Using an Innovative Model to Automate an Efficient Taxi Dispatching System in Mauritius

www.ijemr.net ISSN (ONLINE): 2250-0758, ISSN (PRINT): 2394-6962
119 Copyright © 2018. IJEMR. All Rights Reserved.
Volume-8, Issue-4, August 2018
International Journal of Engineering and Management Research
Page Number: 119-125
Using an Innovative Model to Automate an Efficient Taxi Dispatching
System in Mauritius
Pillay Kanaksabee1
and Aviyash Domah2
1
Lecturer, Department of Business Informatics and Software Engineering, University of Technology Mauritius (UTM),
MAURITIUS
2
Software Engineer, Agileum, MAURITIUS
1
Corresponding Author: pkanaksabee@umail.utm.ac.mu
ABSTRACT
The world is rapidly changing with the introduction
of innovative technologies to meet the ever-changing demands
of consumers. This research focuses mainly on the study of
the actual taxi system in Mauritius to manage their everyday
activities in an efficient way to earn a good and a decent
earning. With over 6000 drivers in Mauritius, this has caused
a major issue in the transportation sector. According to the
survey conducted in Shanghai, China, the percentage of
empty miles ((total miles-service miles)/total miles) for taxi is
32%, while the waiting time for passenger may extend more
than 1 hour in the rush hours at the worst. The taxi employed
under this actual system is rarely capable of reaching the
customers in the shortest possible time. This is an exigent
problem actually confronting current taxi systems. Thus, an
alternative Interactive Taxi dispatching system is proposed,
whereby the system will be fast and cost effective to transfer
the customer’s request from the operator to the most suitable
vehicle. This dispatch ensures that customers are served
within the shortest time, resulting in satisfying customer’s
demands. Since existing taxi system are working manually,
there is a need to have automatic taxi trip system which
would be easy and efficient. [14] Therefore, the aim of this
paper is to research the pertinence of the use of an Interactive
Taxi Dispatching System as an alternative mode for taxi
drivers to easily be connected with their riders.
Keywords--- Taxi Drivers, Taxi Dispatching System, Taxi
Booking, Quick Transport, Cost Effective Taxi
I. INTRODUCTION
Taxis have an imperative role in contributing
constructive services within Mauritius transportation
sector. In the midst of the growing prominence for
providing satisfaction to customers, it is fundamental for
taxi drivers to continuously improve their current systems
and facilities to guarantee quality and superior service. Our
everyday life has changed rapidly and is connected always
with devices around us. We can have anything we need at
any time and people simply expect to be able to connect
wherever they are. Thus, this interactive taxi dispatching
system helps riders to quickly find the most suitable taxi
for their transportation and allow drivers to simply connect
to passengers through a fair reward system. In the end,
what we want is to have a smarter, faster, and more
effective transportation platform for all people to use daily.
Nowadays, taxis are available everywhere in
Mauritius to provide services to serve their customers in
each region. However, many taxis roaming on the streets
counter to request passengers on their way, and take
passengers to their specific and desired destinations. As a
result, existing taxi networks are dependent on drivers to
drive around in the region randomly to pick up passengers
on the roads. This service model has successfully served
up to 25% public passengers in metropolitan such as San
Francisco and New York [1], [2].
In TAS (Taxi Automation System), online
dispatching of available taxis to meet the current customer
bookings are done normally with the help of a satellite-
based taxi automation system to meet the current customer
bookings; that is the system uses a Global Positioning
System (GPS) to repeatedly locate taxis location in factual
time. [3]. The current method consists of assigning the
nearest taxi in first come first serve queue basis without
taking into consideration the passengers who are in the
request queue. To improve performance of taxi fleet
service, preferably, the assignment taxis should be done
concomitantly and optimally in order to service all
customer bookings within the time frame. [4]
By means of comparing to the existing dispatch
system, all the existing direct distances of the taxi to the
current customer locations are generally calculated.
However, this process needs pretty little space for storage
and computation time. Conversely, for the anticipated
dispatch system, the calculation of the shortest possible
path for every available taxi required a significant larger
room for memory and calculation time. [5] In regards to
the speed and memory restrictions of computers, it was not
a practical alternative to establish and calculate the shortest
paths for all the available taxis.

The existing taxi operation approach features
elevated number of empty taxies wandering around the
streets, whilst plenty of potential riders are irritated with
the fact of not finding a taxi when needed. [6] The key
reason for this inadequacy is that the current taxi dispatch
system is unable to exchange appropriate information in a
real-time manner between taxi drivers, riders and the
dispatching system.
The inefficiency in dispatching system not only
makes a major impact on passenger waiting time and
passenger satisfaction, but also increases the taxi driver
idling time. Though, drivers said that though they do not
mind waiting for hours, but still they find waiting at the
terminal enormously time-consuming and tiring. [7] The
drivers are likely to wander near residential areas during
morning peak and wait until they get a passenger. The
incompetency of the system arises; even when the market
is correctly synchronized and harmonized, due to the
deficiency of the perfect widely spread information sharing
between taxi drivers and passengers. [8]
To improve taxi fleet service performance,
ideally, we should concurrently and optimally assign taxis
to service all customer bookings that are made within the
time window. This is an exigent problem confronting
current taxi dispatch systems. [12]
II. PROBLEM STATEMENT
The inefficiency in the current approach of
booking a taxi is that it is incapable of reaching the rider in
the shortest time and determining the actual rider’s real-
time location. As a result, many people are deprived from
using taxi service. it is difficult to gain the distribution of
passengers since passengers can appear at any place and
any time in the city. [15]
III. OBJECTIVES
The objective of this Interactive Taxi Dispatching
system is to provide riders with taxi services effectively
and safely, based on the location of the taxi driver and
rider. In other word, match the rider and available taxi
properly in term of distance and availability of the nearest
taxi first.
IV. HYPOTHESIS
An Android app for both drivers and riders along with
the implementation of an online booking website and the
online operator dispatch system will ease the task for the
booking process.
V. RESEARCH QUESTIONS
1) What are the actual procedures for
searching for a taxi in Mauritius?
2) What are the problems faced by
Mauritians when searching for a
taxi?
VI. LITERATURE REVIEW
A. Procedures for seraching for a Taxi
In this random searching mode, passenger is
waiting for a taxi on street and empty taxi is moving
around looking for a passenger, both passenger and taxi do
not know each other’s location. Passengers and taxis are
randomly searching for each other as shown in the figure
below.
Figure 1.0: Procedures for searching a taxi randomly
In this mode, when a passenger needs a taxi, the
only thing he or she can do is standing on street and
waiting for a vacant taxi passes by. The waiting time of the
passenger depends on the probability that a vacant taxi
passes by during his/her waiting. [9]
Another mode is when a taxi is waiting for a
passenger at a heavy traffic center or stop point. The
location for taxi is known but the arriving passenger is
unknown as shown in the figure below.
Fig 1.1: Taxi waiting for a passenger
For a vacant taxi, it reduces empty driving
searching for passengers but the waiting time could be very
long. After several interviews with taxi drivers, we find
that the average day waiting time of taxi at Pudong
International Airport in Shanghai to get a passenger is 15
minutes. [10]
B. Problems faced when searching for a taxi
More cars may pass on a busy street, and during
rush hours, thus there may be few vacant taxis passing by.
An unlucky passenger may wait long time without meeting
any available bypass taxi. For the driver of a vacant taxi,
he/she has to keep moving the taxi and hopes to find a
passenger waiting on the bypassing street. Since the
location of potential passenger is unknown, the taxi driver
has to decide which street to roam through in order to
increase the probability to meet a passenger. The
unproductive idle driving and waiting time is caused by

lack of taxi and passenger location information. In a big
city, usually there are so many idle taxis roaming around
causing traffic jam and air pollution. This may also cause
accidents when a taxi driver is looking for a passenger
without paying enough attention on driving. [11]
From a taxi driver’s point of view, an efficient
taxi system is one in which his or her taxi is never empty.
By contrast, a customer’s idea of an efficient taxi system is
one that delivers an empty taxi to his or her location at the
instant one is desired. Obviously, these goals are in
conflict. [13]
VII. METHODOLOGY
In order to develop a suitable system, requirement
determination process was taken into consideration.
Software tools used are HTML/HTML5, CSS3,
Javascript/Bootstrap/AngularJS, JQuery/Jquery Mobile,
PHP 5.0, Wamp Server, Android Studio 5.0 ADT for
executing the application, MySQL for storing the database
of users and Ionic framework/ Cordova environment for
coding the program for the application. Moreover, Agile
approach was used for the software development process.
Additionally, to contrast between the current booking
system and the Dispatching system, a table is provided
below:
Table 1.0: Contrast between current system and Dispatching Centre
Current system Dispatching Centre
Reservation You wander around the street to find a taxi or
phone a taxi driver to come and pick you up.
You connect to Fast Taxi app on your
Smartphone.
The Ride The driver is barely polite and he doesn’t care
because he may never see you again. The car
is not always clean.
The driver is nice and polite and the car is tidy.
The drivers know that they will be evaluated after
the ride
Paying The price is determined by the driver himself
and sometimes these prices may exaggerate.
Fast Taxi calculates the tariff based on the
distance travelled.
Emergency Wait until someone comes for help. Press the alarm button and the operator will take
immediate actions
Afterwards That’s it! It’s over! Rate the driver to improve our reputation for
future customers
VIII. PROPOSED SYSTEM
The ideal solution is to develop an interactive
dispatching system which will be android-based and web–
based applications and which can be accessible on mobile
devices such as smart phones and tablets. Several APIs
will be integrated in the system. The content can also be
customized to be used on any devices, including PDAs,
mobile phones, etc., that are connected to the internet
depending on the users’ device. Furthermore, extending the
user’s ability to receive and interact with the information.
Fig 1.2: The Proposed System
Initially, there are three modes for a booking a taxi.
1) Instant Booking using the app
2) Advanced booking using the website
3) Direct call to the DISPATCHIG CENTER
By knowing both rider’s and taxis’ locations, the
DISPATCHING CENTER can determine which taxis are
available nearby to serve the rider. Explicitly, this system
will work as follows. Preliminary, when a rider needs a
taxi, he can simply send a booking request for a taxi
through a wired phone, a mobile phone or Internet.
DISPATCHING CENTER will take the rider’s request and
automatically attain the current location of both the rider
and taxis. After both of their locations are known,
DISPATCHING CENTER searches the nearest taxis. Then
if there are some taxis available, the DISPATCHING
CENTER will choose the nearest taxi to the waiting rider,
and order the taxi to pick up the rider. Then,
DISPATCHING CENTER will send the taxi details and
the expected arrival time of the driver back to rider. When
the driver arrives at the rider’s location, he notifies the
rider. Once the ride is completed, the rider will receive
another notification for the completion of the ride and the
rider can rate his/her booking. If there is no available taxi,
the DISPATCHING CENTER will notify the rider the sad
news and wait for 15 minutes until a taxi is available. Both
the rider and the driver can cancel a booking at any time
due to some reason. But, if a rider cancels a ride more than
3 times, he/she will be block automatically and has to
contact the site admin to reuse the system. Furthermore,
the rider can call the driver and the DISPATCHING

CENTER in case some problems arise. Both the rider and
the driver can make use of the panic button to alert the
DISPATCHING CENTER in case of emergencies.
IX. RESULT
The splash screen of the Fast Taxi app will
appear, and will check if the phone’s WI-FI is on and if it’s
not it asks users to turn it on.
The booking process takes place as follows:
1) Once both the drivers and riders open their app,
their current location is displayed.
2) Both drivers and riders must login to use their
app.
3) Rider can make an instant booking using the app
as shown below.
Assume that the booking is being done from UTM BPML
Road, Port Louis, Mauritius and its destination to Triolet,
Pamplemousses, Mauritius
4) Once the rider will click on the book button, the
following notification will appear on his screen.
5) The rider receives a confirmation about his/her
particular booking and will receive another notification to
confirm the booking.
6) At the same time, the driver who can satisfy the
criteria mentioned by the rider and having the shortest
distance path will be alerted.

7) Once the driver confirms the ride, the rider will be
alerted on his app.
8) Simultaneously, the driver will receive a
notification to pick up his rider.
9) A rider can also cancel a booking. However, the
rider cannot cancel a ride more than 3 times in a month or
else he will be blocked. The driver will receive a
notification that the booking has been cancelled.
10) Upon arrival, the driver will notify the rider by
clicking on the notify button
11) Concurrently, rider will receive the notification
that the driver has arrived and the driver will receive a
confirmation message that he has notified the rider.
12) During the ride, both the rider and the driver can
view the booking details on their app.
13) Use the panic button in case of an emergency or
even make a call to Fast Taxi.

14) On completion of ride, the driver will click on the
ride completed button and the rider will receive the
following message. At the same time, the driver will also
be notified to login for a new ride.
15) When rider will click on Rate button, the following will
appear to rate the booking.
16) Both the rider and the driver can view live news, sport
news, and play games.
X. CONCLUSION
This Interactive Taxi Dispatching system is an
innovative approach to the Mauritian people. As a whole,
this proposed system will allow the Dispatching center to
lessen the time to pick up the rider, consequently
improving the accuracy and efficiency of their dispatch
systems accordingly, and thus delivering a superior level of
guaranteed service to the riders. With the diminution in
waiting times, more riders would agree to take on taxis
through bookings, rather than using other means of hiring
them when needed. This would let the taxi operators to
better manage and optimize the taxi fleets in an efficient
way, as a much better match of demand and supply of taxis
could be met to satisfy customer’s needs.
In the near future, to further improve the
performance of this proposed system and at the same time
to use the system to its full potential, advanced research
can be carried out to gain and ascertain new methods or
ways to obtain accurate estimations of the signal delays on
the road, so that they can be incorporated into the
estimated travel times to give the rider much more accurate
information about the drivers’ arrivals. This will help
during the decision-making moment of dispatch to give a
better estimate of the arrival time of a taxi. This needs a
function that will predict the delays caused by signals that
normally an individual vehicle experiences at intersections,
would enhance the accurateness of the proposed system.
REFERENCES
[1] Hara Associates Inc. and Corey, Canapary &
Galanis.(2013). Taxi user surveys. Available at:
http://www.sfmta.com/sites/default/
files/Draft%20SF%20UserSurvey%2055%20WEB%
20version04042013.pdf, 2013
[2] Taxi of tomorrow survey results. New York City Taxi
and Limousine Commission. (2001). Available at:
http://www.nyc.gov/html/tlc/
downloads/pdf/tot_survey_results_02_10_11.pdf, 2011.
[3] Marshall Brain & Tom Harris. How GPS receivers
work. Available at:
http://electronics.howstuffworks.com/gadgets/travel/gps.ht
m
[4] Anurag Mandle, Akshay Jaiswal, Bhushan Dod, &
Roshan Lokhande. (2014). Taxi automation using real time
adaptive scheduling. International Journal on Recent and
Innovation Trends in Computing and Communication,
2(3), 592-594. Available at:
http://www.ijritcc.org/download/Taxi%20Automation%20
Using%20Real%20Time%20Adaptive%20Scheduling.pdf
[5] Clarke G. & Wright J. (1964). Scheduling of vehicles
from a central depot to a number of delivery points.
Operations Research, 12(4), 568-581.
[6] Para Kalpana & Dr. A.Lakshmi Devi. (2013).
Placement and sizing of distributed generators in
distributed network based on lric and load growth control.
Journal of Theoretical and Applied Information
Technology, 47(1), 1-9.
[7] Anil Yazici et. al. (2011). Challenges in managing
centralized taxi dispatching at high-volume airports: A
case study of John. F. Kennedy international airport. 1-18.
http://docs.trb.org/prp/12-2980.pdf
[8] Xianyuan Zhan, Xinwu Qian, & Satish V. Ukkusuri.
(2014). Measuring the efficiency of urban taxi service
system. 1-8. Available at:
https://pdfs.semanticscholar.org/f24b/758084a9db94e4118
7f7042b1d880382908a.pdf

[9] Y. Yuan, J. Zhang. (2003). Towards an appropriate
business model for m-commerce. International Journal of
Mobile Communication, 1(1-2), 35-56.
[10] Marco Kouwenhoven. (2008). The role of
accessibility in passengers' choice of airports. OECD, 1-
38.Available at: https://www.itf-
oecd.org/sites/default/files/docs/dp200814.pdf
[11] XU Zhengchuan, YUAN Yufei, JIN Huiliang, &
LING Hong. (2005). Investigating the value of location
information in taxi dispatching services: A case study of
DaZhong tax. Available at: http://www.pacis-
net.org/file/2005/273.pdf
[12] Khadaroo, A. J. & Seetanah, B. (2007). Assessing the
contribution of land, sea and air transport capital to the
economic performance of the small island state of
Mauritius. Applied Economics Letters, 14(15), 1151-1155.
[13] D. Santani, R. K. Balan, & C. J. Woodard.
(2008,June). Understanding and improving a GPS-based
taxi system. 1-2. Avalable at:
https://apollo.smu.edu.sg/papers/MobiSys08_poster.pdf
[14] Maryam Ahmed, Zahara Batool, & Mahrukh Raheel.
(2015). Automatic taxi trip sensing and indicating system
through gsm. American Journal of Computer Science and
Engineering, 2(5), 42-48.
[15] N. Streitz, C. Stephanidis (Eds.). (2013). Distributed,
ambient, and pervasive interactions. Berlin, Heidelberg:
Springer-Verlag, 326–335. Available at:
https://www.springer.com/gp/book/9783642393501
[16] M. Qu, H. Zhu, J. Liu, G. Liu, & H. Xiong. (2014). A
cost-effective recommender system for taxi drivers. In 20th
ACM SIGKDD International Conference on KDD, New
York: USA, 45-54. Available at:
https://www.cse.cuhk.edu.hk/irwin.king/_media/presentati
ons/p45-qu.pdf
[17] Google Traffic API. (2015, June). Available at:
https://developers.google.com/maps/documentation/
avascript/examples/layer-traffic

Using an Innovative Model to Automate an Efficient Taxi Dispatching System in Mauritius

More Related Content

What's hot (20)

Similar to Using an Innovative Model to Automate an Efficient Taxi Dispatching System in Mauritius (20)

More from Dr. Amarjeet Singh (20)

Recently uploaded (20)

Using an Innovative Model to Automate an Efficient Taxi Dispatching System in Mauritius