Trap Rev
Trap Rev
TRAP
REVIEW-1
COURSE CODE-SWE3999
SUBMITTED BY
M.SIVANI-16MIS0071
P. MEDHAVINI-16MIS0166
B. TEJASWINI-16MIS0191
A. KIRTI REDDY-16MIS0406
R. PRATHYUSHA-16MIS0423
PROBLEM STATEMENT
The main objective of this project is to create a system which senses the garbage capacity filled
of a dustbin and this data is collected from various places in the city. This is done by using IOT
enabled devices.This data is then given to an algorithm which based on which dustbin is fuller
than others, and the locations of those dustbins in a city will find an optimized path for trucks
to follow to empty the garbage.
OBJECTIVES
1.Create a system which will sense the garbage capacity in a dustbin and upload it to the cloud
in a Thing-Speak channel.
3 Export the data collected in the Thing-Speak channel into a CSV file for route optimization.
4 Design a system which implements the Capacitated Vehicle routing problem with pickup and
delivery for a given csv file.
5 Feed the csv file exported to the route optimization system to get the efficient route for the
garbage trucks to follow.
ARCHITECTURE
The data of the capacity of the bins is collected from the sensors installed in the bin and they
are uploaded to Thing-Speak platform with the coordinates of the dustbin (stored in the sensor
system). Then the municipal hub of the city will be able to see this information in the form of
a route for the garbage trucks to follow in the city.
The Hub PC or android phone will be able to connect to the internet to get this route. The
route is calculated by the app or the program automatically by collecting the latest sensor data
from Thing-Speak and applying the route optimization algorithm to find the route. The route
optimization algorithm will read the csv files as shown in figure and consider the latest
capacity data of each bin by maintaining an array for that purpose.
Then this data is taken and the weights of the graph are updated by giving some weightage to
these capacity values. i.e., a full dustbin at a certain destination or endpoint of an edge
reduces that edge weight. After recalculation of the weights, the CWRPWPD is applied on
the graph, to find the route for each truck to follow.
LITERATURE REVIEW
NATIONAL STATUS
Mahmuda Akhtar et al. in [2] have come up with a method to optimize routes by using back
tracking search algorithm. They also came up with a Capacitated Vehicle Routing Problem
(CVRP) model like in [1] then they applied the Backtracking search algorithm (BSA). It is a
new population based metaheuristic algorithm. The BSA algorithm consists of initialization,
selection, mutation + crossover, selection 2 stages. All these steps analyse data collected from
the neighbourhood bins and trial populations to compare the fitness values of trial
populations with the initial population and the population is updated based on that.
Anitha A in [10] came up with a garbage monitoring system using IOT enabled devices
namely Arduino UNO. She used IOT parts like Arduino, WiFi module and android devices to
connect with each other in order to send and receive the data collected be the ultrasound
sensors which are placed on the top of the bin to collect the capacity of the bin which is full
and send the data collected to the wirelessly connected android device by which the user be
able to view the collected data in an interactive GUI format so that he/she can take further
steps in the waste management of it.
INTERNATIONAL STATUS
M.A Hannan et al. in [1] have discussed route optimization of scheduled wasted collection
trucks using PSO algorithm. The PSO algorithm is population-based algorithm and it is a
meta-heuristic optimization method that performs very efficiently in routing problems. In
PSO the algorithm produces a cluster of particles that are moved close to the optimized value
by some velocity which is initialized. By remembering the two values of value found by
particular particle and the best solution of the neighbourhood particles this algorithm works.
Radek Fujidiak et al. in [3] have proposed a method of using genetic algorithm for efficient
municipal waste collection in smart cities. Then they mapped the data collected by the
position of the bins on a map and they provided this map to their algorithm as input in the
form of Scalable Vector Graphics with coordinates. They used classical graph representation
of the map. Then XML parsing is used on the input graph and then Floyd Warshall’s
algorithm is applied for distance recalculation and then it is solved like a Travelling Salesman
problem. This was their first attempt at arriving at a path. Then they used their proposed
genetic algorithm on the data and compared the two methods to find that their method is more
efficient.
SOCIETY RELEVANCE
Garbage collection is a problem in India due to limited number of garbage trucks and the
lower capacity of garbage trucks. So more often than not some places have dustbins full of
garbage for days and not emptied.
Garbage truck drivers follow shifts to collect the garbage from different areas on different
days of the week, but areas with high garbage disposal are not given special attention.
So this problem can be solved by giving more priority for full bins and emptying them first.
For that the capacity data is collected using IOT enabled devices and used to get an optimal
path for trucks.