IEEE Computational Intelligence Magazine - May 2022 - 84

massive transportation demand, which
yields large-scale optimization problems.
Second, complex application scenes yield
NP-hard optimization problems, which
typically make mathematical methods
perform unsatisfactorily or even make
them inapplicable. Tuning the model of
the optimization problem to make it suitable
for mathematical methods will cause
an unacceptable modeling error. Third,
many transportation optimization problems
have complex features (e.g., dynamic
environment and multiple objectives) due
to real-world application demands, making
them more difficult to solve.
Evolutionary computation (EC) is a
kind of computational intelligence and
artificial intelligence technique that imitates
intelligent behaviors or phenomena
in nature and human society. A salient
feature of EC algorithms (approaches) is
that they do not require a delicate mathematical
model (e.g., linear programming
model) and can obtain optimal or
near-optimal solutions in an acceptable
time. With a powerful search capacity,
great robustness, and potential parallelism,
EC has been widely used to solve
various kinds of complex optimization
problems [7]-[9].
In recent years, EC algor ithms
have also been extensively used to solve
transportation optimization problems,
helping to achieve intelligent transportation
in smart cities. However, to the best
of our knowledge, there are still few literatures
that comprehensively survey the
conjunction of EC algorithms and intelligent
transportation in smart cities. To
fill this gap, a two-layer taxonomy is
introduced to classify related studies, as
shown in Fig. 1. In the first layer, the
related studies are classified into three
categories based on the application
scene of the optimization problem.
These three categories are " For Land
Transportation, " " For Air Transportation, "
and " For Sea Transportation. " The
construction of smart cities is oriented
to different types of cities. Land transportation
is the most basic transportation
mode in cities. Air transportation is
essential to cities with airline hubs, such
as Los Angeles and Beijing. Sea transportation
is essential to coastal and port
cities, such as Shanghai. Thus, it is necessary
to review the related studies on
all three transportation modes. Then, in
the second layer, related studies are further
classified from the perspectives of
government, business, and citizens. The
government perspective focuses on
public infrastructure and municipal
management, the business perspective
focuses on the profit and service quality
of transportation companies, and the
Government Perspective
Vehicular Network
Traffic Light
Electric
Vehicle Infrastructure
For Land Transportation
Business Perspective
Metro Transportation
Route Scheduling
Citizen Perspective
(Taxi, Carpool, Parking)
Trip Recommendation
Car Service
Business Perspective
Two-Layer Taxonomy of
EC for Intelligent
Transportation
in Smart Cities
For Air Transportation
Citizen Perspective
Landing Scheduling
Airline Management
Route Design
For Sea Transportation
Business Perspective
Berth Allocation
Path Planning
Container Stowage
FIGURE 1 Two-layer taxonomy of EC for intelligent transportation in smart cities.
84 IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE | MAY 2022
IEEE Computational Intelligence Magazine - May 2022

Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - May 2022
