Recent advances in traffic optimisation: systematic literature review of modern models, methods and algorithms

Abstract

Over the past few decades, the increasing number of vehicles and imperfect road traffic management have been sources of congestion in cities and reasons for deteriorating health of its inhabitants. With the help of computer simulations, transport engineers optimise and improve the capacity of city streets. However, with an enormous number of possible simulation types, it is difficult to grasp valuable, innovative solutions which are of the greatest value to city citizens. In this work, the authors expose various problems within this area having reviewed and analysed over 130 papers selected out of 1200 works in the field of urban simulations. The study describes the selection process of important papers and highlights characteristics of microsimulations, macrosimulations, computation optimisations and other approaches found in the literature which are especially useful and should be further built on in the future. They present and compare results provided in reviewed works in terms of throughput improvement, queue, waiting and travel time reduction, vehicle speed increase, speed‐ups as well as assumed simulation parameters. Finally, they focus on research gaps, such as a small number of works considering crisis simulations, few real‐world scale simulations as well as on software architectural changes and low‐level optimizations.

Full text

IET Intelligent Transport Systems
Review Article
Recent advances in traffic optimisation:
systematic literature review of modern
models, methods and algorithms
ISSN 1751-956X
Received on 24th May 2020
Revised 20th October 2020
Accepted on 30th November 2020
E-First on 2nd February 2021
doi: 10.1049/iet-its.2020.0328
www.ietdl.org
Rydzewski Aleksander1 , Czarnul Paweł1
1Department of Computer Architecture, Gdańsk University of Technology, Gabriela Narutowicza Street, Gdansk, Poland
E-mail: Aleksander.Rydzewski@pg.edu.pl
Abstract: Over the past few decades, the increasing number of vehicles and imperfect road traffic management have been
sources of congestion in cities and reasons for deteriorating health of its inhabitants. With the help of computer simulations,
transport engineers optimise and improve the capacity of city streets. However, with an enormous number of possible simulation
types, it is difficult to grasp valuable, innovative solutions which are of the greatest value to city citizens. In this work, the authors
expose various problems within this area having reviewed and analysed over 130 papers selected out of 1200 works in the field
of urban simulations. The study describes the selection process of important papers and highlights characteristics of
microsimulations, macrosimulations, computation optimisations and other approaches found in the literature which are
especially useful and should be further built on in the future. They present and compare results provided in reviewed works in
terms of throughput improvement, queue, waiting and travel time reduction, vehicle speed increase, speed-ups as well as
assumed simulation parameters. Finally, they focus on research gaps, such as a small number of works considering crisis
simulations, few real-world scale simulations as well as on software architectural changes and low-level optimizations.
1Introduction
Since 1886, when Karl Benz patented the car [1], the population
has been growing steadily [2] and the demand for vehicles is
growing rapidly. This demand makes city authorities notice the
problems caused by the effects of the most popular transport type.
These problems include not only traffic jams that burden the
economy of countries [3], but also the deterioration of the health of
city residents. According to the Eurostat case study [4], with the
current sales results and methods of recycling old vehicles [5], 15
million vehicles enter city streets every year. Due to the
aforementioned problems and alarming data, many scientists are
doing research that can help reduce the negative effects of road
transport by improving safety, capacity and average speed during
congestion. All these factors have a direct impact on exhaust gas
emissions [6] and the life satisfaction of city citizens. In this work,
we focus on the latest reports in the field of road simulations,
believing that the latter currently constitute the most effective way
of managing city traffic. In our analysis, we aimed at finding works
that show the greatest percent of improvement on urban roads.
Additionally, we looked for problems that appear in most of the
studies in this domain. Finally, the work describes simulation
methods that have been used in recent years and highlight the most
valuable ones.
In our work, we described each of the simulation methods that
have been related to city simulations over the last few years.
Besides that, we describe the cores of these works and unusual
findings. Section 3 analyses problems and solutions related to the
movement of vehicles in the urban environment, but also those
concerning simulators or smart cities in the future as well as goals
set for them. Section 4 allows to track the latest trends in the
literature. In each section, we described each method from the
simplest used in the literature to the most complex one. In the
subsection dealing with microsimulations, we proceeded with
works focusing on cellular automata (CA), moving to multi-agent
systems (MASs) that have a very wide range of applications ending
with evolutionary algorithms and fuzzy logic. The next subsection
focuses on macrosimulations in which we included a cell
transmission model (CTM) which is similar to CA, a macro
fundamental diagram (MFD) which is based on estimating the flow
of vehicles on individual road elements, and we finished by
presenting Petri nets which show the representation of the road as a
network of states reflecting city fragments. Description of tests and
systems allowing for parallelisation of simulations in the CPU and
GPU environment follows. We also included ways of performing
calculations in cellular networks, clouds and a fog environment. To
complete the work by including other methodologies that have
emerged in this branch of science, we described other approaches
using operations research, game theory, graph theory, optimisation
of city grids and other useful tools needed for integration
simulation environments and modern web systems.
The following Section 5 includes observations that allow the
reader to spot problems in the research on simulations. There we
can find key issues crucial for this branch of science, also
presented in the form of diagrams with superimposed particular
work's results. We then outline directions of research that are
neglected such as a small number of papers dedicated to crises and
driver safety, a small number of user-friendly modern simulators
for academic environments (the direction of development of
simulation as a service) and a large disproportion in the parameter
sets used during city simulations. We believe that this work details
an extremely important problem, i.e. how to create widespread
road analysis systems so that cities can collaborate on modern
solutions that are truly groundbreaking.
In recent years a large number of research works devoted to
urban traffic optimisation have been created. This review describes
results from a most novel and interesting papers and presents these
to the reader in a transparent form which allows understanding
specific techniques of simulation from basic to most modern
approaches in the literature. In our opinion, this kind of study has
considerable value for scientists and should be repeated
periodically. For this reason, in Section 2, which follows, we
described how to repeat the research we have carried out in this
work. The work contained in this paper comes from the three
largest databases of scientific articles, i.e. Elsevier, Springer and
IEEE.
2Literature search method
This paper focuses on research which helps optimise traffic flows
or create new policies regulating traffic jams when accidents occur.
With that intention, the phrase ‘Traffic AND simulations AND
IET Intell. Transp. Syst., 2020, Vol. 14 Iss. 13, pp. 1740-1758
© The Institution of Engineering and Technology 2021
1740