Research Article

A Decision-Making Model Using Machine Learning for Improving Dispatching Efficiency in Chengdu Shuangliu Airport

Table 1

Explanation for notations.

NotationDefinition

Factor of number of passengers in the parking lot
Passenger’s waiting time
Driver’s waiting time in the parking lot
Difference between the time taking the last taxi to pick up a passenger and the time taking the next taxi to pick up a passenger in the parking lot
Boarding time of passengers
Number of passengers who are waiting for taxis
Airport position coefficient 1
Airport position coefficient 2
Waiting time coefficient
Airport size eigenvalue
Number of passengers in the parking lot at some point
Number of taxis in the parking lot at some point
Airport position coefficient 3
Airport terminal size coefficient
Airport position coefficient 4
Driver’s waiting cost
Expected revenue per hour while the driver waits
Driver’s waiting revenue in the parking lot
Distance a driver picking up passengers from airport to city
Mileage of taxi starting price
The fare per kilometre beyond the starting mileage
Taxi starting price
Fuel cost per kilometre of taxi
Revenue for drivers returning directly to the city to pick up passengers
Distance of picking up passengers in the city
Successful times of taxi drivers in picking up passengers per hour in the city
Priority of a taxi
Passenger boarding time in line i
Waiting time of the j th taxi in line i at some point
Priority of the first taxi in line i
Probability of choosing the first taxi in line i
Relative weight for line i
Standardized relative weights for nonempty line i