A new algorithm from Binghamton University considers the passenger’s perspective when scheduling flights. The goal is to create the most efficient travel itinerary for the airport, the airline and the passenger.

The new algorithms aim to lower the passenger's cost of delayed flights.

"Airports mainly focus on increasing the runway throughput and, in general, most other papers focus on delay as a time, but they don't consider the actual cost of the delay," said Duaa Serhan, a graduate student in industrial and systems engineering who developed the models for the paper.

Traditionally, the effect on the passenger is overlooked when scheduling flights, but there personal costs are valuable. Instead, schedules focus on the airport, which is owned by the government and regulated by the FAA, and the airlines.

"From the airport's perspective, the runway throughput is the most important, so they want to quickly send out as many aircraft as possible," said Associate Professor of Systems Science and Industrial Engineering Sang Won Yoon, "From the airline company's perspective, they just want to minimize the delay."

The result: a schedule based on maximizing aircraft throughput and plane turnover, as elected by a decision support system (DSS). The DSS prevents delays and uses time and distance parameters to find the optimal schedule.

The Binghamton team created two algorithms that also include the passenger in this decision making. Passenger delay costs include missed connections, rebooking costs and compensation for missed flights.

"Time is a cost, but airline companies also have to reroute passengers. If they miss the flight, then they have to schedule a new flight or provide a voucher. There are also immeasurable costs, like missing a call with your family, or a business meeting," said Yoon.

The human costs can be immeasurable for airlines, which is why the passenger’s perspective is often ignored. The Binghamton team wanted to bring the passenger’s needs into the equation.

The new models integrated common passenger delay costs: passenger delays, passenger missed connections and additional operating costs. The models were tested under various air traffic conditions and scenarios and compared to three other common scheduling methods.

The model which put the passenger’s perspective in the forefront really stood out. This model had the lowest airline and passenger delay costs, reducing them 6.4% more than other models. It also reduced the number of flight delays by 8.6% to 65.4%, depending on the given situation.

The next step for the team is to take into account how the weather affects delays.

"What we are investigating now is changing air terminal configurations based on the weather uncertainty. We are working on routing and scheduling aircraft when there are uncertain weather conditions, and trying to minimize delays," said Serhan.

The paper on the new models was published in the Journal of Air Transport Management.