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.
B-1 A p p e n d i x B To date, little attention has been devoted to how the various elements of DDFSs vary over time. In general, it has been assumed that DDFS elements will grow or decline in step with annual activity, sometimes with allowances for peak spreading. The research conducted to help develop this guidebook rigorously evaluated the various elements that comprise DDFSs and their variability and predictability over time independent of variations in annual activity. More specifically, it involved the analysis of a 10-year sample of historical OAG schedule and US DOT T-100 data, including a sample of five large-hub, five medium-hub, five small-hub and five non-hub airports, to assess the following DDFS elements: ⢠Stability of airport schedule profiles over time; ⢠Stability of the peak hour, both in terms of percentage of daily activity and timing; ⢠Stability of individual flight times to specific markets; ⢠Stability of the nonstop markets served; ⢠Stability of fleet mix; and ⢠Stability of load factors for each market and airline. There were several purposes to this research effort. First, to determine whether the current practice of using existing flight schedules as the foundation for future flight schedules is appro- priate. Second, to determine whether cloning (which implicitly assumes a continuation of exist- ing daily schedule profiles) is a reasonable alternative to manually identifying and adding new flights. The third purpose was to provide guidance on selecting new flight times to new or exist- ing markets. The fourth purpose was to provide guidance on selecting new nonstop markets, adjusting equipment types, and adjusting load factors. The final purpose was to provide a quan- titative basis for establishing confidence intervals around each element to assist in the evaluation of risk and uncertainty (see Appendix D). B.1 Stability of Schedule Profiles Over Time The analysis determined that, in general, the hourly pattern of arriving and departing opera- tions tended to be consistent at most airports. That is, the peaks and valleys in the daily profile of activity tended to occur around the same time across the 10-year period of analysis. This is not to say profiles are absolutely rigid. Variations in each hourâs share of design day scheduled operations or scheduled seats of five or 10 percent were not unusual. Other findings from the analysis included: ⢠Airport size is positively correlated with the stability of schedules (e.g., large hubs have more stable schedules than medium-hubs and so on). ⢠Total operation schedules are more stable than separate arrival and departure schedules. Stability and Predictability of Critical DDFS Factors
B-2 Guidebook for preparing and Using Airport design day Flight Schedules ⢠Total operation schedules are more stable than arriving/departing seat schedules; however, the gap in variability between scheduled operations and scheduled seats decreases as airport size decreases. This is because the type of aircraft and number of seats per aircraft become more uniform as airport size decreases. ⢠The stability of domestic operation schedules holds very close to that of total operations, while international operation schedules are more variable. ⢠Individual carrier schedules are less stable than total airport schedules. ⢠There is no discernible trend for carrier-specific schedule stability; the scale of individual car- rier operations at an airport is a more important driver than the specific carrier. For example, there is no evidence that, given an equal number of operations, a low cost carrierâs schedule is more stable than a legacy carrierâs schedule. As noted, international schedule profiles tend to be less stable than domestic schedule pro- files. Flights to specific international regions (e.g., Europe, Northeast Asia, and southern Latin America) tend to operate within specific schedule windows, but if the mix of flights to each region changes, the overall profile of international operations may change significantly. See Appendix N in ACRP Web Only Document (WOD) 14: Guidelines for Preparing Peak Period and Operational Profiles (technical report accompanying ACRP Report 82) for additional detail on international profiles. In addition, major structural changes to an airline schedule, such as an increase or decrease in the number of connecting banks, or a transition to a rolling hub, will cause changes to the schedule profiles over and above those suggested earlier. The research findings suggest that imposing top-down controls on the distribution of DDFS flight times may be too rigid and may generate results that fail to incorporate more subtle trends that can emerge from bottom-up DDFS development. A potential compromise may be to estab- lish soft controls to establish bounds from which DDFS-generated profiles could not deviate from without good cause. B.2 Stability of Peak Period Over Time The guidebook research analysis also examined the stability of the peak hour over time, both in terms of the hour of occurrence and the magnitude as a percentage of daily activity. The analysis determined the following: ⢠Large-hub airports tend to peak earlier in the morning than other airports, with the peak time occurring around 9:00 or 10:00 am. ⢠Medium-hub, small-hub, and non-hub airports all tend to peak later in the afternoon with the peak time tending to occur around 2:00 to 4:00 in the afternoon. ⢠With respect to the stability of the time that the peak hour occurs, there is no discernible dif- ference between airport size groups. ⢠The percentage of daily activity that the peak hour represents tends to be negatively correlated with airport size, indicating that larger airports typically have a more even distribution of activity across the time of day while smaller airports have higher peaks of activity. ⢠Larger airports tend to have more stability in the percentage of daily activity represented by the peak hour, relative to the airports of smaller size. The analysis indicated that although airports exhibit certain tendencies regarding the timing of the peak hour within the day, shifts in the timing of the peak of 2 or 3 hours can still occur. This has implications for air quality analysis, where average meteorological conditions can change depending on the time of day, and also for many terminal and landside facility requirements that are dependent on passenger O&D traffic. An early morning departing peak is likely to have more
Stability and predictability of Critical ddFS Factors B-3 originating passengers than a mid-morning peak with the same number of scheduled seat depar- tures. Likewise, a late evening arriving peak is likely to have more terminating passengers than an afternoon arriving peak with the same number of scheduled seat arrivals. Like the schedule profile analysis, the peak hour analysis indicated variations in the peak hour share of daily activity ranging from three to almost 20 percent, depending on the size of the air- port. Again, this suggests that imposing top-down controls on the distribution of DDFS flight times may be too rigid, and may generate results that fail to incorporate more subtle trends that can emerge from bottom-up DDFS development. B.3 Stability of Scheduled Flight Times to Individual Markets A key task in the preparation of a DDFS for future conditions is the adjustment of existing flight times or the estimation of new flight times. The guidebook research analysis examined how scheduled flight times to individual markets varied over the period of analysis. To make it manageable, the study was limited to markets where daily flight frequency did not change and focused on the first and last flights of the day. The analysis generated the following observations: ⢠The average schedule change between sequential months is significant, with changes of 10 to 15 minutes or more where frequencies stay constant. ⢠The standard deviation for scheduled flight time changes between periods is very high, sug- gesting that changes in flight times are variable and difficult to predict. ⢠The last flight of the day has a greater average change in scheduled flight time than the first flight of the day. ⢠Scheduled flight times for medium- and small-hub airports generally have greater inter-period change than for large-hub airports; however, non-hub airports have relatively simple operation profiles and their scheduled flight times have more stability as a result. Connecting banks have a significant effect on flight times at airline connecting hubs and at small airports where much of the service consists of flights to a connecting airport. The scheduled flight time analysis suggests that the first and last flights, and intermediate flights by inference, tend to remain within the same connecting bank. Within the bounds of that connecting bank, however, there is a tremendous amount of variability. This suggests that the selection of DDFS flight times at a connecting airport could be a two-step process, with the first step involving the selection of the connecting bank(s) and the second step randomly selecting times that fit within that bank. Appendix Q in ACRP Web Only Document 14 contains a complementary analysis that looked at the impact of changes in flight frequency upon scheduled flight times. The analysis found that, when the initial flight(s) were in the early morning or late evening, their schedule times were rela- tively unaffected by the addition of a new flight frequency. However, if the initial flight(s) were in the afternoon, the addition of a new frequency caused the initial flight time to shift by more than an hour almost 50 percent of the time. This suggests that midday flight times from an existing schedule are poor guides for a future DDFS in instances where flight frequencies to an existing market are expected to change. B.4 Stability of Nonstop Markets Depending on the ultimate use, market selection can be an important element of a DDFS. It affects directional headings for airspace analysis and can affect load factors and O&D connecting splits on individual flights. The guidebook research analysis looked at the stability in the number
B-4 Guidebook for preparing and Using Airport design day Flight Schedules of flights to individual markets and the number of nonstop markets served by airport category. The analysis determined: ⢠Large airports were more stable than medium or small airports in terms of the number of markets served. ⢠The frequency of flights to individual markets was also more stable at large airports than at small airports. ⢠The number of nonstop markets served appears to be increasing at large-hub airports but declining at all other airports. The 10-year study period encompassed a challenging time for the airline industry, during which the overall number of scheduled departures was dropping. Therefore, the guidance above may not apply to potential changes in nonstop markets during times of growth. B.5 Stability of Aircraft Equipment Types Equipment type can be an important consideration for DDFSs, as they affect aircraft delay, gate requirements and passenger loads. The guidebook research involved an analysis of the stabil- ity of general equipment types (widebody, narrowbody, regional jet, and turboprop) by airport category and determined that: ⢠The mix of equipment types is more stable at large airports than small airports. ⢠Equipment types that account for a large portion of an airportâs activity tend to be more stable (in percentage terms) than those that account for a smaller portion. ⢠Airlines at large-hub and medium-hub airports predominantly use narrow bodies, followed by regional jets. ⢠Airlines at small-hub airports predominantly operate regional jets, followed by narrow bodies. ⢠Airlines at non-hub airports generally fly regional jets, followed by turboprops. The 10-year period of analysis encompassed a time when many carriers were shifting service away from turboprops and small regional jets, resulting in elimination of service to small com- munities and shifting to larger aircraft with less frequency at other markets. This secular change in equipment use suggests that the variation in equipment type might be less in more stable times for the industry. B.6 Stability of Load Factors One of the key drivers of passenger loads is the enplaning or deplaning load factor. Apply- ing specific load factors to each market, as opposed to using an airport average, can generate greater precision but at the cost of additional effort. One of the guidebook research tasks was to determine how stable market load factors were over time, and whether the additional precision associated with assessing load factors by market could be confounded by variations over time. The general findings of the analysis were: ⢠There is a clear positive correlation between airport size and average load factor, with larger airports having higher load factors. ⢠There is a negative correlation between airport size and load factor variability, with larger airports having more stable load factors than small airports. ⢠Load factor stability is consistently higher when measured on a carrier basis than a market basis, regardless of airport size. ⢠Load factors have become more stable over time for all airport groups.
Stability and predictability of Critical ddFS Factors B-5 A likely reason for the difference in market and carrier load factor stability is that while market demand may have seasonality, resulting in lower load factors in certain months and higher over- all variability, carriers can compensate for this seasonality on a system-wide basis by reallocating their capacity to other markets. Load factors tend to be increasing at the same time that the standard deviation associated with load factors is decreasing. This is consistent with the ongoing industry trend in which airlines are increasingly matching capacity to demand, and service to markets with low load factors is being reduced. This suggests that differentiating load factors by market for DDFSs may become less of an issue in the future, as the differences between markets decline. It also suggests that if a DDFS preparer does differentiate load factor by market, they should consider increasing load factors at markets with low factors at a higher rate than at markets with high load factors. B.7 Summary The analysis of the stability of DDFS critical factors indicates that there is material variation in these factors over time, and that the degree of variation follows rational and predictable trends. This predictability can in turn be used to establish confidence intervals to help quantify the uncer- tainty associated with DDFSs. See Appendix D for more discussion of confidence intervals.
