EcoSort: Advancing Smart Bins Waste Segregation and Management with Image

Recognition and Incentive Rewards

A Capstone Project Presented to

The Undergraduate Program
College of Technology
Pamantasan ng Lungsod ng Maynila

In Partial Fulfillment of the Requirements for the Degree

Bachelor of Science in Information Technology

By

De Guzman, Julius Bryan B.

Engco, Barveily Joanabeth V.
Grefiel, Dharlene I.

Prof. Mark Anthony S. Mercado

July 2023
 Chapter One

INTRODUCTION

1.1 Background of the Study

The Philippines, named as the fourth largest waste contributor in Southeast Asia
and one of the world's top ocean polluters, is definitely facing a solid waste management
(SWM) problem. This is due to a number of factors, including increased garbage
production and gaps in waste collection, segregation, and recycling in many local
government units (LGU). (Rebuelta-Teh R., 2022). According to a 2020 World Wildlife
Fund assessment, 35% of total plastic trash produced in the Philippines is spilled into the
environment, 33% is collected and disposed of in sanitary landfills, and only 9% is
recycled. Improper trash disposal directly adds to global warming and climate change by
generating methane during waste breakdown and burning. Metro Manila generated up to
9,500 tons of garbage per day (tpd) in 2020, with a projected increase to 10,400 (tpd) in
2025, based on the data provided by the National Solid Waste Management Commission
(NSWMC). Waste collection in Metro Manila is done collaboratively by barangays and
LGUs. As required by RA 9003, barangays collect segregated garbage and transport it to
materials recovery facilities (MRFs) for recycling and composting of biodegradables.
Meanwhile, local governments are responsible for collecting residual waste. However,
due to the restricted resources of barangays and the small number of MRFs, this criterion
has not been achieved throughout the 17 LGUs of Metro Manila. (Reyes, M. 2023).
Furthermore, there is still a lack of awareness among households regarding the
importance of waste segregation, resulting in no changes in disposal practices.
(Philippine News Agency, 2022).

Waste segregation is the process of separating waste from other rubbish in order to reuse,
recycle, and minimize waste. (Kihila et al., 2021). Despite the fact that people have
always produced waste, waste management has become more complicated as industry
and technology have advanced. (Abdel-Shafy & Mansour, 2018). Mismanagement of
waste segregation at the local level contributes to natural resource depletion. Traditional

1
waste management approaches frequently fall short of adequately sorting recyclable
materials from general waste, resulting in increased landfill waste and missing recycling
and resource recovery opportunities. Not everyone has been properly educated about
waste management practices. Recycling is a key component of waste reduction, and it is
essential for environmental sustainability. The 3Rs of recycling - Reduce, Reuse, and
Recycle - are often forgotten, even by people who have been educated about them since
they were young. (Hin L., et al, 2021).

Waste containers currently in use are unable to distinguish between recyclable and
non-recyclable materials. Due to this, a sizable percentage of recyclables end up in
landfills. Citizens’ are deterred from actively adopting sustainable waste management
techniques by the lack of real-time monitoring and response. In recent years, the idea of
smart bins fitted with cutting-edge technologies has drawn a lot of interest. To enable
effective trash segregation and management, smart bins combine cutting-edge capabilities
like image recognition, data analytics, and IoT connectivity. (Bhutada, S., et al., 2018).
EcoSort, a ground-breaking trash management system that uses image recognition and
incentive rewards to encourage sustainable waste disposal practices, is one such solution.
With the help of image recognition technology, the study intends to create Smart waste
segregator bins that can accurately process classification of waste products through
computer vision using IoT. Real-time image processing and garbage classification as
recyclable, or non-recyclable will be possible with the Ecosort web system. Along with
encouraging people to actively participate in waste segregation and fostering a sense of
personal responsibility towards environmental sustainability, the study aims to develop a
module that will SMS the authorities when the bins are full as well as incentive rewards
coupons that can be exchanged for potential recycled materials.

2
1.2 Statement of the Problem

1.2.1 Inefficient Waste Segregation: Traditional waste containers lack the

ability to differentiate between recyclable,and non-recyclable waste, which leads to
resource depletion and environmental deterioration.
1.2.2 Limited real-time monitoring and feedback mechanisms: Waste
management officials struggle to identify overflowing bins or areas with high garbage
generation rates without real-time monitoring. This makes waste collection, recycling
planning and resource allocation difficult. Inadequate trash disposal is sometimes caused
by a lack of awareness.

1.2.3 Lack of knowledge, engagement and awareness of citizens: Current

waste management systems lack incentives for citizens to participate in ecologically
sustainable waste disposal. People often remain uninformed of recycling requirements
and proper waste segregation, which limits their involvement in recycling programs.

3
1.3 Objective of the Study

1.3.1 General Objective

The objective of this study is to develop a Smart Bin with an IoT computer
vision-based waste management system and incentive coupon rewards to promote waste
segregation and management. The objective is to assess the impact of EcoSort on waste
management efficiency, public engagement, and sustainable waste disposal practices,
revolutionizing waste management systems for a sustainable future.

1.3.2 Specific Objectives

1.3.2.1 Implement a Web-System with Image Processing through

Computer Vision for Waste Recognition using IoT: Ecosort System will utilize
deep learning algorithms (CNN) to distinguish between recyclable materials
(Plastics, Papers, Cans, Metal, Glass), ensuring real-time and proper waste
segregation at the point of disposal.

1.3.2.2 Establish an SMS Bin Fill Notification System: Allowing

real-time monitoring of smart bin fill levels using sensors or weight measuring
equipment. When a bin hits capacity, an automatic SMS is sent to waste
management staff, enhancing efficiency, reducing overflow, and allowing for
timely waste disposal.

1.3.2.3 Introduce Coupon Incentive Rewards Exchangeable for

Potential Recycled Materials: Allows citizens to exchange coupons for recycled
materials, fostering sustainable practices and reinforcing environmental
responsibility for long-term behavior change.

4
1.4 Significance of the Study

This study aims to advance waste segregation and management practices,

promote sustainable behaviors, encourage public participation, leverage technology,
contribute to a positive environmental impact, and beneficial specifically to as follows:

Citizens: This study aims to promote sustainable waste management practices and
encourage active participation. Image recognition technology guarantees proper waste
segregation, contributing to a cleaner and healthier living environment. The incentive
rewards program encourages residents to engage in appropriate garbage disposal,
promoting a sense of environmental responsibility. This study serves citizens by
providing a user-friendly and effective waste management solution.

Local Government Unit (LGU): This study aims to provide an advanced approach to
waste segregation and management. By incorporating image recognition and incentive
rewards, it enhances the efficiency and effectiveness of waste management practices.
This can assist LGUs in achieving their waste management goals, improving
environmental sustainability, and promoting a cleaner and healthier community.

Materials Recovery Facilities (MRFs): This study intends to help MRFs, which play
an important role in waste management. The system's image recognition technology
enables accurate waste categorization, which can help MRFs streamline sorting
processes. The incentive rewards scheme also promotes proper waste disposal, which
results in higher-quality recyclable materials at MRFs. As a result, their operational
efficiency and the whole recycling process improve.

National Solid Waste Management Commission (NSWMC): This study intends to

correspond with the NSWMC's goals and objectives for increasing waste management
methods. By proposing a unique and practical strategy to waste segregation and
management through incorporating image recognition and incentive rewards. The ability
to increase waste management efficiency, encourage sustainable habits, and minimize
environmental impact makes it critical for the NSWMC in meeting its waste
management targets.

5
Future Researchers: This study aims to provide insights into the use of image
recognition and incentive incentives in waste segregation and management. The
research findings, methods, and lessons learned can be used to inform future studies on
waste management, technological improvements, and the efficacy of incentive-based
systems. This can help to progress waste management methods and the creation of new
inventive solutions in the industry.

1.5 Scope and Limitations

1.5.1 Scope

The general intent of this capstone which is entitled “EcoSort: Advancing Smart
Bins Waste Segregation and Management with Image Recognition and Incentive
Rewards” encompasses the development, implementation, and evaluation of the EcoSort
system within the context of waste segregation and management. The focus is on the
utilization of image recognition technology in smart bins to accurately identify and
categorize different types of waste materials, promoting proper waste segregation at the
point of disposal. The scope also includes the integration of an incentive rewards
program to encourage active participation and sustainable waste management behaviors
among users. The study may involve the design and testing of the EcoSort system, as
well as the collection and analysis of data related to waste segregation, user
engagement, and the effectiveness of the incentive rewards program.

1.5.2 Limitations

While the study aims to advance waste segregation and management practices,
there are certain delimitations to consider:

Geographical Focus: The scope of the study is restricted to a certain geographic region
or environment, such as a city, community, or educational institution, such as the city of
Manila. The findings and recommendations may be context-specific and not
immediately applicable to other regions without taking into account their distinct waste
management systems and dynamics.

6
Technical Constraints: The study focuses on the utilization of image recognition
technology within smart bins. It may not cover other technological aspects or methods
of waste segregation and management. The scope does not extend to the development or
implementation of physical smart bin infrastructure, as it primarily centers on the image
recognition and incentive rewards components.

Legal and Regulatory Factors: The study operates within the legal and regulatory
frameworks of the relevant jurisdiction. It takes into account any local regulations,
policies, or guidelines related to waste management and data privacy.

Stakeholder Limitations: The study's scope focuses primarily on the perspectives and
experiences of users, such as individuals or institutions implementing the EcoSort
system. It may not encompass an exhaustive analysis of all stakeholder perspectives,
such as waste management authorities, recycling facilities, or local government units.

7
 Chapter Two

REVIEW OF RELATED LITERATURE

2.1 Related Literature

2.1.1 Related Foreign Literature

The Importance of Waste Segregation

According to Davies' article from 2023, the article emphasizes the significance of
waste segregation for efficient waste management. It underlines the significance of
classifying garbage according to its composition and features. The article illustrates the
several advantages of waste segregation, including enhanced health, sustainability, and
environmental protection. It promotes waste segregation as a strategy for people,
organizations, and communities to help create a cleaner, more sustainable future.

Smart self-power generating garbage management system using deep learning for
smart cities

According to a recent journal conducted by Thamarai and Naresh (2023), an

innovative waste management system for smart cities is proposed, combining
Convolutional Neural Network (CNN) and Internet of Things (IoT) technologies. The
system is designed to autonomously handle household waste, utilizing a CNN to
categorize it into organic and inorganic components. The organic waste is further utilized
to generate electricity, while the inorganic waste is sent for recycling. The system
comprises several components, including a garbage collector, a waste segregation unit, a
power generation unit, an IoT sensor-equipped bin for inorganic waste collection, and an
electronic control unit. The suggested CNN model performs satisfactorily, achieving a
waste segregation accuracy of 98%, surpassing other pre-trained models. The system is
highly adaptable for managing municipal waste in smart cities and contributes to

8
electricity production through the utilization of biogas derived from decomposing organic
waste.

An Internet of Things Waste Collection System

According to Elechi, P., et al. (2022), the ultrasonic sensor plays a vital role in the
automatic waste system by sensing waste objects and providing feedback to the
IoT-based automatic waste controller. The sensor operates using a reflective method,
where transmitted ultrasound interacts with an object and reflects back to the sensor’s
receiver. The research focused on accurately measuring ultrasonic wave time and distance
to ensure the system’s efficiency. The system effectively detects waste objects at a
distance of 30 cm, aligning with predefined design parameters.

Intelligent waste management system using deep learning with IoT

According to Rahman, M.W., et al (2022), the integration of deep learning and

IoT technology in waste management offers an efficient solution for waste classification
and real-time data monitoring. The proposed architecture combines a convolutional
neural network (CNN) with IoT technology to enable intelligent sorting of digestible and
indigestible waste. A smart trash bin design incorporating a microcontroller and multiple
sensors is introduced, with IoT connectivity allowing for real-time data control and
Bluetooth facilitating short-range data monitoring through an Android application. The
model’s effectiveness is evaluated based on waste label classification accuracy, sensor
data estimation, and the system usability scale (SUS), achieving a classification accuracy
of 95.3125% using the CNN model and receiving a SUS score of 86%. This adaptable
smart system holds potential for real-time waste monitoring in household activities.

Waste Classification for Sustainable Development Using Image Recognition with

Deep Learning Neural Network Models

Based on the journal by Malik, M., et al (2022), an architecture for classifying

litter into specific categories using the EfficientNet-B0 model is presented. The
EfficientNet-B0 model, pretrained on ImageNet, offers high accuracy levels ranging from

9
74% to 84% in top-1 over ImageNet. The study proposes tuning the model for images
specific to different demographic regions, enabling efficient classification through
transfer learning. This allows for the creation of a customized model optimized for a
particular region, demonstrating comparable accuracy to the larger EfficientNet-B3
model while utilizing significantly fewer parameters. As a result, the proposed technique
achieves a 4X efficiency improvement in terms of FLOPS (floating-point operations per
second), and fine-tuning the model using region-specific litter images enhances
classification performance.

Smart Segregation Bins for Cities Using Internet of Things (IoT)

Based on the article by Srisabarimani, K., et al (2022), the operation is divided

into two parts: trash segregation and real-time monitoring of the dustbin using IoT. Trash
segregation is carried out with the aid of sensors, including an IR sensor, an ultrasonic
sensor, and a capacitive sensor, to detect the existence of waste, gauge its volume, and
determine the presence of plastics. The segregation process is carried out using
components such as an air blower, a microprocessor, a rotating disk, and a servo motor.
Dustbin data is stored on a cloud server system and can be viewed through a Wi-Fi
module. Waste collection from the garbage bin is done manually to avoid interfering with
the work of everyday laborers and potentially leading to unemployment. This concept
aims to benefit society by separating waste into plastic and reusable wastes.

Smart bin: Waste segregation system using deep learning-Internet of Things for
sustainable smart cities

According to Mohammed, A., et al. (2022), existing smart bins that use IoT and
wireless sensor networks (WSN) are heavily reliant on numerous types of sensors and a
console for connectivity, which is dependent on programming and operating systems. The
limitations of embedded smart bins are overcome by merging IoT with artificial
intelligence technologies such as deep neural network (DNN) systems. The article
demonstrates a Friendly Waste Segregator using Deep Learning and IoT to categorize and
separate waste objects as biodegradable and nonbiodegradable. The proposed solution

10
employs a powerful deep learning network to accurately categorize garbage and IoT for
monitoring and connectivity via various sensors. With initial training, the technique can
recognize and separate real-time garbage objects with an average accuracy of 97.49%.
The smart bin aims to provide optimized waste management of bio and non-bio waste
while contributing to the creation of an environmentally safe society.

Smart Prediction and Monitoring of Waste Disposal System Using IoT and Cloud
for IoT Based Smart Cities

According to John, J., et al. (2022), an IoT-based smart garbage disposal solution
for smart cities is proposed. The system measures fill levels in bins of any size using
off-the-shelf components, with an Arduino microcontroller communicating with IR, UV,
and weight sensors, as well as a GPS module, to monitor bin status. Data is sent over a
cluster network to a master module linked to the backend through Wi-Fi. An intelligent
LSTM algorithm anticipates waste generation trends and Firebase Cloud Messaging
notifies when bins are full. The system also includes automated bin height calibration, a
dynamic web data display, and a distributed real-time firebase database.

IOT Based Waste Management System Using Node MCU

Based on the journal by K, V. S. R. (2022), a novel IoT-based waste management

system is presented to address the issue of garbage segregation and monitoring in Indian
cities. The current system is inefficient, resulting in expensive transportation and
collection costs and overflowing rubbish bins. The proposed method employs a NodeMcu
microcontroller, an ultrasonic sensor to detect garbage levels, and a moisture sensor to
separate dry and wet waste, programmed using the Arduino IDE and the C-Language.
This intelligent garbage monitoring system provides an effective option for improving
waste management and segregation.

Smart Dustbins - Automatic Segregation & Efficient Solid Waste Management using
IoT Solutions for Smart Cities

11
 According to Karwar, Mr. V. H. (2022), an IoT-based smart dustbin monitoring
system for solid waste management in smart cities is proposed. The system uses dynamic
scheduling and routing of trash collectors, moisture detection for automatic wet and dry
waste segregation, and real-time status updates to the cloud. A mobile application assists
trash collectors with optimal routes, and municipal corporations can monitor the waste
collection process. The proposed system aims to improve sustainability by efficiently
using technology to address global environmental problems.

Smart Waste Segregation System using Convolutional Neural Networks

Based on the journal by Shenoy, A., et al. (2022), a smart trash segregation
system based on CNN is presented to address waste management difficulties in India.
The system identifies waste using OpenCV and a Pi camera and separates it with a
robotic arm. An ultrasonic sensor checks waste levels when the bin is full and warns an
authorized individual via a GSM module. This device improves upon existing systems
that use moisture sensors, which can mistake dry or metal waste for wet due to
microscopic water droplets, and contributes to the Swachh Bharat project.

Garbage Management System using Internet of Things

According to Meet, P., & Patel, K. D. (2022), the relevance of smart waste
management in modern cities is discussed, which face issues related to smart grids,
environments, living, and transportation networks. Traditional trash management
procedures are insufficient, and smart waste management is a primary issue for cities due
to its impact on people’s lives, health, and the environment. The paper examines different
potential approaches for smart bin systems, such as smart garbage collection, monitoring,
and segregation, and proposes a framework for a smart waste management system. The
rubbish Monitoring System aims to improve society cleanliness by preventing incorrect
rubbish management.

An Intelligent Smart Bin for Waste Management

12
 According to Hin, L.C., et al. (2021), the integration of Intelligent Systems into
the smart bin enables automatic classification and sorting of various waste categories. By
leveraging Internet of Things (IoT) technology, the smart bin efficiently communicates
with devices, transmitting data to a monitoring computer for analysis, data collection, and
information dissemination. The incorporation of machine learning continuously enhances
the performance of the smart bin, while the implementation of robotics facilitates
automated sorting and enables long-distance notifications to users. The proposed system
represents a significant contribution to green technology by reducing human effort,
enhancing waste management efficiency, promoting recycling, and ultimately creating a
cleaner living environment.

AUTOMATED WASTE SEGREGATOR

According to this article by Gimonkar, R. (2021), the development of an

automatic trash segregation system for homes and small communities is the main topic.
The system offers a straightforward and efficient solution by dividing waste into three
categories—metal waste, solid waste, and wet waste. The segregation method makes use
of an Arduino Uno, small mechanical arms, and electronic sensors for waste detection.
The purpose of this study is to support community recycling efforts and trash
management efforts.

A CNN-Based Smart Waste Management System Using TensorFlow Lite and

LoRa-GPS Shield in Internet of Things Environment

Based on the journal by Sallang, N. C. A., et al. (2021), the goal is to create a
smart waste management system in an IoT environment that uses a deep learning model
to improve trash segregation and monitor bin status. The SSD MobileNetV2 Quantized
model has been trained to classify waste and has been integrated with a Raspberry Pi 4
and a camera module. A servo motor sorts waste, an ultrasonic sensor measures fill
percentage, and a GPS module pinpoints the location. Bin status is sent through a LoRa
module, and the components are safeguarded by an RFID-based locker.

13
Proceedings of the 2021 4th International Conference on Computing and
Communications Technologies, ICCCT 2021

According to the Proceedings of the 2021 4th International Conference on

Computing and Communications Technologies , among the subjects covered are
character recognition in Tamil using convolutional neural network (CNN) analysis, face
recognition and mask detection technology for surveillance systems, share market
prediction using long short-term memory and artificial neural network, convolutional
recommended neural network system based on user reviews for movies, automated
data-driven preprocessing and training of classification models, reward-based smart bin,
and more.

Detection of Waste Materials Using Deep Learning and Image Processing

Based on the journal by Mitra, A. (2020), an automated waste detection

framework that utilizes deep learning algorithms and image processing techniques was
introduced. The framework was implemented by working with a large dataset of images
and training algorithms to identify patterns for object detection and classification. Waste
materials can be classified into six categories (Cardboard, Metal, Glass, Paper, Plastic,
Trash) in multiple objects within a single image using the Faster R-CNN algorithm,
maintaining a high level of accuracy. The detection of waste materials is not limited to
images alone, as the framework can also detect and classify waste materials from video
feeds or real-time webcam streams.

Arduino Based Garbage Monitoring System using IoT

According to Sinha, A., et al. (2020), the increasing population and waste
generated by hotels, industries, and other sources have led to the issue of overflowing
dustbins. To resolve the problem, a smart system utilizing IoT technology to connect
multiple dustbins located across various areas of the city was introduced. Each dustbin is
equipped with affordable embedded devices capable of monitoring and sensing its fill
level using ultrasonic sensors. The gathered information is transmitted to the municipality
officer, who can promptly alert the assigned truck driver for waste collection.
14
Waste Segregation Using Smart Bin and Optimization of Collection Routes

According to Gounder, J. P., et al. (2020), the system is designed to detect and
segregate various types of waste at the user end and notify the authorities when the waste
bins reach their capacity. Users are provided with appropriate dumping technology to
ensure proper disposal into the designated bins, and the system assists users by guiding
them to the nearest bin in case they encounter any issues with the bins.

Waste Segregation and Waste Management Using Smart Bin

According to this book by Verma, S., et al. (2020), an in-depth and

interdisciplinary assessment of the most recent research in sensor and detection
technologies is provided, with an emphasis on new materials, methodologies,
architectures, and properties. The desire to build sensors at both the macroscopic and
nanometric scales has fueled research into sensing materials and technologies for reliable
detection in varied phases and combinations. The introduction of new materials and
processes, such as spin across organic, inorganic, and composite materials, is critical for
reduced-scale detection approaches. The book’s 15 chapters investigate the impact of
material properties on sensing capabilities, as well as the significance of intermolecular
interactions, polymer formation, and behavior under diverse situations in the processing
of ultra-sensing materials. The content is organized into three main sections: principals
and perspectives, new materials and methods, and advanced structures and properties for
sensor devices.

Smart Waste Bin

According to Kulkarni, Dr. N., et al. (2020), a smart trash can for use in homes is
suggested to enhance waste management and segregation. The bin makes use of an
Arduino UNO, Arduino Nano, servo motors, and batteries to distinguish between wet and
dry garbage. Trash separation makes waste management easier, safeguards the
environment, recognizes the value of waste economically, and lessens workers’ health
issues related to trash separation. In India, there is no other technology that separates dry

15
and wet wastes in a single household bin at such a low cost and ease of use. The gathered
garbage can be processed right away.

Smart Bin with Automated Metal Segregation and Optimal Distribution of the Bins

According to Saranya, K. C., et al. (2020), the absence of containers or other

disposal facilities continues to be a problem, and several governments are making a
concerted effort to promote a cleaner nation by assisting in the reduction of pollution and
enhancing public sanitation. A crucial and major application for providing cleanliness is
the smart trash can, modeled as an automatic metal and non-metal segregator for a smart
bin. The level of the bin is continuously tracked and uploaded to the cloud for additional
analysis using NodeMCU. A novel mathematical approach has also been used to
determine the most effective manner to distribute the bins in one dimension. The network
of trash cans can be used in tourist destinations, parks, schools, and other public spaces to
promote sanitation.

Smart Waste Segregation using ML Techniques

According to Patel, Mr. D. N., et al. (2020), an approach integrating IoT and ML
is suggested for lowering landfill waste and greenhouse gas emissions in India. The goal
is to develop a smart trash can for home usage that uses sensors and machine learning
models to distinguish between biodegradable and non-biodegradable waste, separating
recycling and landfill waste from the non-biodegradable waste. Compared to large-scale
segregation, this early segregation at the home level is easier to control and more precise.
The smart bin attempts to offer an environmentally friendly approach to waste
management.

Potential integrated smart waste segregation for all Stakeholders

According to Sadri, B., et al. (2020), the use of smart technologies to improve
trash management and recycling in cities is discussed. Waste separation is handled by
professional and informal recycling personnel, but their operations may result in
contamination and debris, reflecting a lack of citizen participation in garbage separation.

16
The suggested smart waste recycling model employs service design approaches
consistent with local infrastructure and incorporates smart identification of all
stakeholders. The system aims to enhance waste management in cities facing similar
difficulties by enacting legislation, providing monitoring and measurement tools, and
using 4R procedures (reduction, reuse, recycling, and recovery).

Smart Waste Management System Using Image Processing

According to Sridevi, A., et al. (2020), a smart waste classification system based
on IoT and an AI-based deep learning model is proposed to improve garbage
management in large cities. The technology replaces the traditional trash management
system with sensors and artificial intelligence to improve garbage classification. Bin
status is monitored by hardware sensors, and an image processing software written in
Python performs object detection and classification using a pre-trained object detection
model. Each bin’s garbage filling level is monitored by an ultrasonic sensor, and the
system tries to get residents’ support for household garbage pickup.

IoT Based Automated Waste Segregator for Efficient Recycling

According to Saminathan, T., et al. (2019), the journal’s findings reveal that the
Smart bin prototype municipal trash segregator can immediately sort dumped garbage,
resulting in more recyclable paper. The system sends information about the deposited
rubbish and takes appropriate action by utilizing several sensors and motors interfaced
with an Arduino board. The simulation of practical conditions reveals how the bin works
during the segregation process.

Development of smart garbage bins for automated segregation of waste with real
time monitoring using IoT

According to Jain, R., et al. (2019), a real-time garbage collection and

management system based on smart bins is proposed, in which waste is first separated
and collected in dedicated bins that are monitored in real-time on the cloud. When the

17
garbage cans are full, an email notification is received as an alert, delivered to an
authorized individual. If a fire is detected, an email notice is also issued.

Design of IoT based Garbage Segregation for Automatic Smart Trash Bin using NI
LabVIEW

In this journal by Devi, K., et al. (2019), a Smart Trash Bin is proposed that
segregates trash into degradable and non-degradable wastes with the help of sensors and
motors interfaced with NI myRIO. Rapid advances in technology, low initial cost, and
intentional obsolescence have resulted in a rapidly expanding surplus of wastes all over
the world. People throw trash on the side of the road, which is not picked up on a regular
basis by those in charge. It is proposed that wastes be separated into categories such as
biodegradable and non-biodegradable in an effectively closed manner. The entire process
of waste input, level, and segregation is monitored and controlled by the reconfigurable
IO using sensors, thereby reducing open decomposition of organic wastes and limiting
the growth of microorganisms.

Smart Waste Management using IOT

According to Bhutada, S., et al (2019), the Smart Waste Management system is an

innovative solution for keeping cities clean. It monitors garbage bins and provides
real-time information about the amount of garbage collected through a desktop
application. Ultrasonic sensors are placed under the bin lids to detect the garbage level,
which is then compared to the bin’s depth. The system is controlled by an Arduino
microcontroller and data is transmitted to the server using a Wi-Fi modem. Each bin has
an embedded device that tracks the garbage level and has a unique ID for easy
identification. When the garbage level reaches a certain limit, the device sends the level
and ID to the server. Authorities can access this information remotely and take immediate
action to clean the affected bins.

Smart Waste Management System using IoT

18
 Prof. S.A. Mahajan, et al. (2017) proposed a system where public trash cans are
equipped with a device that can monitor waste levels in real-time. This data is used to
create the most efficient route for garbage collection trucks, reducing fuel costs. Load
sensors improve the accuracy of the waste level data, while moisture sensors provide
information on waste segregation in the bin. By analyzing the data collected through
various system-generated reports, municipalities and government agencies can improve
their smart waste management plans.

2.1.2 Related Local Literature

Manila gov’t reminds public on proper waste management

Calucin, D.I. (2023) highlights the importance of proper waste management in

Manila and the government’s efforts to involve the public in keeping the city clean and
sustainable. The goal is to promote responsible waste disposal and create a cleaner, more
environmentally friendly city for all residents.

Worsening garbage problem

According to Reyes, M.A. (2023), the Philippines faces challenges in managing

solid waste, including issues with garbage segregation and collection at the municipal
level. A World Bank study found that Metro Manila has lower waste collection rates
compared to other cities, due to limited resources and coordination. An efficient
collection system by both barangays and LGUs is necessary for effective waste
management. Addressing these issues is crucial for resolving the solid waste disposal
problem and related concerns.

GULP: Solar-Powered Smart Garbage Segregation Bins with SMS Notification and
Machine Learning Image Processing

In their study, Sigongan, J., et al. (2023) successfully designed and implemented a
smart bin that automates solid waste disposal in a specific institution. This bin has
advanced features such as monitoring fill levels and preventing overflow by sending SMS

19
notifications. Additionally, the accompanying application can function independently by
using solar energy.

No Time To Waste: A collaboration

According to Rebuelta-Teh, A. (2022), the Philippines is facing challenges in

managing solid waste, which contributes to ocean pollution and climate change. Efforts
to address this issue include closing dump sites, establishing sanitary landfills, and
building materials recovery facilities. The No Time To Trash initiative, in partnership
with Nestlé Philippines, aims to educate LGUs on efficient waste management through
webinars and modules. Collaboration between the public and private sectors is crucial for
achieving a waste-free future and reducing environmental concerns.

Philippine solid waste segregation/disposal as a community response to the impacts

of climate change.

Singson, A. (2022) published a journal on solid waste management in the

Philippines as a way to reduce the negative impact of climate change. The study
highlights the importance of waste segregation, which helps with recycling and prevents
all garbage from being disposed of together. Proper waste segregation and disposal are
crucial for protecting the environment, as improper disposal can lead to pollution and
contamination of land, water, and air.

Solid waste segregation remains major challenge in PH: DENR chief

According to Sampulna, J. (2022), solid waste management is a major issue in the

Philippines due to poor waste segregation at the local level. Strict enforcement of waste
management laws by local government units (LGUs) is crucial. The Department of
Environment and Natural Resources (DENR) has closed all open dumpsites and requires
LGUs to establish sanitary landfills. However, there is still a lack of understanding about
waste segregation among households. It is important to properly separate biodegradable,
recyclable, residual, household hazardous, and household healthcare wastes. The public is

20
encouraged to use color-coded bins or labels. The DENR offers the Basura Buster app to
help young people learn about waste management.

Concealed Automated Trash Bin with Shredder for Solid Wastes

According to Villame, E., et al. (2022), improper waste management practices are
a common problem in densely populated developing countries like the Philippines. This
includes the incorrect use of waste bins and poor waste segregation, often due to people’s
irresponsibility. A proposed solution is the development of an automated segregation
system. The waste bins are designed to be concealed to save space, slow down
decomposition, and reduce odor. The system is fully automated to minimize direct
contact with waste and uses a gripper, servo motors, ultrasonic, capacitive, and
photoelectric sensors to collect and segregate waste. To save power, segregated waste is
stored in a bin before being shredded and sent to transport bins for collection. Ultrasonic
sensors provide data on the transport bins’ capacity and a GSM module sends an SMS to
the concerned authority about their status, facilitating easier waste collection. Tests on the
prototype’s response time and garbage level detection showed that it performed well and
achieved its intended function.

Self-Monitoring and Automated Real-Time Trash Bin (SMART Bin) with

Solar-Powered Integration for COVID-19 Infectious Waste Prevention

According to Simpal, M., et al. (2022), this research study developed a

Solar-Powered Self-Monitoring and Automated Real-Time Trash Bin (SMART Bin) to
improve waste management and reduce the spread of COVID-19 infectious waste. The
system uses ten hardware components, including an Arduino Uno R3, Ultrasonic Sensor,
Servo Motor, Solar Panel, and others. The bin’s lid automatically opens when a user’s
hand approaches within the acceptable range, allowing for touchless waste disposal. LED
indicators, a buzzer, and an LCD display provide real-time monitoring and notifications
of the bin’s status. The experiment successfully demonstrated the SMART Bin’s
functionality and effectiveness in reducing COVID-19 waste contamination.

21
Automatic Segregation of Limited Wastes through Tiny YOLOv3 Algorithm

In this journal, Macasaet, N. A. G., et al. (2022) developed a device that uses
small YOLOv3 to segregate restricted wastes. By using a pre-trained Tiny YOLOv3
model on a Raspberry Pi 4B, they were able to detect, classify, and separate limited
biodegradable wastes from non-biodegradable wastes. Based on the data collected, the
system achieved macro-averaged precision, recall, and F1 score values of 100%, 88%,
and 93%, respectively, for detection and classification. It also had an overall detection
and classification accuracy of 88% and a waste segregation accuracy of 100%.

Automatic Segregation of Limited Wastes through Tiny YOLOv3 Algorithm

In this journal, Macasaet, N. A. G., et al. (2022) developed a device that uses
small YOLOv3 to segregate restricted wastes. By using a pre-trained Tiny YOLOv3
model on a Raspberry Pi 4B, they were able to detect, classify, and separate limited
biodegradable wastes from non-biodegradable wastes. Based on the data collected, the
system achieved macro-averaged precision, recall, and F1 score values of 100%, 88%,
and 93%, respectively, for detection and classification. It also had an overall detection
and classification accuracy of 88% and a waste segregation accuracy of 100%.

Solid Waste Management Awareness and Practices among Senior High School
Students in a State College in Zamboanga City, Philippines

Alejandro, R., et al. (2021) published a journal on solid waste management and its
challenges in terms of health, environment, and socio-economic factors. The study
highlights the importance of integrating solid waste management concepts into science
education, as mandated by R.A. No. 9003, with educational institutions seen as agents of
change. A descriptive-quantitative approach was used, employing a researcher-made
instrument called the Solid Waste Management Awareness and Practices Questionnaire
(SWMAPQ). The study involved 332 Grade 12 students from a State College, divided
into STEM, TVL, and GAS streams. The results showed that students had sufficient
knowledge about solid waste definition, the effects of improper disposal, prohibited
activities, school initiatives, the importance of solid waste management, and their
22
responsibilities. However, their knowledge of relevant laws was relatively low.
Television/radio, parents, and social media were identified as key sources of awareness.
The study also found that students demonstrated good solid waste management practices
in terms of segregation, reduction, reuse, recycling, and disposal.

A Long-Standing Problem: A Review on the Solid Waste Management in the

Philippines

According to the journal by Coracero, E.E., et al. (2021), the goal was to identify
methods and procedures for addressing solid waste management issues in the Philippines.
Specifically, the study aimed to identify the various challenges of solid waste
management, evaluate the solutions currently in use, and suggest additional potential
solutions to waste management concerns. The purpose of the study was to assess the
current state of solid waste management in the Philippines, identify difficulties and gaps
in the management system, evaluate the current solutions and practices, and recommend
alternative solutions to improve how the problems are addressed.

E-Recycle Bin: An IoT Based Model for Solid Waste Management

According to the published study by de Leon, M.G., et al. (2021), the e-Recycle
Bin, a Smart Bin Prototype, was developed using IoT technologies and a BBMRF. The
bin generates receipts based on the amount of recyclable waste deposited and uploads
data to the cloud for monitoring. IT professionals evaluated the project and found that it
complied with ISO 25010:2015 to a Very High Extent. The system aims to reduce solid
waste issues and promote recycling in the City of Ilagan.

Design and Implementation of Automated Waste Segregator with Smart

Compression

According to the published study by Endaya, J.P., et al. (2020), the goal was to
design and implement an efficient system for properly segregating waste. Two
interconnected Arduino Unos and other components were used to achieve the desired
output. Push buttons were used to determine the type of waste being disposed of, and a

23
Stepper Motor rotated the trash bins attached to a circular metal plate. An Ultrasonic
sensor determined if paper or plastic bins needed to be compressed, which was done
using a Linear Actuator to maximize space in the trash bin. LEDs served as indicators
when the bins were full.

COMMUNITY-BASED WASTE MANAGEMENT

According to the publication by Cities, P. S. (2020), in low- and middle-income

countries where regular garbage collection is not available, people may resort to dumping
or burning their waste nearby. This is harmful to the environment, public health, and the
local economy. In situations where municipal waste management services are lacking or
inadequate, community-based waste management programs offer an affordable, efficient,
and engaging solution. These systems involve direct community participation and
accountability, tracking waste collection and reduction, segregating waste at the
household level, using valuable and reusable items on-site, composting biodegradable
organic waste, and providing a public collection service that transports waste to a
treatment facility.

Development of Waste Management System using the Concept of ‘Basura

Advantage Points’ through Artificial Neural Network

In this journal, Castro, R.C., et al. (2020) developed an automatic separating

machine using an Artificial Neural Network (ANN) as a machine learning method and
incorporating the concept of “Basura Advantage Points.” The ANN serves as the
machine’s brain, classifying plastic bottles into one category and other waste products
into another. The “Basura Advantage Points” concept allows people to earn points for
disposing of waste in the segregating machine, which can be redeemed for rewards
specified by policymakers. Survey results showed that the machine was popular with the
general public, with an accuracy of approximately 80% based on the sample trash. The
study identified a successful approach to reducing improper waste disposal and
encouraging people to participate in effective waste segregation, based on positive
feedback and successful evaluation.

24
Waste segregation schemes launched in BASECO

In the article by Daroy, P. (2019), the importance of waste segregation in reducing

environmental pollution and improving community well-being is emphasized. By
educating residents about the importance of separating waste into categories such as
recyclables, organic waste, and non-recyclables, the article aims to promote effective
waste management and encourage sustainable practices. The positive impact of these
waste segregation schemes on the BASECO community is highlighted, including a
cleaner environment, reduced health risks, and the potential for generating income
through recycling initiatives.

INTENSIFIED ENVIRONMENTAL PROTECTION: SOLID WASTE

MANAGEMENT

Miguel, M. (2019) discusses the Ecological Solid Waste Management Act (RA
9003) in the Philippines, which aims to improve waste management by implementing a
comprehensive program. The act helps local governments (LGUs) develop waste
management strategies, close dump sites, establish Materials Recovery Facilities (MRFs),
and promote environmentally sound disposal methods. It also provides recommendations
and technical assistance for Waste-To-Energy plants, promoting economic development,
job creation, reduced dependence on imported oil, and increased electricity generation.

Solid waste mismanagement in the Philippines

According to Mawis, S.M. (2019), the Philippines is facing challenges in

managing solid waste due to population growth and urbanization. Despite the existence
of waste management regulations such as RA No. 9003, local government units (LGUs)
struggle to comply, leading to issues such as open dumps and a lack of landfill space. To
address these challenges, individual responsibility and community participation are
crucial in promoting proper waste management practices. Collaboration between local
governments and the public is necessary to ensure effective garbage disposal methods.

25
Development of a cost-efficient waste bin management system with mobile
monitoring and tracking

According to Africa, A.D. (2019), the Internet of Things (IoT) is an emerging

technology that can provide solutions to make daily life easier, including in the area of
sanitation and waste collection through the development of smart waste management
systems. The study aimed to create an affordable waste management system and a mobile
application to track and monitor garbage levels. By using low-cost sensors and
microprocessors, the team was able to develop a cost-effective system. Additionally, they
created a mobile application with a graphical user interface (GUI) that displays waste
levels and sends notifications when the trash reaches a certain level, using firebase
database and android studio.

Resource Recovery through RDF: Current Trends in Solid Waste Management in

the Philippines

According to the findings of the journal by Sapuay, G. (2017), Refuse-Derived

Fuel (RDF) has the lowest global warming potential (GWP), while incineration has the
highest and landfill falls in between. Landfill is the worst water polluter, while
incineration poses the greatest hazard. Recycling and RDF consume the most truck fuel.
The paper concludes that waste prevention is the best solution. The study also discusses
the current state of solid waste management in the Philippines, the potential of RDF as a
resource recovery option, challenges in implementing the Ecological Solid Waste
Management Act, and best practices in RDF-based solid waste management.

26
2.2 Related Studies
2.2.1 Related Foreign Studies

A Vision Based Approach to Localize Waste Objects and Geometric Features

Exaction for Robotic Manipulation

According to the study by Aarthi and Rishma (2023), a method for real-time
automatic detection and segregation of waste materials has been developed to facilitate
easier recycling. The approach uses the Mask RCNN deep learning architecture, known
for its high performance, to locate and classify waste objects in natural environments.
Geometric features such as centroid, orientation, and clamping points of the objects are
also extracted to assist a robotic arm in grasping the waste objects effectively. The study
explores methods for ordering the objects in the scene to optimize energy efficiency in
the automated system. This technique has potential for implementation in robotic
machines used in waste management, allowing for efficient waste sorting on road
pavements and streets. The technology offers enhanced safety, productivity, and risk
reduction for sanitation workers involved in waste management processes.

Evaluation of practical edge computing CNN-based solutions for intelligent

recycling bins

According to the study by Li and Grammenos (2023), a cutting-edge smart

recycling bin has been proposed that automatically sorts urban rubbish to increase
recycling rates. Neural network classification models were trained using more than 1800
photos of recycling rubbish and an existing public dataset for two embedded systems, one
containing a Jetson Nano and the other a K210 unit. The models achieved an accuracy of
94.61% on the K210 and 93.99% on the Jetson Nano. A user interface application was
also developed to gather user feedback. The system using the K210 only required 0.89 W,
while the system using the Jetson Nano used 4.7 W, a 30% decrease from previous
research. The study demonstrates a low-cost, highly accurate, and energy-efficient smart
recycling bin with potential for commercialization to improve urban waste recycling.

27
Smart Waste Management and Classification Systems Using Cutting Edge
Approach

According to the study by Cheema et al. (2022), a real-time smart waste

management and classification mechanism called SWMACM-CA has been proposed.
The current lack of a proper waste-segregating mechanism results in a low recycling rate,
despite the potential to recycle 75% of waste in America. To address this issue, the
proposed system uses IoT, deep learning, and cutting-edge techniques. It includes a waste
grid segmentation mechanism that maps waste yards into grid-like segments. By
capturing images and using deep learning algorithms such as VGG16, waste items can be
accurately classified and segregated. The system achieves over 90% accuracy and aims to
provide more effective results than existing solutions. The adoption of hybrid and
decentralized computing models helps reduce delays and optimize computational
resources, ensuring efficient waste management.

Smart Bin for Segregation of Household Wastes

According to the study by Rukkumani et al. (2022), a smart waste bin has been
developed for the separation of household garbage. The bin is able to recognize the type
of trash, whether it is degradable, non-degradable, or metallic waste, for an efficient
waste management system. The total amount of municipal garbage generated in
metropolitan India is very large, and segregating this bulk waste is a time-consuming and
expensive process. To address this issue, the researchers developed a device called a
“Segregating bin” that can separate garbage immediately from the household, potentially
providing an effective way to reduce the burden on a large scale.

Smart Trash Bin Segregation and Identify and Create Alerts on the Level of Waste
Present in the Trash Bin

The concept of a “smart city” that utilizes IT technologies is evolving, according

to Dasari et al. (2022). Concerns about the health impacts of acidification and ecosystem
quality are growing among the population. To address this, the researchers established
electronic waste segregation using sensors and generated an alarm notification based on
28
the level of waste and leftovers to prompt the bin to be emptied. The implementation of a
waste management system is essential for smart cities.

Smart Waste Bin: Mechanical and AI Based Waste Segregation

As cities become more populated, poor waste management has led to the spread
of diseases and increased costs of processing waste, according to Zubair et al. (2022).
The initial stage of source segregation in waste management often fails. To address this,
the researchers proposed a smart waste segregation system at the source that uses image
processing, artificial intelligence (AI), and mechanical sorting. A prototype was
developed to demonstrate how the separation process works in a laboratory setting, with
the potential for application in commercial waste management.

Internet of Things based Intelligent Waste Segregation and Management System for
Smart Home Application

The Internet of Things (IoT) has a significant impact on real-time data

monitoring, including trash segregation and control in large cities, according to
Bhuvaneswari et al. (2022). The study proposes an IoT-enabled smart bin that uses a
microcontroller and various sensors to control the process. Temperature and humidity
sensors are used to separate wet and organic waste, while inductive proximity sensors
identify metal waste. Infrared sensors are used to monitor the fill level of the bin.
Bluetooth allows for short-range monitoring via a smartphone app, while IoT enables
remote control of real-time data collection. An intelligent GUI tracks the unfilled levels
of each bin, while a central controller manages filled bins using a deep learning
technique.

An IoT Based Automatic Waste Segregation and Monitoring System

Environmental risks associated with a growing population include the mixing of

biodegradable and non-biodegradable waste, according to Megalan Leo et al. (2022). In
India, over 150,000 metric tons of waste are generated each day. To improve waste
management efficiency and support the Swachh Bharath Scheme, which aims to keep

29
India clean and green, the researchers propose an automated trash segregation bin. The
IoT-based system monitors the waste process and separates waste using sensors and servo
motors. The system categorizes waste into three types: dry, wet, and e-waste. It notifies
the company when the bin is 50% full and escalates the warning at 90%. This application
can assist in achieving waste management and segregation.

BinWare: Intelligent Waste Segregation Using Machine Learning, Image Processing

and IoT

Proper waste segregation is essential, and intelligent waste segregation systems

have great potential thanks to machine learning and IoT, according to Aggarwal et al.
(2022). Despite much work being done in this area, there are still significant gaps in the
accuracy of separation techniques used. These works use convolutional neural networks,
the inception approach, and other techniques. Various types of waste are used to verify
the accuracy of the designed binware. The paper proposes a viable approach to intelligent
waste segregation that utilizes machine learning and IoT techniques.

An IoT-Based Approach in Automatic Garbage Segregation to Develop an

Intelligent Garbage Segregator

Effective waste management is a significant concern worldwide due to limited

dumping space, low awareness, and abundant financial resources, according to Aditya et
al. (2022). Waste segregation is the first step, but many people lack the necessary
knowledge to carry it out. At dump sites, waste is often manually separated, requiring
additional labor and money. With the primary goal of separating waste into biodegradable
and non-biodegradable materials, this study proposes a smart device to sort trash using an
IoT-based technique.

A Smart Waste Disposal System: To Encourage Proper Waste Disposal

Inappropriate waste disposal is becoming a problem in Sri Lanka due to a lack of

segregation, insufficient collection techniques, and a lack of public support, according to
Alwis et al. (2022). To address this issue, the researchers propose an Internet of Things

30
(IoT)-based solid waste management system that tracks trash cans, directs garbage trucks,
forecasts the amount of waste generated, and rewards proper disposal. One of the
prototypes is a smart bin that automatically opens and closes and measures waste levels.
Other components include machine learning prediction models, mobile applications for
garbage collectors and the general public, and web applications for statistical data and
reporting. This inexpensive system manages large amounts of waste using already
available resources.

Accuracy level of smart trash bin with organic and inorganic waste segregation
feature using color and humidity sensors

One way to simplify the waste handling process and reduce waste is to use
intelligent trash cans with a system for separating organic and inorganic waste, according
to Amal and Soewito (2021). However, previous efforts have shown that currently
available Smart Trash Bins only use one sensor, either the color sensor or the humidity
sensor alone, resulting in inaccurate segmentation. To increase the accuracy of the waste
segregation process, the researchers integrated two sensors, the color and humidity
sensors together. The results of data testing conducted using the methodologies employed
in this study showed that they performed better than those reported in other studies.

Waste Segregation Using Smart Bin and Optimization of Collection Routes

Waste management is a significant global issue due to increasing population

growth, urbanization, and complex living patterns, according to Gounder et al. (2021).
India has policies in place to address waste, but they are not effectively implemented or
enforced. The aim of this project is to develop an optimized model for waste management
and collection that focuses on waste segregation. The proposed system detects different
types of waste, separates them at the user end, and alerts authorities when trash cans are
filling up. In case of an issue, it directs users to the nearest trash can and provides proper
dumping technology.

31
Smart Solid Waste Management System Using Blockchain and IoT for Smart Cities

To make waste segregation easier, Paturi et al. (2021) proposed a novel smart
waste management system that integrates blockchain and IoT with smart bins. The
system uses smart contracts to reward users for properly disposing of waste. A prototype
model was tested on several networks, with the Matic test network outperforming the
Ropsten and Binance Smart Chain test networks. The proposed system offers
transparency, traceability, and scalability while eliminating single points of failure.

Reward Based Smart Bin with Waste Segregation for Encouraging People in Waste
Management System

Separating waste into biodegradable and non-biodegradable waste is one of the

biggest challenges in countries like India, according to Swankg et al. (2021). Plastic
products pose a serious threat to both humans and the environment and should be
properly segregated and recycled. Non-biodegradable plastic waste blocks the ground’s
ability to absorb rainwater, reducing the groundwater level. When these wastes are
burned, carbon monoxide is released into the air, endangering nearby individuals. This
idea aims to motivate people to segregate waste into biodegradable and
non-biodegradable waste by offering them reward points, which they can use for things
like EB bills and rations.

IoT-Enabled Smart Waste Management Systems for Smart Cities: A Systematic

Review

As urbanization, wealth, and consumption increase, so does the production of

waste, according to Sosunova and Porras (2021). Smart waste management (SWM)
involves managing waste trucks, infrastructure, and truck routes, as well as collecting and
processing data from sensors on smart garbage bins. This report provides a summary of
the body of knowledge on SWM systems through a comprehensive analysis of 173
original studies. The study outlined the methodologies, services, sensors, actuators,
stakeholders, data sharing, and research needs in SWM systems and provided
recommendations for implementing SWM systems at the city and SGB levels.
32
Friendly Waste Seggregator Using Deep Transfer Learning

The study by Faiza et al. (2021) presents a deep transfer learning-based friendly
trash segregator. The basic idea is that the system will take a picture of the trash can when
it is full and use a trained convolutional neural network to identify the type of waste.
There are several classifications of waste, including plastic, paper, metal, and so on,
which are further separated into biodegradable and non-biodegradable waste.

Smart Trash Segregator Using Deep Learning on Embedded Platform

India’s growing population generates an increasing amount of waste, most of

which is disposed of in landfills rather than being recycled, according to Mahakalkar and
D, R., (2021). To address the segregation issues with traditional waste management
systems, this paper proposes a Smart Trash Segregator (STS) at the source level. The STS
uses an embedded platform with a sensor module and a trained deep neural network to
autonomously separate waste into six categories. Mechanical actuators deposit the
categorized waste into the appropriate container. The technology is effective for recycling
and environmentally friendly waste management, achieving 85-96% accuracy and being
more efficient than conventional segregation systems.

Computer-Vision Enabled Waste Management System for Green Environment

Waste management is essential for keeping the environment green, but managing
municipal waste is challenging, according to Hewagamage et al. (2021). Most urban
councils use a manual method, which has negative impacts on residents and cleaning
personnel, contaminated waste, hinders information management, and causes untimely
waste removal. This research proposes an automated system that uses an ICT-based
design to address these issues and create an efficient waste management system. The
proposed model introduces a real-time monitoring smart waste bin system based on
computer vision that combines GIS, sensor-based IoT devices, and computer vision. The
system enables automated waste segregation and category expansion, informs users in
real-time of the status of their waste, calculates the best routes for collection, and presents
these routes in a mobile app. A low-cost prototype is demonstrated.
33
AIoT-Based Smart Bin for Real-Time Monitoring and Management of Solid Waste

The study by Bano et al. (2020) presents a proposed smart bin mechanism (SBM)
for smart cities, aiming to ensure city cleanliness and enable real-time monitoring of trash
bins using Artificial Intelligence of Things (AIoT) principles. The SBM operates based
on the 3R concept of Reduce, Recycle, and Reuse. It allows for obtaining real-time
information about each bin, preventing overloading issues. By implementing this
framework, the system can reduce labor costs, save time and energy, and mitigate the
spread of diseases by maintaining clean cities. Decision-making for the installation of
trash bins in appropriate locations within cities is facilitated through the use of fuzzy
logic. The framework is implemented within the NetLogo multiagent modeling
environment.

A Smart Waste Management Solution Geared towards Citizens

The study by Pardini et al. (2020) introduces a comprehensive solution

comprising hardware, software, and communication components, with the aim of
enhancing waste management while actively involving citizens in the process. The
system adopts an Internet of Things (IoT) approach, where smart bins are equipped with
sensors to continuously monitor the fill levels of individual compartments in real-time.
The solution consists of three key elements: the smart bin itself, integration with an IoT
middleware, and a user application. By developing a functional prototype of the smart
bin, creating a waste management application, and conducting a real-world experiment
for evaluation and validation, the study demonstrates the system’s ability to effectively
transform people’s waste management habits and optimize both economic and material
resources.

Iot Based Smart Dustbin

This paper by Maddileti et al. (2020) presents a prototype for a Smart Dustbin
project, which incorporates various components to enhance waste management
efficiency. The prototype features an automated lid that opens upon detecting human
hand movement and waste. Additionally, the dustbin uses ultrasonic sensors to measure
34
the waste level, and this information is conveyed through LED notifications. The key
components used in this prototype include Arduino, NODEMCU, Servo Motor, and
Ultrasonic Sensors. The Blynk application serves as the software component, enabling
the reception of notifications. This Smart Dustbin prototype serves as a foundation for a
larger Smart Waste Management System, where officials can proactively clean or empty
the dustbins based on the notifications received. This approach eliminates the reliance on
individuals within the community to make calls to inform garbage trucks for waste
collection, thereby improving overall waste management efficiency.

Energy saving smart waste segregation and notification system

A significant issue is the increase in waste production brought on by urbanization

and industrialization, according to Sarker et al. (2020). Waste separation is essential for
environmental protection. Bangladesh’s traditional manual approach requires a lot of time
and can be harmful to health. The proposed smart segregation system alerts the garbage
collector when the trash container reaches a certain level, and if it is not collected
promptly, it alerts the City Corporation. The device uses solar energy on sunny days to
save electricity and help reduce greenhouse gas emissions.

IOT Assisted MQTT for Segregation and Monitoring of Waste for Smart Cities

According to an article by Jaikumar, K., et al. (2020), an architecture for waste

segregation and monitoring using Message Queuing Telemetry Transport (MQTT) can
help manage garbage collection. This is important because poorly planned and haphazard
rubbish collection is a significant problem. The solution is an embedded Internet of
Things (IoT) system that monitors waste deposition and separates it into dry and moist
trash. The bins contain sensors, and the bin level information is sent using the IoT.
MQTT is a simple messaging protocol designed for embedded systems, sensor networks,
and mobile apps.

35
Smart Waste Management System using ARDUINO

According to a paper by Vijay, S., et al. (2019), a system has been developed to
effectively manage waste in large cities without the need for constant manual monitoring.
The solution involves an embedded IoT system that individually monitors each dumpster
for the amount of waste deposited. An automated system is used to segregate wet and dry
waste using a mechanical setup and sensors, including an IR sensor to detect waste and a
moisture sensor to detect wet waste. Based on the sensor readings, a motor rotates to
direct the waste into the appropriate container. Ultrasonic sensors are used to measure the
amount of waste present in each container, and if one reaches its full capacity, an alert
message is sent to the personnel responsible for waste management.

Smart Bin (Waste Segregation and Optimisation)

A study by Pereira, W., et al. (2019) describes a smart bin with advanced features,
primarily focused on waste segregation. The bin has four separate compartments for
plastic waste, wet waste, dry waste, and wastewater from the auto clean feature.
Ultrasound sensors allow the bin to open automatically when a person approaches,
providing a hands-free and hygienic experience. The smart bin also has data analysis
capabilities, giving users insights into the quantity and types of waste they dispose of
through graphs and real-time data reception. A connected app sends reminders to users
when it’s time to dispose of their garbage. This innovative concept promotes separate
waste disposal and distinguishes between recyclable and non-recyclable waste. Overall,
the smart bin is an efficient and hygienic waste disposal and segregation system that
contributes to waste optimization.

A Comprehensive Study on Waste Segregation Techniques

According to a paper by Gangwani, M., et al. (2019), segregating waste materials

is important to prevent contamination of landfills with toxic substances. When materials
like polythene bags, old furniture, and e-waste mix together, hazardous substances can
leak into the environment. Proper segregation allows wet waste to be converted into
useful resources like compost or methane gas, which can be used as a natural alternative
36
to chemical fertilizers or as an energy source. Metallic, plastic, and paper waste can be
reused or recycled, reducing the demand for new materials. The paper compares various
automated waste segregation methods that use different technologies and considers
implementing an automated waste segregation process a fundamental requirement for
efficient waste management.

Smart Bin (Waste Segregation and Optimisation)

A study by Pereira, W., et al. (2019) describes a smart bin with a primary feature
of dividing rubbish into four sections for plastic, wet, dry, and wastewater from the
auto-clean feature. The bin also has other features, such as ultrasound sensors that allow
it to open hands-free when someone approaches, improving its hygienic properties. The
bin evaluates the quantity and type of garbage disposed of through live data reception and
provides daily, weekly, and monthly data. A connected app sends reminders to take out
the trash. The smart bin is an effective and hygienic solution for waste disposal and
segregation, separating recyclable garbage from non-recyclable waste and improving
waste management.

Smart garbage bin systems – A comprehensive survey

According to an article by Soni, G., & Kandasamy, S. (2018), modern cities face
challenges related to smart transportation, smart grid, smart environment, and smart
lifestyle. A clean environment is necessary for a city to be considered a “smart city.”
Proper waste management is a serious challenge, and conventional methods are not
sufficient. Smart waste management is crucial because it impacts lifestyle, healthcare,
and the environment. The article presents a framework for a smart garbage management
system that can be implemented in large cities after reviewing numerous proposed
methods for smart bin systems.

IoT enabled dustbins

The study by Mirchandani, S., et al. (2018) proposes a solution to the health risks
posed by garbage piled high in public areas, such as stink and sickness. The solution

37
involves IoT-enabled trash cans that use RFID tags to track waste and update each bin’s
status on the city server. The server determines points based on the weight of waste added
and updates a virtual wallet. When a bin is full, the system alerts the municipality and
suggests the quickest way to empty all bins based on the capacity of municipal garbage
loading vehicles. The user receives assistance in categorizing waste, and the smart bin
can communicate its contents to the recycling network. The technique aims to make
collecting and sorting waste as inexpensive as possible.

2.2.2 Related Local Studies

SOLID WASTE MANAGEMENT AWARENESS AND PRACTICES AMONG

ZONE 2 HIGHER EDUCATION INSTITUTION IN ZAMBALES

The study by Gantang, M.A. (2022) evaluated the knowledge and practices
related to solid waste management among respondents in Zone 2 Higher Education
Institution (HEI) in Zambales. Data was collected through a survey questionnaire, with a
total of 612 teachers and students participating. The findings showed that the majority of
respondents, who were from public colleges, strongly agreed with trash segregation
policies. There was consensus on waste disposal, waste reduction, recycling, and reuse,
as well as awareness of solid waste management practices. The survey also found a
strong link between students’ solid waste management methods and awareness.

Practices on Waste Disposal and Its Environmental Effects

The study by Aming-Hayudini, M. E., et al. (2022) explored garbage disposal

habits among residents of Barangay Upper Sanraymundo, Jolo, Sulu, Philippines. The
researchers used a quantitative approach to survey 99 residents from a population of
8,308. The data showed that the majority of respondents had a high school education,
consistently practiced garbage disposal, and were aware of the environmental impact of
improper waste management. The study concluded that solid waste has a major
environmental impact and recommended increasing the implementation of Republic Act
9003, the Ecological Solid Waste Management Act of 2000, and promoting the 3Rs
(reduce, reuse, and recycle) in Jolo, Sulu, Philippines.
38
A Collaborative Application for Assisting the Management of Household Plastic
Waste through Smart Bins: A Case of Study in the Philippines

The study by Sidhu, N., et al. (2021) suggests a collaborative infrastructure to

monitor domestic plastic waste, which can be difficult to manage for the elderly or
disabled. The system uses simple smart bins equipped with weight scales and a smart
application that estimates plastic generation and determines ideal routes for waste-pickers
using a route-planning algorithm. The concept was tested in a simulated setting in
Quezon City, Philippines, where plastic waste is a major issue. The results showed that
the technique allows waste-pickers to collect more than 80% of domestic plastic waste
before the bins are completely full.

THE DESIGN AND IMPLEMENTATION OF A SMART TRASH BIN FOR

QUARANTINE CENTERS IN BOHOL

The study by Batabat, J.V., et al. (2021) proposes a safety design for an intelligent
trash container bin for quarantine centers in Bohol to handle highly contagious waste
generated during the COVID-19 pandemic. The no-touch trash can uses a servomotor and
PIR sensor to detect a person’s presence and dispose of dirt and contagious waste. An
ultrasonic sensor is used for real-time monitoring of the garbage level within the bin and
alerts the user when it reaches maximum capacity. When the bin is full, a red LED and an
LCD screen message indicate this.

Internet of Things (IoT) Smart Bin: a Prototype to Garbage Segregation Focusing

on Recyclable Materials

The study by Villadolid, H., et al. (2021) describes the development of a

prototype IoT smart bin for garbage segregation, focusing on recyclable items. The smart
bin uses sensors to distinguish between different types of waste and has separate
chambers for plastic waste, wet waste, and dry waste. Recycling is emphasized as a
viable method for managing solid waste disposal, providing both economic and
environmental benefits.

39
An Evaluation of Automated Waste Segregation Systems: A Usability Literature
Review

The study by Carpenteros, A.E., et al. (2021) assesses the effectiveness, benefits,
and drawbacks of automated waste segregation systems using data from 14 studies
published between 2011 and 2020. The systems scored highly in effectiveness and
efficiency (87.5% and 82.5%, respectively), but poorly in learnability (57.5%). The study
emphasizes the importance of effective waste management, especially as waste
accumulation in urban areas has increased due to the pandemic.

Garbage Practices in Higher Education for Healthy Academic Environment

The study by San Jose, A. (2021) analyzes the garbage management behaviors,
ideas, and recommendations of 33 participants at home and at school. Using a
qualitative-Heideggerian phenomenological methodology, the researchers found that the
participants understood the value of waste facilities, were aware of current laws and
policies on garbage management, and believed in being responsible citizens. They
advocated for stricter regulations, the application of penalties, and education. The report
suggests that educational institutions should provide structural trash facilities, encourage
proper garbage disposal, and play a significant role in garbage management by
integrating it into their corporate social responsibility (CSR) and curriculum.

Trash Bin Compactor: Creating Alternative Solution To Garbage Segregation

Management

According to a study by Amila, P.J., et al. (2021), the Self Monitoring Automated
Route Trash Bin (SMART Bin) is an electronic enhancement of existing trash bins. The
aim of the study is to design a trash bin that can monitor its own trash level and navigate
freely along corridors without external control. The SMART Bin uses Gizduino 4.0 as its
main processing unit and follows an ultrasonic-sensing mechanism to generate a route for
it to follow. It is a wall-following and obstacle-detecting device.

40
Development of Smart Waste Bin Segregation using Image Processing

The study by Jimeno, F.N., et al. (2021) aims to create a Smart Trash Bin
Segregation using Image Processing to support trash segregation through waste
identification and segregation based on machine learning. The prototype uses image
processing with three categories - biodegradable, non-biodegradable, and unknown - to
separate solid waste into the correct bins and carries out accuracy testing using statistical
techniques. The device, intended for use in schools but applicable in other
establishments, can help reduce the issue of trash segregation and contribute to a more
environmentally friendly society without sacrificing health and cleanliness. Function tests
obtained an overall accuracy of 97.33%.

Smart E-Waste Bin Development Based on YOLOv4 Model

According to the study by Sampedro, G.A., et al. (2021), waste management

improvements are a major topic of research when establishing smart cities. The goal of
the research is to create a smart e-waste bin that can recognize the type of e-waste
deposited and compensate consumers for their deposit. The system uses You Only Look
Once v4 (YOLOv4) to distinguish between different types of e-waste. The device is
tested for precision, recall, and accuracy, achieving an average accuracy of 93.33% in
correctly detecting the type of e-waste submitted.

e-TapOn: Solar-Powered Smart Bin with Path-based Robotic Garbage Collector

According to the study by Samonte, M.J., et al. (2021), the Internet of Things
(IoT) enables everyday devices like trash cans to be controlled online. The smart bin
solution allows cleaning staff to detect hygienic issues in real time, improve waste
segregation cleanliness, and increase productivity. The study adapts the use of an
automated guided vehicle (AGV) and color sensor to create a practical and affordable
trash management system. A solar-powered Raspberry Pi microprocessor with an infrared
sensor is combined with a path-based robotic garbage collector to produce a smart bin for
managing solid waste. The e-TapOn smart bin system, which includes a robotic garbage

41
collector, a functional smart bin, and a web-based application, is developed, tested, and
integrated to contribute to smart cities.

Awareness and Compliance on Waste Segregation: Implication to a Waste

Management Program in a University

According to a study by Dolipas, B., et al. (2020), the knowledge and adherence
of students to garbage classification and segregation at a university in the Philippines was
evaluated. A questionnaire was used to determine the level of awareness, and trash audits
were used to assess compliance. The findings showed that most people were aware of
biodegradable garbage, except for third-year students who had limited knowledge.
Students showed very little awareness of trash classification and compliance with waste
segregation, regardless of where they lived. The study suggests increasing awareness
initiatives and encouraging education to improve compliance.

Design and Implementation of Automated Waste Segregator with Smart

Compression

According to the study by Endaya, J.P., et al. (2020), the goal is to create and
implement an automated trash can with smart compression to correctly segregate waste.
The system uses two interconnected Arduino Unos, push buttons to identify the type of
garbage, LEDs to indicate proper input, a stepper motor to rotate the trash cans, and an
ultrasonic sensor to determine when compression is needed. A linear actuator compresses
waste to maximize bin space, and LEDs indicate when the bins are full. The technology
can help municipalities and local government units create a greener environment and
improve solid waste management.

VENDOBIN: An IOT-based Plastic Bottle Waste Disposal Vending Machine

According to a study by Dacay, W.J., et al. (2020), the goal of the research is to
encourage correct plastic bottle disposal with a reward system based on VendoBin, a
machine controlled by a Raspberry Pi 3. Infrared sensors distinguish between plastic
bottles and non-plastic items, while ultrasonic sensors detect if the bin is full. Users earn

42
points for each transaction, stored in a text file database, and can redeem them for usable
items with a unique code. When the bin reaches its maximum capacity, it sends a text
message to authorities requesting disposal. The VendoBin correctly classifies 100% of
plastic bottles and provides codes for new and returning users.

Literature Review of Automated Waste Segregation System Using Machine

Learning: A Comprehensive Analysis

According to the study by Flores, M., and Tan, J. (2019), efficient waste
management and disposal have been researched using hardware components like
raspberry pi and algorithms for object recognition and classification. However, these
devices only work with maximum accuracy on images containing single objects.
Convolutional Neural Networks play an important role in object recognition, but existing
systems aim to classify only single objects in an image. In real-world scenarios, it is
difficult to separate individual objects from a garbage pile. There is a need to recognize
and segregate multiple objects in a single image and classify recyclable materials into
different categories to make the recycling process easier. The increase in electronic waste
also presents a challenge for recycling.

i-BIN: An Intelligent Trash Bin for Automatic Waste Segregation and Monitoring
System

According to the study by Pamintuan, M., et al. (2019), an autonomous trash

segregation system was created using the Internet of Things. The project specifically
created a sensor-equipped trash can that can intelligently separate waste and provide
monitoring information about rubbish pickup. Automatic classification of rubbish was
processed using image recognition and machine learning techniques. More than 2000
samples were used to train the produced prototypes to distinguish between biodegradable
and non-biodegradable garbage, and they were successful.

43
Microcontroller-Based Sorting System of Plastic Bottles, Tin Cans and Paper
Cartons through Ultrasonic Sensing

According to a study by Barsaga, R.G., et al. (2018), the “Microcontroller-Based

Sorting System of Plastic Bottles, Tin Cans and Paper Cartons through Ultrasonic
Sensing” by Vonn Vincent Quinones proposes an electronic system that aids in sorting
recyclable materials like plastic bottles, tin cans, and paper cartons. The system uses
ultrasonic sensing to determine the type of material to be segregated and servo motors to
automatically place the material in its correct bin. The study found that ultrasonic sensing
is an effective approach for segregating plastic bottles, tin cans, and paper cartons.

Microcontroller Based Coin Counter with Segregator and Packing System

According to a study by Sequera, M., et al. (2018), the “Microcontroller Based

Coin Counter with Segregator and Packing System” focuses on providing automation and
eliminating errors in coin sorting and handling. The research uses the Gizduino
ATMEGA 328 microprocessor and the Arduino programming language. The machine
accepts Philippine peso coins with denominations of one, five, and ten pesos.

Automated Waste Segregator (Smart Bin) in Classroom Level

According to a study by Paniza, J., et al. (2017), the primary goal of the research
is to create an effective waste segregator using different types of detectors. The study,
which will span 3-4 months, will focus only on waste segregation, specifically plastics
and papers, as these are the most common types of waste. The researchers devised a
method of trash segregation that is more efficient, user-friendly, eco-friendly, and
time-efficient.

44
2.3 Conceptual Framework

Figure 1: Conceptual Model of the Study

The conceptual framework, as illustrated in Figure 1, shows the different stages

that the researchers need to accomplish to achieve the objectives of the study. The input
stage includes knowledge, hardware, and software requirements for the development of
the EcoSort System using IoT. Knowledge requirements include Basic concepts of IoT,
Deep Learning Algorithms(CNN), C++ Programming, and Python (Anaconda)
Language. Developing the system requires computers, Arduino Board, GSM Module
and Computer Vision device as hardware requirements. As part of the software
requirements, Jupyter notebook will be used to train the Smart Bin using Image
Processing and Computer Vision techniques which will be the main objective of the
study in identifying the type of waste. Arduino will be used to integrate IoT in the study,
the Smart Bin will consist of two containers which will only open when the System
identifies the type of waste. Additionally, Visual Studio Code will be used for

45
programming in Django Framework. Under Agile Software Development, there will be
phases that need to be undergone before presenting and deployment. The system is
evaluated in terms of functional stability, performance efficiency, usability, reliability,
and portability from ISO 25010 software quality model. The evaluation of the system’s
performance is done through an evaluation questionnaire. The output of the study is to
develop an Advanced Smart Bins Waste Segregation and Management with Image
Recognition and Incentive Rewards

2.3 Theoretical Framework

Figure 2: Theoretical Model of the Study

This theoretical framework integrates the following theories:

Incentive Theory: Theories such as Self-Determination Theory (SDT) and Operant

Conditioning can be used to explain how incentives can motivate people to engage in
effective waste segregation activities.

46
Human-Computer Interaction (HCI) and User Experience (UX): Theoretical
frameworks in HCI and UX can be used to study how users interact with the EcoSort
system. Theories such as User-Centered Design (UCD) and Information Processing
Theory can help guide the design and evaluation of the user interface and overall user
experience.

Sustainable Behavior Theories: To investigate the factors influencing sustainable waste

management practices, theoretical frameworks such as the Norm-Activation Model
(NAM) or the Theory of Planned Behavior (TPB) can be used. These ideas concentrate
on the role of personal norms, subjective norms, and perceived behavioral control in
fostering pro-environmental behavior.

Systems Theory: The relationships and dynamics of the EcoSort system can be
understood using systems theory. It can assist in analyzing the interconnection of smart
bins, image recognition technologies, incentive mechanisms, user behavior, and the
broader waste management ecosystem

2.3 Operational Definition of Terms

Arduino: In the study, the use of Arduino Boards and GSM modules will be the main
controller on the function of the lids of each bin. The lid of the bin will only open based
on the type of waste identified in the system.

Citizens: The users are the beneficiary in which the proper waste management practices
and incentives can enable them to have enough knowledge and participate in a more
sustainable environment.

Computer Vision: In the system, the use of Camera will enable the system to be
connected to the IoT Smart Bins to process the images of the waste real-time and classify
the waste so that the Smart Bins will only open based on what type of waste is identified.

Deep Learning Algorithms : The use of Deep Learning Algorithms such as

Convolutional Neural Network (CNN) for training and processing the images viewed
through the Computer Vision in the system.
47
Image Recognition: In the system, the Image Recognition is under the term of Image
Processing wherein it is tandem with the Computer Vision for Processing the images
within the system.

Incentive Rewards: The rewards will be coupon based where Citizens can exchange
their coupon for Potential Recycled Materials (Ecobag, etc.).

Internet of Things(IoT): The EcoSort SmartBins will serve as a new approach to the
existing IoT. With the use of Deep Learning algorithms and Arduino based for the
functionality of the Smart Bin.

Smart Bins: In the study, the Smart Bins will serve as the beneficiary of the system
wherein EcoSort is integrated to advance smart bins. The Smart Bins will consist of five
containers which will be labeled based on Recyclable materials (Plastics, Paper, Cans,
Glass, Metal).

Waste Management Practices: The study enables citizens to have knowledge about
proper waste management practices through interacting with the Smart Bins using IoT.

Waste Segregation : This is the process where the Image Recognition feature will be
used so that the type of waste is classified accordingly with Computer Vision using IoT.

48
 Chapter Three

METHODOLOGY

3.1 Research Design

3.1.1 Context Diagram

Figure 3: Context Design

The Context Diagram, as illustrated in Figure 3, shows how the study uses the
basic concept of IoT in integrating the Smart Bin to the Waste Management System
(EcoSort). The Arduino is the main feature on how the System will classify the waste and
connect the transmission through the lids of the Smart Bins, the role of Arduino is to
monitor real-time the Bins fill level and send SMS notification through the GSM Module
to collect the Bins that are full Real-time. The Computer Vision connected in the System
will enable the data to be stored in the database wherein the system will get data for

49
Processing the image for identifying the type of waste to integrate waste segregation
practices.

3.1.2 System Architecture

Figure 3.1.2 System Architecture Design

The System Architecture, as illustrated in Figure 4, shows how the Citizens can
interact with the Smart Bins through Arduino and the System with computer vision. The
system enables Image Recognition through Image Processing and computer vision where
the data will be processed in Deep Learning Algorithms(CNN) to classify the type of
waste and send the classification through the Smart Bins. The Smart Bin’s Fill Level will
be monitored by Sensors and will send SMS notification through a GSM module to the
Garbage collector to provide real-time updates about the Bin statues.

50
3.2 Research Development

Figure 3.2 Agile Software Development Diagram

The researchers opted for Agile Software Development as the system

development methodology in their study titled "EcoSort: Advancing Smart Bins Waste
Segregation and Management with Image Recognition and Incentive Rewards."
Following Highsmith's (2001) rationale, this methodology aims to accelerate the delivery
of functional software and research prototypes, enabling early testing, feedback, and
validation of ideas. By employing this approach, the researchers anticipate reduced costs
and improved efficiency throughout the development process. Agile Software
Development encompasses several phases such as planning, requirement analysis, design,
development, testing, and deployment, each of which significantly influences the
research's progress.

Planning

This phase involves focusing on the concept of the research. In terms of planning,
it mainly considers and analyzes the existing issues of current methods for Waste
management and the objectives that are needed to be fulfilled and achieved in each
timeframe. Moreover, data collection is needed in this particular phase in order to justify
that the problems have a huge impact on the process.

51
 1. Data Collection
2. Identifying the problems
3. Indicating the objectives
4. Defining the scope and delimitation

Requirement Analysis

The researchers will be able to identify requirements, features, and functionalities

that must be considered in the system development of this research in accordance with
the data collected in the previous phase.

1. Determining both the developers’ and users’ hardware and software requirements

2. Identifying the characteristics, features, and functionalities of the system

3. Finalizing the goals and objectives

Designing

In this phase, choosing the appropriate design layouts for the user interface and
user experience in accordance with the Existing Smart Bins and Waste Management
Systems was taken into consideration. The researchers ensure that the system is scalable,
available, and reliable for the administrators and easy to navigate and use for the students.

1. Establishing the architecture of the system

2. Considering data attack prevention, and securely store data security.

Development

This phase is the development of the web system itself through coding and
integration of applications and libraries and compiling these in order to produce a running
and executable system. Using Visual Studio Code as our compiler and xampp server in
order to establish connection to the database.

52
 1. Development of the web-based system through the use of PHP programming
language and Laravel framework for faster progress. Using Anaconda Python in
Jupyter Notebook for training the Model in processing data of waste.
2. Use of CSS and Bootstrap for the front-end framework
3. Integration of several libraries into the system

Testing

In this testing phase, this is where the web-system features and functionalities of
the Smart Bin satisfy the objectives that must be met in the research. It must be in line
with the expected output. Also, this phase is recognized as a crucial part of the system
development to an extent that all bugs and errors must be resolved to prevent future
issues. Furthermore, the finalized system has undergone system evaluation done by the
potential users, which are the Citizens of Manila.

1. System Testing

2. Debugging

3. Checking of objectives

4. System Evaluation

5. System Deployment

Deployment

The deployment phase of the research study plays an important role in getting
user feedback in a real-world situation. This step allows researchers to identify any
unforeseen concerns that may have been missed during testing and supports system
modifications. The deployment phase ensures optimal performance and project
requirements by assuring correct installation and configuration.

53
3.3. Data Collection
The researchers decided that quantitative research method designs will be
conducted in order to identify the issues with the current system and implement
adjustments. This will significantly help the researchers in terms of data collection.

3.4 Sampling Technique

In this research, three hundred (300) citizens in Metro Manila, will be the
respondents. Moreover, a random sampling technique was used in order to obtain the
needed information that has a significant impact on this research. This type of sampling
method is applicable because it evaluates the citizens from the said region regarding the
functionalities and features of the system.

3.5 Data Analysis and Procedures

The researchers will collect and obtain the data for this research through primary
resources. It was made simple for the researchers to compile the required and necessary
information and employ the subsequent study techniques. The main source of this
research is the LGU of Manila, which will be approached by the researchers through a
semi-structured interview to learn more about the current state of the waste segregation
management. With this knowledge in hand, the decision was made to conduct the
research. Moreover, the researchers are currently coordinating with the said beneficiary to
ensure that this research will be in line with the policies, rules, and regulations set by the
organization. Furthermore, the researchers will conduct an evaluation survey exclusively
for Citizens around Metro Manila in accordance with the ISO 25010 model. The
information provided by the respondents will be treated with the utmost confidentiality
based on the Data Privacy Act of 2012.

Upon completing the survey, the researchers will be able to gather the required
data provided by the respondents and analyze the data in order to produce reliable results
that are necessary for interpretation and findings. The method that will be used by the
researchers to obtain the average value of the population is through weighted arithmetic
mean.

54
3.6 Evaluation
Evaluation is a crucial step in the development of system software to ensure that it
aligns with the research's goals and objectives, meets quality standards, and fulfills
specified requirements. The International Organization for Standardization (ISO) 25010
provides a set of characteristics and qualities that serve as the standard for evaluating
system software and determining relevant metrics. By adhering to these standards, the
EcoSort System, utilizing IoT technology, focuses on specific attributes such as
functional stability, usability, reliability, security, and maintainability, which are most
suitable for the system in question.

Scale Rating

4.51-5.00 Excellent

3.51-4.50 Very Good

2.51-3.50 Satisfactory

1.51-2.50 Fair

1.00 – 1.50 Good

Table 1: Descriptive Scale Evaluation Table for interpreting Numerical Findings

The researchers will conduct the ISO 25010 evaluation on the Citizens of
Maynila. The survey will use a 5-point Likert Scale in order to evaluate and assess the
system in accordance with the selected ISO 25010 attributes.

55
 Upon computing the evaluation results, it will be determined based on the table
shown above.

3.6.1. Statistical tools

In order to evaluate and determine the average value of the entire population of
respondents, the assessment findings will be calculated and analyzed using the mean
percentage. To evaluate the image recognition model, precision and recall will be utilized,
the scores will then be used to assess how accurate the training of the model.

56