Journal of Physics: Conference Series

PAPER • OPEN ACCESS

Analysis of Dijkstra’s Algorithm and A* Algorithm in Shortest Path

Problem
To cite this article: Dian Rachmawati and Lysander Gustin 2020 J. Phys.: Conf. Ser. 1566 012061

View the article online for updates and enhancements.

This content was downloaded from IP address 178.171.20.221 on 04/07/2020 at 13:24

ICCAI 2019 IOP Publishing
Journal of Physics: Conference Series 1566 (2020) 012061 doi:10.1088/1742-6596/1566/1/012061

Analysis of Dijkstra’s Algorithm and A* Algorithm in

Shortest Path Problem

Dian Rachmawati1* and Lysander Gustin2*

1
Departemen Ilmu Komputer, Fakultas Ilmu Komputer dan Teknologi Informasi,
Universitas Sumatera Utara, Jl. Universitas No. 9-A, Medan 20155, Indonesia

*Email: [email protected] and [email protected]

Abstract. Finding the shortest path in direction effective is essential. To solve this shortest
path problem, we usually using Dijkstra or A* algorithm. These two algorithms are often used
in routing or road networks. This paper's objective is to compare those two algorithms in solving
this shortest path problem. In this research, Dijkstra and A* almost have the same performance
when using it to solve town or regional scale maps, but A* is better when using it to solve a
large scale map.

1. Introduction
Technology has been helping us a lot, from sending email, text messaging, social media, and even
maps. People using maps for many things, e.g., searching for places such as searching for the nearest
restaurant, café, gas station, or finding direction from home to work. When requesting a route from
one point (starting point) to another location (destination point), usually, the result that comes out is
the "shortest path" from starting point to destination point.
The shortest path problem has been studied for many years. The shortest path problem is the problem
that finds the minimum distance or pathway between nodes or vertices in a graph (for this case, road
network). A graph is an abstract mathematical object, which contains sets of vertices and edges [1].
Regarding to journal from Parveen Sharma and Neha Khurana, the shortest path problem is defined as
that of finding a minimum-length (cost) path between a given pair of nodes [2] and according to
Karishma Talan and G. R. Bamnote, shortest path problem is the problem of looking for the quickest
way to get route from one location to another [3]. Expressed more formally, in a graph in which vertices
are joined by edges and in which each edge has a value or cost, it is a problem of finding the lowest-
cost path between two vertices [4]. Distribution is one of marketing activities in aim to facilitate the
producers in sending goods to the customers. In the distribution process, a set of customers is served by
a producer where located in an area. A set of crews operates a set of vehicles which is located in the
producer place called a depot to fulfill the each customer demand. As a producer, for minimizing their
total operational transportation cost of distribution, the depot has to find an appropriate road network
for the vehicles [5].
Many algorithms solve the shortest path problem. Dijkstra’s algorithm is one form of the greedy
algorithm. This algorithm includes a graph search algorithm used to solve the shortest path problem
with a single source on a graph that does not have a negative side cost and produces the shortest path
tree. This algorithm is often used in routing [6]. A heuristic is a method designed for solving a problem
more quickly. This is achieved by trading accuracy, optimality, completeness, or precision for speed.
A* algorithm is an algorithm that finds a path in a graph from a given initial node to a given goal node.
It employs a "heuristic estimate" h(x) that gives an estimate of the best route that goes through that
node. It visits the nodes in order of this heuristic estimate [7]. This paper will compare Dijkstra’s
algorithm and A* algorithm to solve the shortest path problem. There are four stages in mathematical
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ICCAI 2019 IOP Publishing
Journal of Physics: Conference Series 1566 (2020) 012061 doi:10.1088/1742-6596/1566/1/012061

modelling: problem identification, constructing a mathematical model, determining mathematical

solution from the model, and interpreting mathematical solution into real problem points of view [8].

2. Method

In this implementation, we will use 2 examples from the University of Sumatera Utara at Jl. Universitas
No. 9-A (starting point) to Wisma Benteng Restaurant at Jl. Kapt. Maulana Lubis No. 6 (1st destination
point) and from the University of Sumatera Utara at Jl. Universitas No. 9-A (starting point) to the State
University of Medan at Jl. William Iskandar Ps. V (2nd destination point). Between the starting point
and destination point, we manually generate plenty of points (nodes), which is an intersection between
2 or more roads and find the distance between them (edges). After that, we apply the data to Dijkstra's
Algorithm, A* algorithm, and make a comparison between the two algorithms based on the running
time, the number of loops with the number of points as the factor. The processes are as below.

2.1 Steps for getting the distance between intersection and distance between point to destination point:

First, we need to manually generate point which is the intersection between 2 or more roads that is
possible to be crossed between starting point to both destination points (1st destination point is marked
with green color and 2nd destination point is marked with orange color). We use Google Maps or
Geographic Information System to map the point, as shown below.

Figure 1. Mapping all the possible intersection with map background

2
ICCAI 2019 IOP Publishing
Journal of Physics: Conference Series 1566 (2020) 012061 doi:10.1088/1742-6596/1566/1/012061

Figure 2. Mapping all the possible intersection without map background

After that, we calculate the distance between points connected by the road and estimate the distance
between point to the destination point for heuristic value by using the help from Google Maps.

Table 1. Distance between points connected by the road

From point To point Distance / Weight (in km)

start 2 0.23
start 1 0.084
1 5 0.5
2 3 0.5
… … …
101 65 0.14

Table 2. Distance between point to 1st destination point

From Distance / Weight (in km)

1 3.283108
2 3.441482
3 3.018227
4 2.889003
… …
34 0.677927

3
ICCAI 2019 IOP Publishing
Journal of Physics: Conference Series 1566 (2020) 012061 doi:10.1088/1742-6596/1566/1/012061

Table 3. Distance between point to 2nd destination point

From Distance / Weight (in km)

1 7.91655
2 8.185879
3 7.889952
4 7.696507
… …
101 4.543907

2.2 Steps for calculating the shortest path:

Dijkstra’s Algorithm steps:

1. Set all points distance to infinity except for the starting point set distance to 0.
2. Set all points, including starting point as a non-visited node.
3. Set the non-visited node with the smallest current distance as the current node "C."
4. For each neighbor “N” of your current node: add the current distance of “C” with the weight of
the edge connecting “C” – “N." If it's smaller than the current distance of “N," set it as the new
current distance of “N."
5. Mark the current node “C” as visited.
6. Repeat the step above from step 3 until the destination point is visited.

A* algorithm is just like Dijkstra’s algorithm, and the only difference is that A* tries to look for a
better path by using a heuristic function, which gives priority to nodes that are supposed to be better
than others while Dijkstra's just explore all possible ways.

A* algorithm steps:
1. Set all point distance to infinity except for the starting point set distance to 0.
2. Set all points, including start point as a non-visited node.
3. Set the non-visited node with the smallest current distance as the current node "C."
4. For each neighbor “N” of your current node: add the current distance of “C” with the weight of
the edge connecting “C” – “N” and the weight to the destination point (heuristic). If it's smaller
than the current distance of “N," set it as the new current distance of “N."
5. Mark the current node “C” as visited.
6. Repeat the step above from step 3 until one of the neighbors “N” is the destination point.

We build the Dijkstra’s Algorithm program and A* program as shown below.

After the program finish, we input all the data from the previous step to the program.

4
ICCAI 2019 IOP Publishing
Journal of Physics: Conference Series 1566 (2020) 012061 doi:10.1088/1742-6596/1566/1/012061

Figure 3. Dijkstra’s Algorithm Source Code

Figure 4. A* Algorithm Source Code

3. Results and Discussions

The experiments were performed on laptop Windows 10 pro with 8GB of RAM and Intel i5-6200
processor with PyCharm version 2018.3.5 64-bit version evaluation copy.

3.1. Starting point to 1st destination point (Total of 36 points include start and destination point)

Running Time
We take 3 tries for each Dijkstra and A*, running times are shown below.

5
ICCAI 2019 IOP Publishing
Journal of Physics: Conference Series 1566
st (2020) 012061 doi:10.1088/1742-6596/1566/1/012061
Table 4. Running Time from starting point to 1 destination point using Dijkstra and A* algorithm

Test Number Running Time Running Time A* Total Distance (in

Dijkstra (nanoseconds) km)
(nanoseconds)
1 997.200 996.800 4.504
2 998.200 996.900 4.504
3 997.800 997.400 4.504

Loop Counts
We also take 3 run of each algorithm and the result shown below.

Table 5. Loop counts from starting point to 1st destination point using Dijkstra and A* algorithm

Total Distance (in

Test Number Loop Count Dijkstra Loop Count A*
km)
1 36 26 4.504
2 36 26 4.504
3 36 26 4.504

3.2. Starting point to 2nd destination point (Total of 103 points include start and destination point)
Running Time
We also take 3 tries for each Dijkstra and A*, running times are shown below.

Table 6. Running Time from starting point to 2nd destination point using Dijkstra and A* algorithm

Test Number Running Time Running time A* Total Distance (in

Dijkstra (nanoseconds) km)
(nanoseconds)
1 2,991.400 2,992.000 10.138
2 3,987.800 2,992.800 10.138
3 3,989.100 2,992.100 10.138

Loop Counts
We also take 3 run of each algorithm and the result shown below.

Table 7. Loop counts from starting point to 2nd destination point using Dijkstra and A* algorithm

Total Distance (in

Test Number Loop Count Dijkstra Loop Count A*
km)
1 103 72 10.138
2 103 72 10.138
3 103 72 10.138

4. Conclusions
In conclusion, the use of Dijkstra’s algorithm and A* algorithm in the shortest path is essential will give
the same output in no time when being used on the town or regional scale maps. But on a large scale
map, A* will provide the solution faster than Dijkstra. A* scan the area only in the direction of
6
ICCAI 2019 IOP Publishing
Journal of Physics: Conference Series 1566 (2020) 012061 doi:10.1088/1742-6596/1566/1/012061

destination because of the heuristic value that counted in the calculation, whereas Dijkstra searches by
expanding out equally in every direction and usually ends up exploring a much larger area before the
target is found resulting making it slower than A*. This can be proven by the loop count of Dijkstra and
A*, the more points (nodes) the higher the difference between the loop count nor the time.

5. Acknowledgments
The authors gratefully acknowledge that the present research is supported by Fund Dissemination
IPTEKS Research Results for Lecturers / Researchers Universitas Sumatera Utara.

References
[1] Kairanbay M, and Mat Jani H 2013 A Review And Evaluations Of Shortest Path Algorithms. Int.
J. of Sci. & Tech.Res.2 6 99.
[2] Sharma P, and Khurana N 2013 Study of Optimal Path Finding Techniques. Int. J. of Adv.in Tech.
4 2, 124.
[3] Talan K, and Bamnote G R 2015 Shortest Path Finding Using a Star Algorithm and Minimum
Weight Node First Principle. Int. J. of Innovative Res. in Computer and Communication
Engineering 3 2 1258.
[4] Hart E, Nilsson N J and Raphael, B 1968 A Formal Basis for the Heuristic Determination of
Minimum Cost Paths Systems Science and Cybernetics 4 2 100-107.
[5] Pratiwi A B, Pratama A, Sa'diyah I and Suprajitno H 2019 Vehicle routing problem with time
windows using natural inspired algorithms Department of Mathematics, Faculty of Science
and Technology, Universitas Airlangga, Surabaya, Indonesia
[6] Wahyuningsih D and Syahreza E 2018 Shortest Path Search Futsal Field Location With Dijkstra
Algorithm. Indonesian J. of Computing and Cybernetics Systems 12 2 161-170.
[7] Goyal A et al 2014 Path Finding: A* Or Dijkstra’s? Int.J. of Innovative Trends in Engineering,
2 1, 1-15.
[8] Windarto, S. W. Indratno, N. Nuraini, and E. Soewono 2019 A comparison of binary and
continuous genetic algorithm in parameter estimation of a logistic growth model AIP Conf. Proc.
1587 pp. 139–142 doi: 10.1063/1.4866550.