Current situation
Delays are a very important issue at Boston’s Logan Airport, the fifth most delayed airport in the United States. Approximately 4.75% of total flights at Logan Airport are delayed and with total annual operations projected to increase over the next 15 years, it is important to address this issue today. Delay costs, which include operational and human costs, prove to be extensive and therefore a solution has to be found to ensure the survival of Logan airport.
The major causes of these delays include the adverse weather conditions as well as the fact that demand tends to be clustered at certain periods of the day. There are several approaches that can be taken to conquer this delay problem. The first option, which has been determined to be the best alternative, is to build another runway. This would ensure at least two operating runways at all times, regardless of weather conditions.
The second option proposed for dealing with the delay issue is that of peak period pricing (PPP), which entails increasing airline fees during periods of high demand in order to spread the demand for flights more evenly throughout the day.
Problem analysis
Weather is the foremost cause of delays in the case of Logan Airport. The airport’s operational capacity is significantly impacted by these adverse weather conditions (declining from approximately 120 operations per hour in normal conditions to approximately 80 operations per hour). In some cases, entire runways must be shut down depending on the extent of the weather adversity. Approximately 70-75% of flights at Logan are delayed due to weather issues.
In addition, Logan can be distinguished from other airports due to the fact that it is commonly used by smaller aircrafts (non-jet airplanes) carrying smaller amounts of passengers than commercial flights. Besides carrying fewer passengers, these aircrafts also fly more slowly, contributing further to the delay problem.
Demand for flights in Logan is projected to increase substantially between 2010 and 2015, and if it reaches the forecasted levels, delays will be inevitable as the airport is not prepared to function at full capacity. They already have delay problems, and with an increasing demand for flights, it will be nearly impossible for Logan to be able to cope. Peak Period Pricing appears to be a reasonable solution to this problem, as it would ensure that departing and arriving flights will be scheduled at a rate which the airport can handle without excessive delays.
Some secondary problems that must be taken into consideration include the high delay costs to the airlines and the personal costs for the individual passengers and the disruptions to the regional community.
Recommendations
Currently, only two out of the three Logan runways handle arrivals, at a rate of 30 incoming flights per hour each (60 in total). However, if the arrival rate reaches 59 per hour, extreme delays of over one hour (or 45 minutes, according to the FAA definition of delay) are to be faced by the arriving planes (see Appendix A). On the other hand, arrival rates of 50 and 55 planes per hour face only 6.5 and 12.5 minutes of delay respectively (neither of which are classified as delays under FAA regulations), which suggests that peak period pricing may be a viable solution to the problem if it can reduce arrival rates to approximately 50 planes per hour. However, when northwest winds are severe, only one of the runways at Logan remains in operation, and capacity drops to 20-30 arrivals per hour , in which case arriving planes will face extreme delays, regardless of whether PPP had been implemented or not.
The costs to the airlines, associated with delay, must also be taken into consideration. Assuming an arrival rate of 59 planes per hour, different types of aircraft face extra costs ranging from $495 to $3,490 depending on the type of aircraft (see Appendix B). Since PPP, while effective for peak periods when the weather is good, cannot influence when adverse weather conditions would occur, airlines would not only find themselves paying higher landing fees during peak periods, but also high delay costs whenever adverse weather results in the shutting down of one or more runways. This further supports our view that PPP will not be an effective solution to the delay problem at Logan.
If PPP is implemented, airlines face higher landing fees. Assuming a $200 landing fee during peak period is sufficient to reduce the arrival rate to 45 planes per hour, the delay time faced by planes when there are two functioning arrival runways is 4.5 minutes, which appears reasonable. However, if weather conditions permit only one runway to remain functional at Logan, this delay increases to over an hour. Building a new runway would ensure, as mentioned previously, that at least two runways would be in operation at all times.
In addition, smaller aircrafts using Logan may be unable to afford the PPP landing fee and refrain from using the airport altogether, thus hurting the profitability of the airport. For smaller aircrafts carrying fewer passengers, the suggested PPP landing fees represent substantial portions of their total revenues (see Appendix C), and implementing these fees may appear as discrimination in favor of larger, commercial aircrafts.
Conventional jets (150 seats) face the highest costs of all types of aircrafts. Currently, the mix of airplanes is approximately 42% conventional jets, 40% turboprop (15 seats), and 18% regional jets (50 seats). In the following calculations we have assumed that the costs associated with an hour of delay are $352 for turboprop planes, $672 for regional jets and $1,590 for conventional jets, in addition to $30.90 per passenger. We have also assumed that the annual number of operations is 479,000 of which 239,500 are arrivals. The yearly cost savings to airlines resulting from the installation of a third runway range from $40.81 million to $574.7 million, depending on the arrival rates of airplanes (see Appendix D).
If the plane mix is going to change (60% conventional jets, 30% regional jets and 10% turboprop planes) as is predicted to happen by 2015, the cost savings resulting from building a third runway will be even higher, ranging from $54.4 million and $766.5 million depending on the arrival rates. Note that the cost of building the new runway is $100 million.
By installing a new runway and ensuring that at least two arrival rates are operating at all times, we are essentially guaranteeing that delay times will not be any longer than they currently are under good weather conditions. In other words, when weather is adverse, the installation of a new runway will result in the same delay times as are currently experienced under good weather, essentially counteracting the current negative impact of adverse weather.
In addition, once Logan can guarantee that delay times will not exceed the worst-case scenario of 45 minutes even in adverse weather, the implementation of peak period pricing will be more effective because the airlines will be assured that they will get their money’s worth by paying a premium for arrival times during peak hours. If Logan were to refrain from building a new runway, PPP pricing may anger the airlines if adverse conditions occur, because they may experience over an hour of delay despite having paid the extra landing fee for arriving during the peak period.
Our final recommendation is the installation of a new runway in order to decrease delay costs for airlines, as well as to prepare for the increase in arrival rates projected for the next few years. The implementation of PPP, which we deem ineffective for solving the present delay problem at Logan, will always remain an option but is not critically necessary at this time.
Limitations
This analysis was conducted assuming that the additional runway, under each possible weather scenario, would have the standard service rate of 30 arriving planes per hour. This affects the cost calculations associated with the installation of the new runway, and if the new runway were to have an operating capacity of less than 30 arrivals per hour, the cost savings may be less than projected.
There is also the issue that the surrounding community has taken issue with the building of the new runway. There is currently a 1976 court injunction in place which will have to be lifted before the project can be undertaken. The assumption is being made that the FAA will fund the soundproofing of homes and that community groups and any legal action they may pursue will not have an impact on the undertaking of the project.
In addition, fundamental assumptions of queue theory are violated in this model. Firstly, in adverse weather conditions, not all of the runways operate at the same rate of planes per hour. In addition, the arrival rate of the planes is not distributed according to the Poisson distribution.

Appendix A

Appendix B

Appendix C

Appendix D