Smart City IoT System - CollectMyWaste

Abstract

In the current day and age, humanity relies on technology to solve the vast majority of the adversities that it faces. Without exception, the main goal of this research is to discover an efficient and innovative alternative to traditional waste management, resorting to cutting-edge technologies. Due to the nature of the problem, the search for a solution will be broken into three different parts. The first is based on the regular monitoring of waste inside a specific group of containers, and how can the captured information be safely transmitted through the network. The second part of the solution aims to house every bit of information provided, in order to generate optimal and efficient waste collection routes. The third, and last part of the solution, serves as a bridge between the periodically updated information and the users. The web-based platform, CollectMyWaste, will provide to its managers all the information stored and generated, with the intention of maintaining streets clean and prevent unnecessary expense. In addition, a state-of-the-art study will be performed aiming to understand the existing solutions, and which aspects make them succeed or fail. Furthermore, in order to assemble an efficient machine to machine communication network, a deeper study of the possible technologies to employ will be performed. Equally, with the means of generating optimal routes, some of the most searched algorithms and mechanisms will be studied. To conclude, a brief introduction to the developed CollectMyWaste project will be presented, highlighting some of its core functionalities.

Full text

Smart City IoT System - CollectMyWaste
Vladyslav Kalyuzhnyy
ESTCB
IPCB
Castelo Branco, Portugal
s.vladyslav@protonmail.com
João Santos
TIMWE Lab
Covilhã, Portugal
joao.santos@timwetech.com
Mónica Costa
R&D Unit in Digital Services,
Applications and Content
Polytechnic Institute of Castelo Branco
Castelo Branco, Portugal
monicac@ipcb.pt
Pedro Silva
R&D Unit in Digital Services,
Applications and Content
Polytechnic Institute of Castelo Branco
Castelo Branco, Portugal
psilva@ipcb.pt
Abstract—In the current day and age, humanity relies
on technology to solve the vast majority of the adversities that it
faces. Without exception, the main goal of this research is to
discover an efficient and innovative alternative to traditional
waste management, resorting to cutting-edge technologies. Due
to the nature of the problem, the search for a solution will be
broken into three different parts. The first is based on the
regular monitoring of waste inside a specific group of
containers, and how can the captured information be safely
transmitted through the network. The second part of the
solution aims to house every bit of information provided, in
order to generate optimal and efficient waste collection routes.
The third, and last part of the solution, serves as a bridge
between the periodically updated information and the users.
The web-based platform, CollectMyWaste, will provide to its
managers all the information stored and generated, with the
intention of maintaining streets clean and prevent unnecessary
expense. In addition, a state-of-the-art study will be performed
aiming to understand the existing solutions, and which aspects
make them succeed or fail. Furthermore, in order to assemble
an efficient machine to machine communication network, a
deeper study of the possible technologies to employ will be
performed. Equally, with the means of generating opti mal
routes, some of the most searched algorithms and mechanisms
will be studied. To conclude, a brief introduction to the
developed CollectMyWaste project will be presented,
highlighting some of its core functionalities.
Keywords—smart cities, internet of things, waste
monitoring, route optimization, waste management
I. I
NTRODUCTION
The amount of waste produced has been continually
rising over the past years, which requires the implementation
of new waste management policies and new waste treatment
methods. However, despite the upsurge of recycling bins,
collecting points and land disposal units, additional problems
arise, particularly on the level of efficiency and cost of the
waste collection services. The escalation of the number of
waste containers rose the number of fleet vehicles, thus
increasing fuel consumption and overall service expenses.
The CollectMyWaste project aims to minimize aggregated
costs and maximize the efficiency of waste collection
services through the investigation and development of new
and innovative solutions, allowing to smart monitor waste
containers in real time and, in turn, deploy optimal and
efficient waste collection routes. The innovation promised by
this project is set on four different pillars. The first is the data
acquisition, regarding the fill level of each monitored
container, through sensors of diverse types and brands, via
Internet of Things technologies. The second pillar holds the
study of a complementary communication environment,
between the sensors network and the central system (Cloud),
preferably at null costs to the entity responsible for the
solution. The third pillar is responsible for generating
automatic and efficient waste collection routes, based on data
provided by the sensors in real time. The fourth and last
innovative pillar is relative to the articulation between the
application of the two involved agents in urban waste
collection: the manager and the fleet operator.
II. S
OLUTION
D
ETAILS
The scheme of the system can be divided into three
distinct layers (
Fig. 1
): System of acquisition and
communication of data – IoT; Cloud infrastructure and
support; and the applicational layer.
Fig. 1. CollectMyWaste Layered Architecture
A. Data Acquisition
1) Data Capturing
Data is obtained through ultrasonic sensors, that detect
waste volume within a given container through the emission
of high frequency waves, keeping track of time taken to
capture back the reflected signal. In addition, as the ultrasonic
sensors may not be the best suited device for a specific type
of waste, another type of sensor might be helpful in the waste
volume calculations, such as, pression sensors or weight
sensors.
978-1-7281-9090-7/20/$31.00 ©2020 IEEE
Authorized licensed use limited to: b-on: Instituto Politecnico de Castelo Branco. Downloaded on January 19,2021 at 10:29:55 UTC from IEEE Xplore. Restrictions apply.

2) Data Transmission
The transmission of information occurs in two levels,
among the sensors and the gateway, and between the gateway
and the Cloud. The first level relies on the implemented
communication module. If the sensors possess a Global
System for Mobile communications/General Packet Radio
Service (GSM/GPRS) module, information will be directly
sent through the mobile network, however, equipping each
container with those modules will skyrocket the project’s
budget. On the other hand, implementing a Wi-Fi module in
each sensor requires that every container is within the
coverage range of the specific gateway, in order to send data
to the external network. This solution in more affordable, due
to its high usability.
B. Cloud Infrastructure
The Cloud infrastructure will give support to the entire
CollectMyWaste ecosystem, being defined in three main
categories: Events, Core Functionalities and Route
Optimization.
1) Events
The event tracking module will save,
chronologically, every occurrence related to the detection
of a full container, the emptying of a container, the
detection of a malfunction or incident regarding the sensor,
and even the finalization of a given waste collecting route.
2) Core Funcionalities
All the operational data is stored in the Cloud
database and regards the entire logic behind the
CollectMyWaste solution. The rational database grasps all
crucial information about the monitored waste containers,
respective sensors, registered managers, vehicles and
routes. The module referring to the system’s rules contains
all the statistics needed to evaluate the information
provided from the registered events, helping to feed the
intelligent mechanisms in order to generate efficient routes.
In addition, a specific group of key performance indicators
(KPIs) will be emphasized, affecting the process of route
creation. Another critical function of the Cloud
infrastructure is to ensure the orchestrated distribution of
information between its intern modules and the system’s
components in a safe way, ensuring that all the access to
the information is restricted to the managers of the
implemented solution. On the other hand, the integration of
external systems will be possible through the use of
application programming interfaces (APIs), provided by
the services that the solution aims to use. The integration
of these services is needed to enrich the information
provided by the system, such as, meteorology or traffic
congestion.
3) Optimization of waste collection routes
The module responsible for the intelligent
mechanisms contains the essential algorithms to create
optimized routes, allowing them to receive diverse inputs,
with different weights, such as information reported by the
sensors, number of vehicles available, potential route
distance and time taken until completion, quantity of rubbish
collected in one route and available storage in the vehicle’s
container. In addition, every completed route will be stored
with the respective information regarding the collected waste.
C. Applicational layer
The applicational layer is constructed with two distinct
tools: CollectMyWaste Manager, which consists on a web-
based application for the managers of the waste collection
services; and the CollectMyWaste Operator, which consists
on a mobile application dedicated to the fleet members of the
business.
1) CollectMyWaste Manager
The manager will have a unique map view of the monitored
containers, gaining insight into the daily workload and
explore optimal container placement. Each container
represented on the map is colored regarding their fill level,
the manager will also have the ability to switch their state to
on or off, verify their values, possible malfunctions and fill
level. The manager will also be capable of viewing all the
information regarding the business, such as, the total number
of containers, vehicles and users. Another functionality
allows the manager to manually generate a route, selecting
each destination point at a time.
2) CollectMyWaste Operator
The fleet members are able to access a map, with each
container and their measured fill level, view routes that need
to be completed and report eventual problems through the
mobile application, CollectMyWaste Operator. When a bin is
emptied, the system automatically updates its information
and when a route is not completed it is marked as so.
III. W
ASTE
C
OLLECTION AND
M
ANAGEMENT
Generally, the existing solutions for smart waste
management are based on the use of sensors that measure the
fill level of the container, sending notifications to the network
through wireless communication. The information is then
transmitted to the fleet, informing that a specific container
requires emptying.
At commercial levels, some companies already offer a
smart waste management solution. For example, EcubeLabs
[1] offers a solution assembled with three modules: one smart
waste container (Clean CUBE) powered by sunlight, which
has the ability to compress waste in order to maximize
available storage; a ultrasonic sensor (Clean CAP) to measure
fill levels, that are able to communicate through 2G and 3G
networks; and an online platform (Clean City Networks) that
allows to view on a computer, tablet or smartphone the level
of waste in each container, as well as optimize waste
collection routes through the implementation of search
algorithms and prediction mechanisms. Meanwhile, Enovo
[2] employs a solution that allows for the monitoring of the
waste containers through ultrasonic sensors, estimate when
certain containers will be full, optimize routes and time spent
during waste collection, and know, remotely, the type and
quantity of waste collected in each stop point. Smartbin [3]
offers a similar solution, where specialized ultrasonic sensors
are employed to measure the volume of waste within the
container. In addition, the sensors are able to report its
Authorized licensed use limited to: b-on: Instituto Politecnico de Castelo Branco. Downloaded on January 19,2021 at 10:29:55 UTC from IEEE Xplore. Restrictions apply.

inclination, localization and the bin’s temperature. The
communication of those parameters is established through
cellular networks (2G/3G). This solution also provides a web-
based platform, that allows to monitor, in real time, and
detail, the state of each container, optimize routes, directly
infer the expenses of the fleet vehicle responsible to complete
the route. Ecogest [4], developed by Tecmic, uses sensors to
analyze the volume within the monitored container,
transmitting through GPSR the information required by the
fleet. The solution is also based on an online platform that
allows the containers maintenance and optimal route
planning.
From the state-of-the-art analysis, it was possible to notice
that the core functionalities of every solution are based on the
use of sensors to monitor waste and on a web-based platform
to generate optimized routes. However, none of the solutions
relied on the information provided by the citizens. One
possible, and simple approach would be to allow citizens to
provide feedback in regard to the implemented solution. For
example, the citizen may wish to request a container
placement or report an eventual problem or accident,
therefore, allowing the citizens to take part in the solution
may increase its chances of success.
IV. I
NTERNET OF
T
HINGS AND
M
ACHINE TO
M
ACHINE
C
OMMUNICATION
There are many different technologies that can be used in
IoT/M2M, in order to create a scenario of ubiquitous
communication. Some examples of usually employed
technologies are the Radiofrequency Identification (RFID)
[5], Near Field Communication (NFC) [6], Quick Response
codes (QR code) [7], Bluetooth/Bluetooth Low Energy
(BLE) and Wi-Fi. As for the concern for minimal energy
consumption, in 2003 IEEE developed a standard technology
to allow low power communication between devices, the
IEEE 802.15.4 standard [8]. This standard became a
reference to some proprietary low power communication
protocols, such as ZigBee [9]. However, as proprietary
solutions are used in the application layer, interoperability in
networks that communicate exclusively through these
protocols is affected. Solutions to this problem are based on
the use of specialized gateways, mapping between different
protocols. However, this is a costly solution due to its high
requirement of operations between the nodes, which impairs
network performance and energy efficiency.
Nonetheless, there are many possibilities at the
infrastructure level that can be applied to support IoT
networks. Generally, those infrastructures are the Low Power
Wide Area Networks (LPWANs), on which, low
consumption devices may communicate through the wireless
networks [10]. Each one of the available solutions has in
mind different business models, however, they all follow the
same purpose; provide services through networks of
sensors/smart objects that communicate through low power
technologies. For example, SigFox [11] is a French company
that aims to become a global operator in IoT networks,
possessing all the technology to connect IoT devices (from
backend to software and endpoints) within the same public
network. Despite being a low cost solution, its business
model originates a single point of failure to every user, since
everyone has to implement the whole network infrastructure
provided by SigFox. Meanwhile LoRa Alliance [12], relies
on Semtech to develop all the radio components of the
system, leaving the communication protocols between the
devices open to public usage. Therefore, it becomes possible
for any user to acquire endpoints that use LoRa’s
specifications in order to create public or private networks of
low consumption devices. Another possibility is the LTE-M
[13], which is set on the existing network infrastructure used
in LTE radio communications. The benefit of this solution is
the use of an already existent network infrastructure, that
operates on an licensed spectrum. However, the biggest
questions about the use of this infrastructure are related to its
universality, since at the moment there are not many brands
to produce and sell microchips that operate using this
technology.
V. R
OUTE
O
PTIMIZATION
A
LGORITHMS
Companies that have fleets of vehicles, such as those that
perform waste collection, have all the interest in minimizing
fuel and fleet maintenance costs, choosing routes that allow
them to perform their services at minimal expenses. Thus, the
upsurge of solutions to problems such as the Vehicular
Routing Problem (VRP), that proposes to find optimal routes
in regard of the environment and resources. The VRP is
originated from the Traveling Salesman Problem (TSP) [14],
where the goal is to find the shortest route between a group
of destination points, where each one is visited only once, and
the agent must return to its point of origin. Due to the high
complexity of solving the VRP, the techniques to do so are
countless. Thus, there are the exact algorithms, which seek
optimal solutions, but have a high computational cost, such
as the Branch and Bound and Branch and Cut algorithms; the
classic heuristics, which are computationally lighter than the
exact algorithms, and which are usually measured using
attributes such as precision, speed, flexibility and simplicity,
of which can be given the example of the Clarke and Wright,
Fisher and Jaikumar, Multi-Route Improvement, Sweep and
the Petal algorithm; and the metaheuristics, which may
include the Ant Colony System, Simulated Annealing, the
Genetic Algorithms, Tabu Search and Neural Networks [15].
Smart waste collection involves the issues addressed in the
VRPs, with some work and research having to be done in the
area in order to find an appropriate algorithm/mechanism to
generate optimal routes. In many cases, the solution
employed may grasp multiple search algorithms and
mechanisms.
VI. M
ANAGEMENT
S
YSTEM
P
ROTOTYPE
With the intention to facilitate the managers daily job, the
developed web-based management platform must be as easy
and simple to navigate and interact as possible. Therefore, the
developed prototype aims to satisfy those requirements, and
help to understand how a real-world application would work
and be managed.
The common user may wish to navigate through the
solution’s webpage (
Fig. 2
), however, the manager must know
his credentials in order to be granted access to the system’s
information.
Authorized licensed use limited to: b-on: Instituto Politecnico de Castelo Branco. Downloaded on January 19,2021 at 10:29:55 UTC from IEEE Xplore. Restrictions apply.

Fig. 2. Homepage
After logging in, the manager is faced with every
registered container, being capable to interact with each one,
analyzing its fill level (
Fig. 3
).
Fig. 3. Map
With the intention to simplify the analysis process, each
container is colored regarding its fill level, as described in
Error! Reference source not found..
TABLE I.
C
ONTAINER DESCRIPTION
This icon represents a container currently offline, that is,
its sensor is not capturing any data and it won’t be included in
future routes.
This icon represents a semi-empty container, one that its
fill level doesn’t surpass 45%.
This icon represents a container that its fill level is
between 45% and 80%.
This icon represents the almost full container, thereafter it
will be included in future waste collection routes.
The system’s core tool is to generate optimal routes,
including containers that are above its eighty percent of fill
level (
Fig. 4
). In addition, the same tool displays
information, such as the distance to complete the route and
time taken.
Fig. 4. Routing
C
ONCLUSION
As humanity evolves, the solutions to our problems tend
to follow along. This small initiation, to treat our waste with
the attention it deserves, may help us to prevent big
catastrophes in the future to come, especially in
underdeveloped counties, where pollution is a major concern
to its citizens wellbeing. This project not only aims to ease
this concern, but also strives for a cleaner and friendly city.
R
EFERENCES
[1] Ecubelabs, “ecubelabs,” Ecubelabs, [Online].
Available: https://www.ecubelabs.com/solution/.
[Accessed on 20 08 2020].
[2] Enevo, “enevo,” enevo, 2020. [Online]. Available:
https://www.enevo.com/waste-solutions-services.
[Accessed on 20 08 2020].
[3] Smartbin, “smartbin,” smartbin, [Online]. Available:
https://www.smartbin.com/solutions/smart-
monitoring/#. [Accessed on 20 08 2020].
[4] Tecmic, “waste-management.tecmic,” Tecmic, 2017.
[Online]. Available: https://waste-
management.tecmic.com/. [Accessed on 20 08 2020].
[5] C. T. P. Ltd, “clarinox,” 11 2009. [Online]. Available:
https://www.clarinox.com/docs/whitepapers/RealTim
e_main.pdf. [Accessed on 21 08 2020].
[6] C. Faulkner, “techradar,” 09 05 2017. [Online].
Available: https://www.techradar.com/news/what-is-
nfc. [Accessed on 21 08 2020].
[7] D. ADC, “nacs,” DENSO ADC, 2011. [Online].
Available:
http://www.nacs.org/LinkClick.aspx?fileticket=D1Fp
VAvvJuo%3D&tabid=1426&mid=4802. [Accessed
on 21 08 2020].
[8] D. Gascón, “web.archive,” 17 11 2008. [Online].
Available:
https://web.archive.org/web/20120319184855/http://s
ensor-networks.org/index.php?page=0823123150.
[Accessed on 22 08 2020].
Authorized licensed use limited to: b-on: Instituto Politecnico de Castelo Branco. Downloaded on January 19,2021 at 10:29:55 UTC from IEEE Xplore. Restrictions apply.

[9] Z. Alliance, “zigbeealliance,” Zigbee Alliance, 05 08
2015. [Online]. Available:
https://zigbeealliance.org/wp-
content/uploads/2019/11/docs-05-3474-21-0csg-
zigbee-specification.pdf. [Accessed on 22 08 2020].
[10] R. Sanchez-Iborra, “ncbi.nlm.nih,” 17 05 2016.
[Online]. Available:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4883
399/. [Accessed on 22 08 2020].
[11] L. Guzmán, “sigfox,” Sigfox, 2018. [Online].
Available: https://www.sigfox.com/en/news/sigfox-
presents-2017-results-and-2018-roadmap. [Accessed
on 22 08 2020].
[12] L. Alliance, “lora-alliance,” 2019. [Online].
Available: https://lora-
alliance.org/sites/default/files/2020-
02/LA_2019_AnnualReport_External_INTERACTI
VE.pdf. [Accessed on 22 08 2020].
[13] R. &. Schwarz, “rohde-schwarz,” Rohde & Schwarz
USA, Inc., [Online]. Available: https://www.rohde-
schwarz.com/us/solutions/test-and-
measurement/wireless-communication/iot-m2m/lte-
m/lte-m-theme_234034.html. [Accessed on 22 08
2020].
[14] G. B. D. a. J. H. Ramser,
“andresjaquep.files.wordpress,” 28 09 2008. [Online].
Available:
https://andresjaquep.files.wordpress.com/2008/10/26
27477-clasico-dantzig.pdf. [Accessed on 22 08 2020].
[15] M. D. L. M. G. W. J. G. Leonora Bianchi,
“doc.rero.ch,” 19 09 2008. [Online]. Available:
https://doc.rero.ch/record/319945/files/11047_2008_
Article_9098.pdf. [Accessed on 24 08 2020].
Authorized licensed use limited to: b-on: Instituto Politecnico de Castelo Branco. Downloaded on January 19,2021 at 10:29:55 UTC from IEEE Xplore. Restrictions apply.