If possible, the analyst should obtain relevant air passenger characteristics and passenger activity information along with capacity information to calculate total aggregate time savings for the classes of travel shown in Table 6. This information can be transformed into percentages that can then be applied as weighted averages to the values developed next in Step 4. Defining Peak Periods Peak periods will vary by airport. Some airports may have seasonal peaks (for example, ski season); others may peak for a particular week (spring beak). All airports will experience varying levels of passenger activity by time of day. For example, a major business hub may experience peak passenger activity during the first morning bank of arriving and departing flights, as business travelers are departing or arriving for meetings, or boarding connections to smaller cities. Similarly, a peak period may occur in the late afternoon and early evening as business travelers are departing for home and returning home. Because regional characteristics are rarely homogeneous, and local residents on personal travel contribute to peak passenger volumes, determining an airport’s peak period requires an analysis of hourly counts of arriving and departing passengers by day of the week to observe when peaks occur. Weekend days should be part of this analysis, to account for vacationers and other personal trips. This analysis is typically done for the peak month of the year, although it may be appropriate to consider other months as well if travel patterns vary widely during the year. In some cases the airlines serving an airport may be willing to provide passenger counts by flight or by hour of the day. Where these data are not readily available it is common practice to record arriving and departing aircraft seats by hour from the flight schedule and apply the average load factor for each airline and aircraft type for that month (which is available from the monthly data reported by each airline to the USDOT on Schedule T-100 and available on the Bureau of Transportation Statistics website or from a number of commercial sources). However, in performing this analysis there are a number of issues that need to be taken into account: • Peaks of arriving and departing passengers may occur at different times. • Aircraft passenger loads include connecting passengers, who may not be using some of the airport facilities (e.g., airline check-in or baggage claim). Therefore it may be necessary to adjust the aircraft passenger loads for connecting passengers, particularly at airline hubs. • Aircraft load factors (the percentage of seats occupied by passengers) are likely to be higher during peak periods than off-peak periods. However, there is no way to determine how much higher from the USDOT data, which only provide monthly totals of passengers and seats for each nonstop flight segment. Nonetheless, making adjustments to the reported load factors to give assumed peak and off-peak load factors on the basis of judgment is likely to give more accurate results than ignoring this aspect completely. A reasonable approach might be to assume that the Page 21

peak period load factor is mid-way between the average reported load factor and 100%, and adjust the off-peak period load factor accordingly to give the correct average load factor overall (this type of adjustment will depend on the proportion of seats in the peak period). • Departing passengers will arrive at the airport some time before their scheduled flight departure, typically an hour or more. Therefore the peak demand on many airport facilities, particularly those earlier in the departure process, such as terminal access services, airline check-in, and security screening, will typically occur in the hour before the hour with the greatest number of departing seats. On the other hand, it takes some time for all the passengers on a flight to deplane after the flight has arrived and proceed through the terminal to baggage claim and ground transportation, so the peak demand for facilities used by arriving passengers will typically occur in the half-hour after the half-hour with the greatest number of arriving seats. Thus performing the analysis on the basis of half-hourly rather than hourly periods is likely to give more accurate results. Table 6 shows a typical worksheet that can be used to classify passenger activity at an airport by trip purpose and peak/off-peak periods. Table 6: Breakdown of Passengers by Trip Purpose and Period Type of Traveler Peak Period Off-Peak Period Total Business Leisure Total If the Table 5 matrix (shown earlier) shows that a project may affect gate, flight time or flight delays, then it can also be useful to distinguish between two conditions: (a) delay under good weather conditions (referred to as Visual Flight Rules, or VFR), and (b) delay under cases of low visibility, inclement weather or other conditions requiring an increased spacing of takeoffs and landings (referred to as Instrument Flight Rules, or IFR). The delays will be greater under the latter conditions. For these, the probabilities of occurrence and the extent of delays can be calculated based on projected future traffic volumes from the FAA Terminal Area Forecasts, State or Regional Aviation System Plans, or Airport Master Plans and historic occurrence rates for conditions of diminished capacity.7 The analyst can calculate delay using the matrix shown in Table 7. 7 Generally these calculations are performed by a specialized consultant with expertise in airport capacity and delay analysis, and not by a benefit-cost analyst, although the benefit-cost analyst would need to know the right questions to ask and whether additional studies are needed. Recent ACRP and FAA reports listed in the references at the end of the Guidebook provide guidance on the subject. Page 22

Table 7: Breakdown of Airside Delay Weather Conditions and Time Periods VFR IFR* Percent Time Total Pass x Avg. Delay Percent Time Total Pass x Avg. Delay Peak Periods Off-Peak Periods TOTAL * At some airports it may be necessary to define multiple categories of IFR conditions, since delays may be significantly different under each IFR condition (e.g., delays may depend on the wind direction). To use Table 7 under these cases, the analyst should add the appropriate number of columns (“percent time” and “total passenger x average delay”) to the right of the single IFR column shown and identify each IFR condition analyzed. The result of Step 3 should be a total measure of the annual passenger delay for a future design year in a build (with project) and no build (without project) case. The general formula for calculating total delay in a future year is: Aggregate Annual Passenger Time Delay (in person-hours for each category of delay) = Number of Passengers (forecast to be flowing through the airport annually) * Average Delay per Passenger (for that corresponding category and year) The analyst should make this calculation separately for each of the ten time categories that are applicable to the project. If possible, this calculation should be repeated for peak and off-peak time periods. Step 4: Calculate Value of Delay The fourth step calculates the value of the delay estimated in the previous step. For each of the ten time categories, the analyst should select the corresponding unit value of delay shown earlier in Chapter 2, Table 1. Then, the value of the savings is calculated for each time category using the following general formula: Annual Value of Passenger Time (dollars, for each trip segment category) = Aggregate Annual Passenger Time Delay (person-hours, from the formula in Step 3) * Unit Valuation of Time per Hour of Passenger Time Savings (from Table 1) This calculation should be carried out for each applicable trip segment category, and the results then summed over all applicable trip categories, using the worksheet shown in Table 8. Page 23

Table 8: Worksheet for Calculating Value of Enhanced Travel Time (Reduced Delay) Time Category Total Annual Person-Hours of Time Saved x Value per Hour of Time ($2013) Base Case Project Case Difference Terminal landside (Departure) Ground access time Terminal access time Check-in and security time Time to reach gate area Gate time Airside (Flight) Flight time Unexpected flight delay Terminal Landside (Arrival) Time to reach bag claim or exit Baggage claim wait time and exit Ground egress time GRAND TOTAL There are several caveats and additional notes regarding the use of this calculation worksheet. First, it may also be observed that the worksheet does not provide a separate line for valuing time savings associated with transfers between connecting flights. If an airport capital investment project affects connecting passengers, the time savings of those passengers may be valued using the unit value of time for flight time. The ACRP 03-19 research did not specifically estimate a value of time for connecting passengers, but assumed that travelers valued time spent making connections the same as time spent in flight. To the extent that travelers consider the total travel time from their origin airport to their destination airport in making trade-offs between the travel time of different flights itinerary options (particularly between nonstop and connecting flights) and the associated airfares, it seems reasonable that they might value the time spent making connections much the same as the time spent on the aircraft. Second, it should be noted that the worksheet is set up for simplicity purposes without reference to trip purposes or peak/off-peak periods. However, it is recommended that the worksheet should actually be filled out separately for business and for non-business (leisure) travel, and the results then summed. This is preferred over use of the national composite because each airport actually has a unique mix of business and leisure travelers. Some airports are used almost entirely by leisure travelers while others have a much greater use by business travelers. Clearly, the national composite average is not appropriate for those situations. Therefore the analyst should use the composite trip purpose values only when no information on the business and leisure travel mix is available. Besides having information on the business/leisure mix of travelers, some airports may also have information to establish a different trip purpose mix for peak and off-peak travel (which may refer to either times of day or seasons of the year). All available information on customer trip purposes (business and leisure travel mix) and peak/off-peak differences Page 24

should be used, as appropriate, to calculate overall time values that are appropriate for the specific airport. Finally, it should be noted that information regarding the income levels of travelers should be used when available. While not all airports have survey information regarding the income mix of travelers, this information can enable a more sophisticated and accurate value of time savings associated with airport improvement projects. The appropriate unit valuation for time savings by income group was shown earlier in Chapter 2, Table 2. Step 5: Apply to Benefit-Cost Analysis The fifth step is to apply the information from Step 4 in a benefit-cost calculation. It is preferable to calculate the value of travel time savings directly for each future year given the non-linearity in delays as traffic levels increase, as well as likely changes in the values of time as: (1) the reduction of delays affects the overall time a passenger spends at the airport; (2) the relative change spent at the mix of time categories (see Table 5, the Time Impact Mapping Matrix); and (3) possibly the composition of the traffic. However, the value of total annual savings in passenger time calculated in the fourth step can also be shown for a single future year. To use this value in a benefit-cost analysis, it is necessary to transform the single value into a year-by-year time stream of values. This normally involves extrapolating for a number for years beyond the future year for which the value of travel time savings was derived, and interpolating the annual value of travel time savings for intermediate years until the year for which the annual value was derived. For the forecasting to be realistic, the analyst should consider that bottleneck delays start to occur with passenger volumes well below the runway, terminal, or gate facility capacity8, and increase progressively (and non-linearly) as passenger volumes approach or exceed (for short periods of time) the facility capacity. Thus a realistic projection of delay should account for these threshold effects.9 The software or spreadsheet typically used by the analyst for benefit-cost analysis will take the time stream of annual time saving values and calculate a net present value. It is assumed that all values are expressed in constant dollars (excluding the effect of inflation), 8 While, depending on the circumstances at each airport, this effect can start to become significant at a ratio of passenger activity to capacity of around 80%. 9 Bottlenecks for airport landside components may impede circulation in the airport and extend trip time, or affect the efficiency of other components. Airport circulation may affect terminal access time, check-in and security time, time to reach the gate, and time at baggage claim, as well as boarding time. These impacts, depending on where the bottleneck forms, will affect the total value of time expended and possibly more than one component of the airport terminal part of a trip. Page 25

and that the discount rate reflects the time value of money above inflation10. The results should go into a benefit-cost table as shown in Table 9. This example shows only the benefit-side of the table, but the cost-side should also be included. The results of this guidebook should all be reported in the row shaded and labeled as “passengers - time savings.” Depending on the nature of the time savings, there may be some additional diversion of trips from other airports. (See the text box below for a brief discussion of trip diversion.) Table 9: Typical Benefit-Cost Table for an Analysis Component Sum Across All Years Undiscounted Value Discounted Value Net Present Value Airline – Revenue Added* – Staff Time Cost Savings – Operating Expense Savings Airport – Revenue Added* – Staff Time Savings – Operating Expense Savings Passengers – Time Savings – Expense Savings Shifted Trips – Time Savings – Expense Savings Cargo Operators: Cost Savings Salvage Value of Asset * Changes in revenue are not typically included in benefit-cost analysis, since they are considered to be transfer payments, although they may be important for financial analysis. 10 The FAA mandates a seven percent discount rate in BCAs for Airport Improvement Program (AIP) discretionary grants greater than $10 million, and encourages sensitivity analysis with higher and lower discount rates (FAA Airport Benefit-Cost Analysis Guidance, Office of Aviation Policy & Plans, Federal Aviation Administration, December 15, 1999). There is much greater discretion available for airports and airport analysts in selecting discount rates for BCAs in capital decision making when not tied to the AIP grant process. Page 26

Assessing Trip Diversions and the Value of Time Savings Trip diversion refers to passengers selecting a different airport to use as a result of an airport capital investment project or operational changes. Such diversions may occur if a capital investment improves air service and passenger convenience at one airport compared to other airports in the vicinity. Diverted trips arise from new or extended runway projects that enable airports to offer longer distance non-stop flights. In such cases, passengers in the affected markets beginning their trips closer to the airport that has attracted additional air service may opt for the shorter ground trip. For example, a passenger may have to drive 50 minutes to take a non-stop transcontinental flight from the closest airport that offers this service. But if an airport situated 15 minutes from the passenger’s ground origin builds an 8,000 foot runway enabling transcontinental flights (and attracts an air carrier to provide this service) then the passenger may opt to change airports to save the 35 minutes (each way) in ground transportation. Stated preference or revealed preference techniques can be used along with an analysis of air passenger ground trip origins to estimate the volume of any diversion. The value of the time savings of diverted trips includes the difference between ground access time to the original airport and the time to the improved airport. In other cases, trips may be attracted from other airports if a capital investment project allows more frequent flights to the same destination (installation of additional gates may be an example of this). Surveys of business travelers in multi-airport regions typically show that larger airports are preferred by business travelers over the geographical convenience of smaller but closer airports, since they generally provide more alternatives should a flight be canceled. In this case, there may be a tradeoff between the dis-benefit of longer ground access times with the advantages of more flight options at the larger airport and reduced risk of long delays due to canceled flights. In such situations, fairly Page 27