DYNAMIC STRATEGIC PLANNING IN PRACTICE: PEASE INTERNATIONAL TRADEPORT. Joakim Karlsson. Daniel Webster College. Nashua, NH

Transcription

1 DYNAMIC STRATEGIC PLANNING IN PRACTICE: PEASE INTERNATIONAL TRADEPORT Joakim Karlsson Daniel Webster College Nashua, NH

2 Karlsson 1 ABSTRACT Increasingly there has been a recognized need for flexibility in long term airport planning. Past rigid reliance on forecasts has resulted in costly mistakes. These errors have taken two forms: construction of facilities that remain underutilized for years and/or inadequate preparations for rapidly changing trends. Either can be expensive, as limited capital funds are tied up in unused facilities; operations costs are accrued on empty buildings; or, conversely, the airport operates in crisis mode, rushing construction projects to catch up with changing markets. Airport planners must accept that they cannot adequately predict the future. One answer is dynamic strategic planning a form of planning that accepts a wide range of possible future outcomes. This approach has two dimensions: more flexible use of forecasts and construction of facilities that give the airport operator options to handle uncertainties. Despite the increased call for dynamic strategic planning, it is not always clear how to use it in practice. The airport at Pease International Tradeport serves as a case study to demonstrate several aspects of this approach. These include a shift from traditional forecasts to planning thresholds; modifications to industry design standards to take into account unique local circumstances; and the incorporation of real options in the passenger terminal design, including an international/domestic swing gate, non-load bearing walls for easy expansion, and the use of the aircraft apron for overflow parking. The case study also discusses areas where improvement is needed, such as the lack of decision analysis to justify major capital expenses.

3 Karlsson 2 INTRODUCTION The latest call for dynamic strategic planning can be found in Professors Richard de Neufville and Amedeo Odoni s book, Airport Systems Planning, Design, and Management. The authors make a compelling case for this mode of thinking and illustrate their concepts with both historical mistakes and successful implementations (de Neufville & Odoni 2003, 59-88). In summary, dynamic strategic planning recognizes a priori that forecasting is an inherently uncertain process; develops flexible plans that can cope with high levels of uncertainty and multiple future outcomes; and uses analytical tools to measure which decisions to implement and when. Dynamic strategic planning is suitable for all airports. Even facilities that are running near capacity, and therefore have limited growth potential, are subject to the uncertainties of a deregulated airline industry. Some airports, however, face particularly high levels of uncertainty. These include newly constructed commercial airports; existing commercial airports with low levels of airline activity; and closed military bases which are being reused as commercial airports (Karlsson 2002, 2). For these airports, flexible planning is a must. They generally have little advance knowledge of the characteristics of future demand or when demand will materialize. These types of airports are the focus of this paper. When planning fails to adequately account for high levels of uncertainty, the consequences can be costly. There are two basic failure modes involved:

4 Karlsson 3 Overbuild : This occurs when an airport operator constructs facilities in anticipation of rapid growth, which then does not materialize. Underbuild : This occurs when an airport operator does not have adequate infrastructure in place for unforeseen rapid growth or changes in demand. The costs associated with the overbuild scenario are easily recognized: Unused or underutilized facilities absorb capital funds, which could have been used more effectively elsewhere. They also incur operational costs, which may not be recoverable because revenues are depressed. Even mothballed facilities incur some minimal level of operational costs. Overbuild examples include Newark Terminal C and Montréal-Mirabel (de Neufville & Odoni 2003, 69, 143). Mirabel was originally intended as the hub for Montréal s international flights. Its role has since changed to reflect actual market realities. It now targets cargo and charter flights for vacationers (Aéroports de Montréal n.d.), but remains underutilized. The underbuild scenario may be viewed by some as a windfall of passengers. However, even here there are costs. Unexpected demand results in inconveniences for passengers, inability to accommodate new services, as well as a rush to construct. The latter increases the chance of planning errors, premium construction costs, and public relations challenges. Underbuild cases are often associated with airlines opening new hubs or low-cost carriers starting new service. Examples include Delta Airlines establishing a hub at Cincinnati and Southwest Airlines moving into Baltimore-Washington International (de Neufville & Odoni 2003, 79-80). An example close to our case study is New Hampshire s Manchester Airport, which reached record traffic in 2002, despite major declines elsewhere in the industry. Manchester s attempt to keep up with this rapid growth includes construction of a 6,500 m 2 terminal addition, a 4,800 space parking garage, an

5 Karlsson 4 elevated pedestrian walkway, new food and retail concessions, new car rental concessions, a new air traffic control tower, and extensions of both runways (Manchester Airport Authority 2003). Some airports go through both overbuild and underbuild phases. Washington Dulles International struggled to build up significant traffic for almost 20 years (de Neufville & Odoni 2003, 69). By the end of the 20th century however, it had become the twelfth most delayed airport in the U.S., averaging 5.6 minutes of delay per operation (Federal Aviation Administration 2000, 39). While the need to cope with high levels of uncertainty is becoming increasingly clear, the airport planning community still relies heavily on traditional methods. In the U.S., these are promulgated by the Federal Aviation Administration in its Advisory Circular Airport Master Plans (FAA 1985). While theoretically advisory in nature, conditions attached to federal airport grants make its use a mandatory for all practical purposes (FAA 1999). Like most standard airport planning methods, the FAA approach begins with an airport inventory and the development of a forecast. The actual planning activities follow and are tied directly to the forecast results. It is this reliance on specific forecast values that makes most airport plans incapable of dealing with high levels of uncertainty. High uncertainty means that the airport faces many different future outcomes. These outcomes preclude the airport from relying on a single comprehensive solution. Instead, a tool box of methods is needed that allows the airport to remain flexible. The airport need not accommodate all possible scenarios, but only those outcomes that have the greatest probability of occurring. The outcomes may differ significantly and along several dimensions: Both the overall magnitude of activity and its characteristics (e.g., the share of domestic vs. international

6 Karlsson 5 passengers) may vary. The specific tools discussed here include the use of planning thresholds, modifying industry accepted design standards, and incorporating real options. BACKGROUND The closure of military aviation facilities often raises the possibility of civil aviation reuse as a commercial passenger airport or air cargo facility (Hoyle, Tanner & Associates 1993). The facility s market catchment and proximity to existing commercial airports determine whether or not this is feasible. Even when this appears to be the case, such facilities often face high levels of uncertainty, for a number of reasons: Aviation tenants are reluctant to move to new facilities The facility often faces competition from nearby airports The exact level of activity is extraordinarily difficult to predict when it stems entirely from the introduction of new service Our case study focuses on the aviation reuse of Pease Air Force Base in Portsmouth, NH. 1 It features all of the characteristics listed above. A short history of the facility is in order (Pease Development Authority n.d.; HTA 1995; HTA 2001; Mark Rowell, personal communication April 2003): The airport began as the civilian 300-acre Portsmouth municipal airport, built in the 1930s. During the Cold War, the need for strategic bomber bases in northeastern U.S. emerged, and Portsmouth Air Force Base (AFB) opened in The facility was renamed Pease AFB the following year, in honor of World War II Captain Harl Pease, Jr.

7 Karlsson 6 With the end of the Cold War, the U.S. Congress initiated several rounds of Base Realignment and Closures (BRAC). Pease was the first facility to close as part of the 1988 BRAC. As such, it has often found itself in a leading role in the base reuse process. The New Hampshire State government acted quickly to try to minimize the economic losses associated with military base closings. In April 1989 it established the interim Pease Redevelopment Commission. The PRC decided as early as August 1989 that the facility would be redeveloped as a joint civilian/new Hampshire Air National Guard airport. On June 1, 1990, the Pease Development Authority (PDA) was created. A joint use agreement between the PDA and U.S. Air Force was executed July 1991, allowing civil aviation operations to take place. In December 1991, the airport s first civilian airport layout plan was approved. A public benefit transfer from the U.S. Air Force to the PDA occurred on April 14, 1992, allowing the PDA to operate the facility under a long term lease. 2 During this time period, the facility was also given its current name Pease International Tradeport. In 1993, Business Express and Atlantic Coast Airways (dba United Express) began passenger service, using a renovated Air Force building as a temporary terminal. Both carriers stopped their Pease flights within two years however, citing lack of demand. In 1998, the PDA completed construction of an international passenger terminal. In 1999 Pan American Airways began Boeing 727 jet service. Boston-Maine Airways, added regional turboprop service in 2002, doing business as Pan Am. Other aviation activities include general aviation, cargo, and military aviation. 1 The facility actually straddles two communities: The Town of Newington and the larger City of Portsmouth. 2 Transfer of title from the Air Force to the PDA cannot occur until environmental clean-up has been completed.

8 Karlsson 7 The airport is well situated, with reasonable travel distances and good highway connections to Boston, Portland, ME, and Manchester, NH. Approximately two million people live within an hour s drive to the airport. In terms of its infrastructure, the airport offers a 3,450 m runway, 60 ha of apron, an international passenger terminal, a 4,200 m 2 cargo terminal, and an 800 ha business park (PDA 2003). By many measures, the redevelopment has been a success. The Tradeport features more than 160 businesses, over 350,000 m 2 of new or renovated space, and approximately 5,000 jobs created (PDA 2002). Yet, as Figure 1 shows, civilian aviation activity at Pease has been volatile. Initial growth in passengers was encouraging, fueled by low introductory fares. However, as fares were raised to above break-even levels, demand dropped. By 1996, both Business Express and United Express had dropped schedule service from Pease. Significant levels of passenger service did not resume until , with the introduction of Pan Am as a carrier. Since a high in 2000, the number of enplanements has been dropping again. The number of aircraft operations reflects a similar pattern, although since it also includes cargo, general aviation, and military activity, it has generally exhibited less volatile behavior. This high level of volatility in aviation activity is indicative of how uncertain the future is at Pease. The airport has substantial excess capacity at this point: If the airport s marketing efforts result in additional carriers starting service, passenger activity could increase quickly. Conversely, if Pan Am falters, passenger activity could again drop. It is this reality, which has persisted since the closure of Pease AFB that has forced the PDA into adopting dynamic strategic airport planning techniques.

9 Karlsson 8 60,000 50,000 40,000 30,000 20,000 10, Enplanements Operations Figure 1: Pease aviation activity total annual enplanements and operations DISCUSSION This case study discusses three specific examples of dynamic strategic planning at Pease. They are: the use of planning thresholds, modifications to terminal planning guidelines, and the use of real options. The need to include decision analysis and options analysis is also covered. Use of Planning Thresholds The traditional method of airport planning calls for developing aviation activity forecasts, and then evaluating alternatives that provide the capacity to meet the projected demand. Construction is typically scheduled to occur a few years prior to capacity being reached, as predicted by the forecast. In other words, the construction schedule is directly influenced by the

10 Karlsson 9 results of the forecasting process. On the other hand, dynamic strategic planning recognizes that forecasts are never accurate. This is particularly applicable to high uncertainty scenarios, such as Pease. This poses a challenge to airport decision makers: If reliable forecasts cannot be developed, how does one plan ahead for construction? The solution is to change the question from what year to begin construction to what activity level warrants construction. At Pease, this was implemented through the use of planning thresholds. These are activity numbers that, when reached, trigger expansion. In the case of passenger terminal planning, these activity numbers are expressed as peak hour passengers, annual enplanements, and the peak number of gate positions used by aircraft. This method removes the pressure from trying to determine which year to begin expansion, and instead focuses airport development on specific goals measured in terms of aviation activity. At Pease, expansion of the international passenger terminal is planned in two phases. Planning thresholds were developed as triggers for each phase. Because the terminal has the ability to serve both domestic and international passengers, space needs for both were considered. For the Phase 1 expansion, planning thresholds were set to correspond to the activity levels representing 80% of the terminal s existing capacity (HTA 2001, 7). For the Phase 2 expansion, the planning thresholds were increased to account for added capacity planned in Phase 1. Various average aircraft sizes were considered, to give airport decision makers maximum flexibility in accommodating a broad range of future scenarios. Table 1 illustrates the results from this analysis. The example lists results for the domestic-only scenario for the Phase 1 expansion (HTA 2001, 10).

11 Karlsson 10 Table 1: Phase 1 passenger terminal planning thresholds (domestic) Item Capacity Planning threshold Annual enplanements 54,259 43,407 Annual air carrier operations 150 seat aircraft 1, seat aircraft seat aircraft Peak hour aircraft 150 seat aircraft seat aircraft seat aircraft Based on this analysis, specific recommendations for action could be prepared for the airport management staff. An example is provided below, for the Phase 2 expansion (HTA 2001, 12): Expansion strategy for Phase 2: 2a. Add approximately 4,650 m 2 expansion, one passenger boarding bridge, a new baggage carousel, and new curbside elements (including check-in counters) when the peak hour passengers reach 667 and annual enplanements reach approximately 156,000. 2b. Add an additional passenger boarding bridge (for a total of six) when the number of aircraft that have to be served simultaneously on a regular basis increases from five to six. All too often, airport planning documents provide technical results without providing clear and specific guidance to the airport decision makers. An effective airport plan accomplishes the following (de Neufville & Odoni 2003, 60): It defines organizational relationships that will allow management to develop the right facilities according to future requirements as they emerge. Most important, it provides managers with the support they need to program the implementation of overall strategic decision.

12 Karlsson 11 This is a fundamental principle of dynamic strategic planning, and the sample expansion strategy shown above is a specific example of how this can be implemented. Modifications to Terminal Planning Guidelines A necessary extension of dynamic strategic planning is that planners must be willing to modify industry standard guidelines in support of flexible planning. As an example, modified space planning factors were used to plan the Pease passenger terminal expansion. Space planning factors are used for computation of space requirements for various airport facilities, in this case the terminal building. They are usually employed in the preliminary planning phases, in order to estimate order-of-magnitude ranges for the overall size of a terminal facility prior to performing more detailed calculations (Horonjeff & McKelvey 1994, 444). These rules-of-thumb are usually expressed as unit space factors, for example, 0.01 m 2 of gross terminal building space per annual enplanement. Various organizations publish space planning factors, including the FAA (1988) and the International Air Transport Association (1989). The specific challenge in applying standard space planning factors to the passenger terminal at Pease is their inherent assumption about peaking characteristics. They assume that the airport in question is relatively mature in its development, and therefore consistently busy throughout the day. Hourly peaks are assumed to exist, but to be moderate. Mathematically speaking, the FAA and IATA space planning factors include an underlying assumption that the peak hour ratio (i.e., the ratio between activity in the peak hour and the average hour) is close to unity. The solution to this challenge was to examine the peak hour ratio at an airport similar to Pease, but at a more developed stage, namely Manchester Airport. A mathematical model was

13 Karlsson 12 developed to approximate how the peak hour ratio might change as the airport matures (Karlsson 2002). Figure 2 is an example of a typical relationship between an airport s peak hour ratio and its future level of development, as expressed by future annual enplanements (Karlsson 2002, 51) Peak Hour Ratio ,000 1,000,000 1,500,000 2,000,000 Annual Enplanements Figure 2: Peak hour ratio vs. annual enplanements Once a model had been developed to explain this relationship, it was used to adjust the space planning factors. This in turn allowed for the presentation of a preliminary terminal expansion plan that took into account the unique circumstances at Pease. The results were presented to the PDA in the form of a technical supplement to the airport s master plan (HTA 2001, 39-41). This analysis was also used to support the development of the planning thresholds described earlier.

14 Karlsson 13 Use of Real Options A real option is defined by physical facilities or devices that enable the system operator to react to changes (de Neufville & Odoni 2003, 813). It is a physical aspect of the airport infrastructure that has been designed to accommodate a range of future outcomes. Typically these are facilities that allow for multiple or shared uses. They also include uses that are temporary in nature and can be converted into other, long-term uses in a cost effective way. The passenger terminal at Pease features three examples of real options (The H.L. Turner Group Inc. 1997; HTA 1995): A domestic/international swing gate Use of the aircraft parking apron as overflow car parking Non load-bearing walls that allow for easy expansion of the terminal A domestic/international swing gate can be configured to accommodate either domestic or international flights. In domestic operations, arriving and departing passengers are allowed to commingle. In international operations, arriving passengers must be separated from all other passengers until they have passed through the federal inspection services (customs, immigration, and agriculture inspection). As shown in Figure 3, this is achieved through the use of a moveable wall and a system of locking doors to channel passenger flow. This is similar to the system in place at Edmonton International Airport (de Neufville & Odoni 2003, ), except that vertical separation is not used at Pease.

15 Karlsson 14 Figure 3: Real options in the Pease passenger terminal The use of the aircraft parking apron as overflow car parking was the result of higher than expected parking occupancy during the airport s initial years of airline service. At the time, a renovated military aviation building served as the airport s terminal. In order to avoid building new parking facilities prior to finalizing the airport s future terminal configuration, part of the terminal aircraft parking apron was converted to overflow parking. This was achieved by erecting security fencing to create a secure area on the apron. Converting aircraft parking to car parking is generally not desirable, since it limits future expansion. In this case, excess aircraft parking capacity was borrowed to fill current car parking needs. This overflow car parking was designed so that it could easily be returned to its original use, when needed. Although the decision to use the aircraft parking apron was reactive

16 Karlsson 15 rather than proactive, it is included here as an example of a real option that other airports may consider. In the planning, design, and construction of the international passenger terminal at Pease, non-load bearing walls were used to allow for cost effective future expansion (see Figure 3). This represents a real option, because the airport user purchased a choice that accommodates different future outcomes. The use of non-load bearing walls in this way is a relatively common practice in airport passenger terminals, and may be the most frequently used real option in airport design. Need for Decision and Options Analysis As described, Pease has used a number of techniques to maintain flexibility in its airport planning. One of the examples listed here, the use of the aircraft parking apron as interim overflow parking, was a purely reactive decision. However, the others were planned and serve as good examples for how to practice dynamic strategic planning. The dynamic strategic planning methods in place at Pease today were introduced based on the best professional judgment of the planners at the time. Some, such as the decision to build a new international passenger terminal, were speculative in nature. In fact, at the time the terminal was built, there was no air carrier activity at the airport at all. The move to go ahead with the passenger terminal was an inherently risky decision. Pease could reduce the risk in its planning process by adopting quantitative techniques to evaluate its decision making. Two specific approaches to consider are decision analysis and options analysis. In decision analysis, the expected value of a specific decision is computed. This takes into account the economic value of each possible outcome associated with that decision, as

17 Karlsson 16 well as the probability of each outcome occurring (de Neufville & Odoni 2003, ). The decision with the highest expected value is the optimal choice. Options analysis quantifies the benefit of the flexibility provided by either a financial option or a real option (de Neufville & Odoni 2003, 813). Financial options are contractually defined by calls and puts. For example, these may be used by airlines to hedge fuel prices. Real options, as discussed earlier, represent physical choices in the airport infrastructure that buy flexibility to deal with uncertainty. Both financial and real options have a price: For example, the real option inherent in the domestic/international swing gate at Pease has a capital cost associated with its construction, as well as an operational cost. 3 Options analysis is an analytical method, usually implemented in computer software, to determine whether the cost of the real option outweighs the value of its benefit. CONCLUSION Airport planning remains mostly traditional and rigid in its approach. Over-reliance on forecasts has resulted in costly mistakes. This is true even at airports that are well established and have clearly defined trends in their historical aviation activity. At airports such as converted military aviation facilities, the level of uncertainty is higher, making traditional planning methods ineffective. Dynamic strategic planning methods can be particularly valuable at these types of airports. Even when the need for dynamic strategic planning is evident, it can be difficult for airport decision makers to know how to apply it in practice. This case study has highlighted specific examples from Pease International Tradeport, including the use of planning thresholds to avoid

18 Karlsson 17 reliance on traditional forecasting; the modification of standard space planning factors to take into account unique circumstances at the airport; and the use of real options to buy flexibility in the airport s infrastructure. What these methods have in common is that they acknowledge upfront that the future is inherently uncertain and allow the airport to deal with a wide variety of possible outcomes. However, it must also be recognized that there are costs associated with decisions based on dynamic strategic planning. Tools such as decisions analysis and options analysis should be used to ensure that the benefits exceed the costs. 3 As a side note, the manufacturers instructions require that the door be operated by two workers for safety reasons.

19 Karlsson 18 REFERENCES Aéroports de Montréal. N.D. Corporate Information Our History. Internet. Available from accessed 11 May De Neufville, Richard, and Amedeo R. Odoni Airport Systems Planning, Design, and Management. New York, NY: McGraw-Hill, Inc. Federal Aviation Administration Airport Master Plans. Advisory Circular 150/5070-6A. Washington, D.C.: U.S. Government Printing Office Federal Aviation Administration Airport Master Record. FAA Form Internet. Available from accessed 14 May Federal Aviation Administration Assurances: Airport Sponsors. Internet. Available from accessed 15 May Federal Aviation Administration Aviation Capacity Enhancement Plan. Internet. Available from accessed 14 May Federal Aviation Administration Planning and Design Guidelines for Airport Terminal Facilities (FAA Advisory Circular No. 150/ CHG 1). Washington, DC: U.S. Government Printing Office. The H. L. Turner Group Inc Program/Concept: Customs, Immigration & Air Terminal Building. Concord, NH: The H. L. Turner Group Inc.

20 Karlsson 19 Horonjeff, Robert, and Francis X. McKelvey Planning and design of airports. New York, NY: McGraw-Hill, Inc. Hoyle, Tanner & Associates, Inc Pease International Tradeport Airport Master Plan Supplement. Manchester, NH: Hoyle, Tanner & Associates, Inc. Hoyle, Tanner & Associates, Inc Pease International Tradeport: Airport Master Plan Technical Report. Bedford, NH: Hoyle, Tanner & Associates, Inc. Hoyle, Tanner & Associates, Inc Regional Aviation Impacts and Aviation Facility Reuse of Loring AFB. Bedford, NH: Hoyle, Tanner & Associates, Inc. International Air Transport Association Airport Terminals Reference Manual. Montreal, Canada: International Air Transport Association. Karlsson, Joakim Preliminary Space Planning for Passenger Terminals at Developing Airports. International Journal of Applied Aviation Studies 2: Manchester Airport Authority Manchester Airport News. Internet. Available from accessed 14 May Pease Development Authority Airport Facts. Internet. Available from accessed 15 May Pease Development Authority. N.D. History. Internet. Available from accessed 17 March Pease Development Authority (April 2002). News and Events. Internet. Available from events.htm; accessed 15 May 2003.