Received 24 March 2023, accepted 1 May 2023, date of publication 10 May 2023, date of current version 18 May 2023.

Digital Object Identifier 10.1109/ACCESS.2023.3274862

Design for an Intelligent Waste Classifying

System: A Case Study of Plastic Bottles
SUCHADA RIANMORA , POOMPACHARA PUNSAWAT, CHARINYA YUTISAYANUWAT,
AND YOSAKORN TONGTAN
School of Manufacturing Systems and Mechanical Engineering (MSME), Sirindhorn International Institute of Technology, Thammasat University, Pathumthani
12121, Thailand
Corresponding author: Suchada Rianmora ([email protected])
This work was supported by the Thammasat University Research Fund under Contract TUFT 29/2565.

ABSTRACT The use of plastic bottles has become a significant environmental concern, and recycling
them has become a priority. Small and medium-sized recycling companies must collect and categorize large
volumes of plastic bottles and sell them to larger recycling firms, a process that is time-consuming, costly,
and labor-intensive. This manual sorting process can pose health risks, particularly during the COVID-19
pandemic, and can affect worker productivity. To address these issues, this study proposes the development
of an automated conveyor belt system that can rapidly and accurately separate plastic bottles by type. The
system utilizes an opaque and transparent plastic bottle separation platform, which saves time, cost, and
manpower. This system design provides recycling SMEs with a competitive advantage by serving as a
practical application model and a prototype with an easy-to-use concept. Key tools employed in this research
include product design development (PDD), Kansei engineering, manufacturing process design, controlling
system, and fault tree analysis (FTA). The light sensors are critical components in the separation process,
detecting the opacity or transparency of the bottles’ surfaces. The proposed prototype’s reliability will be
assessed by FTA, which considers all potential failures. This study contributes to the body of knowledge
surrounding the integration of conveyor systems and provides valuable information for businesses seeking
to optimize their sorting processes. The guidelines developed in this study can serve as a starting point for
further research on the integration of conveyors in waste sorting plants.

INDEX TERMS Product design and development, Kansei engineering, conveyor belt, microcontroller, light
sensor, fault tree analysis.

I. INTRODUCTION scale chain belt conveyors with robust output for various
The mismanagement of a large amount of waste poses a applications (Figure 1) [6], [7], [8], [9]. Chain belt conveyors
significant risk to the environment, public health, and the are a popular choice in the recycling industry due to their
ecosystem [1]. Despite these challenges, it is crucial to sturdy and durable construction. Their stability is comple-
prioritize environmental protection, public health, hygiene, mented by great flexibility, as they can be easily adapted
material recovery, waste reduction, emissions reduction, or extended to meet specific needs. These conveyors can be
and disease prevention [2]. Effective recycling depends on equipped with rubber, and are built with heavy-duty modular
efficient separation, and the market offers a wide range frames, high vertical side pieces, and quality bearings and
of separation technologies that have been developed and gearboxes. Variable speed controls can also be added to all
applied [3], [4], [5]. Advanced waste separation systems lead chain belt conveyors [10], [11], [12]. Plastic recycling plants
to higher-quality material recovery, resulting in increased use various techniques, such as remolding, to reduce the envi-
revenue. Several companies have introduced medium-to-high ronmental impact of plastic bottles. Before melting the bottles
down, they must first be separated based on their material
The associate editor coordinating the review of this manuscript and properties. Small recycling companies are often relied upon
approving it for publication was Tiago Cruz . to collect used bottles from customers [13], [14], [15].
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.
VOLUME 11, 2023 For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/ 47619
 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

II. RELATED WORKS

A. CONVEYING AND SORTING TECHNOLOGY IN
INDUSTRY AND PRODUCTION
Conveying and sorting technology is highly adaptable and
can process a wide variety of materials. However, different
systems are optimized for specific types of waste, which
highlights the importance of selecting the appropriate con-
veyor system for the task at hand. For instance, pneumatic
conveying systems are ideal for managing small waste gener-
ated during production, such as paper scraps, cardboard, and
plastic waste from a shredding system [18], [19].
Conveyor belts are versatile and can handle irregularly
shaped materials, such as metal scrap, PET, and other waste.
Chain conveyors are better suited for transporting larger
unit loads, such as containers, pallets, or goods carriers.
Conversely, tilt conveyors can handle all materials that are
FIGURE 1. Belt conveyor for waste management [8]. collected in containers or boxes, including small items like
plastic and Styrofoam, as well as bulkier waste like car tires
and wood. Industrial sectors and researchers anticipate that
However, these conveyors are typically medium-to-large in conveying and sorting technology will offer automation, safe
size, and require sufficient space for equipment and materials. and efficient removal of materials, energy conservation, and
Due to budgetary constraints, many small and medium-sized cleanliness. These requirements can be achieved through var-
enterprises (SMEs) choose to rely on manual sorting pro- ious solutions [19], [20], [21], [22], [23].
cesses to separate various types of plastic bottles. However, Pneumatic systems [24], [25], [26], [27], for instance, offer
this approach poses significant health risks to employees, a direct link between the production site and waste disposal
as bacteria can thrive on unwashed water bottles that are and can be combined with other systems like extraction sys-
left unattended for a week [16], [17]. Given the potential for tems, presses, and containers to create an automatic trans-
infections and illnesses associated with manual sorting pro- port and disposal system. Waste generated during production
cesses, it is crucial for SMEs to consider alternative methods can be transported directly to the desired destination via the
of sorting plastic bottles. Automated conveyor systems, for conveyor system, avoiding pollution at the production site
example, can perform sorting tasks quickly and efficiently, or in the factory. The conveyor system can also transport
without exposing employees to health risks. While such sys- waste directly from the factory floor to a container, freeing up
tems may require an initial investment, they can provide valuable space within the production facility for other tasks.
long-term benefits in terms of increased productivity, reduced To automate the disposal of recycled material and residual
labor costs, and improved workplace safety. It is essential for waste, a sorting plant is necessary. Modern sorting technology
SMEs to prioritize the health and safety of their employees can cleanly and efficiently separate different types of residual
when considering methods for sorting plastic bottles. Invest- materials, contributing to an environmentally friendly recy-
ing in automated sorting systems can not only reduce health cling process [22], [28], [29], [30], [31], [32], [33]. Sorting
risks but also improve overall efficiency and profitability in plants typically consist of a multi-stage sorting line that
the long run. removes foreign materials and sorts paper waste. Conveyor
To address these concerns and simplify the process, this technology is integral to the sorting process, transporting
study proposes an alternative sorting conveyor system that the respective materials to their desired points automatically.
is compact in size and capable of categorizing bottles into Clean sorting is essential for a high degree of recycling and
two groups: transparent and opaque. Belt conveyors are a circular economy, particularly for important raw materi-
widely used in industries due to their strong conveying als like paper and cardboard. To support environmentally
capacity [8], [9], [10], making them an ideal solution for friendly recycling, our sorting plants rely on a multi-stage
this purpose. The sorting conveyor system’s specifications sorting line that effectively removes foreign materials and
are determined by the type of waste, structural conditions, sorts paper waste. The plant’s advanced technology, includ-
and existing production machines. The system’s overall con- ing conveyor belts, automatically transports each material
cept focuses on reliability and practicality, ensuring that the to its intended destination, allowing for a fully automated
conveyor runs smoothly and the sorting process remains process - development of automatic sorting conveyor belt
efficient with trouble-free operation in everyday use. These using PLC [28]. Conveyor technology is a vital component of
adaptations are essential when setting up disposal sys- the sorting system, enabling efficient and seamless operation.
tems to improve the efficiency and safety of the sorting For the optimization process of conveyor belts for separat-
process. ing waste, two methods related to the optimization process of

47620 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

conveyor belts for separating waste were found to be inter-

esting and useful for further study [34], [35]. Both studies
aim to optimize the conveyor belt system used in waste sep-
aration to improve efficiency and accuracy, and the proposed
methods have shown promising results in achieving this goal.
Yasin et al. proposed an optimization method for a conveyor
belt system used in municipal solid waste separation. The
study aims to optimize the conveyor belt system’s operational
parameters, including the conveyor speed and the angle of FIGURE 2. Total design model [34].
inclination, to improve the separation efficiency of the waste.
The authors used a simulation model to evaluate the per-
formance of the optimized conveyor belt system and found
and evaluating concepts, manufacturing and testing or imple-
that it could effectively separate municipal solid waste into
menting a product or service. While industrial design is
different categories with higher accuracy and efficiency com-
focused on the aesthetics of a product, such as its form,
pared to traditional conveyor systems. Huang et al. proposed
color, texture, structure, and feeling, product design encom-
an automatic sorting system based on programmable logic
passes a blend of marketing, product management, indus-
controllers (PLC) for waste separation. The authors devel-
trial design, and engineering characteristics [36]. According
oped an optimization method for the conveyor belt system
to Mazumdar, product development involves translating
used in the sorting process to improve the system’s effi-
customer needs into product design and manufacturing
ciency and accuracy. The study proposed a sorting algo-
processes, and managing dependencies throughout the prod-
rithm based on the weight, size, and color of the waste
uct life cycle, including design, manufacturing, distribution,
and optimized the conveyor belt speed and belt width. The
technical support, and disposal or recycling stages [37]. The
simulation results showed that the proposed system could
total design model is illustrated in Figure 2, which starts with
effectively separate different types of waste with higher accu-
market investigation to check the basic requirements, fol-
racy and efficiency compared to traditional conveyor belt
lowed by product design specifications (PDSs) that identify
systems.
components and sub-components of a new design.
Sorting systems utilize various methods such as screens,
In the design process, after the conceptual design is com-
air classifiers or separators, and NIR spectroscopy [32], [33].
pleted, a 3D CAD model is generated based on important
In the latter, infrared sensors detect the type of material on
factors determined in the previous step. The detailed design
the conveyor belt and sort it out accordingly, even checking
phase involves specifying the precise details of each compo-
each individual piece. Automated sorting technology offers
nent, such as materials, manufacturing processes, and phys-
several advantages, such as reducing the need for manual
ical properties required. At this stage, a reasonable cost is
labor and promoting order, making it an integral feature in
assigned for manufacturing, detailing, and selling the model.
waste management. In addition to waste management, con-
This design process is applicable to a range of engineering
veyor technology has become a crucial aspect of recycling
and consumer products, including those in the composite
plants. Its use in combination with tipping devices and press
field.
containers allows for convenient filling, while conveyor belts
can transport residual waste from reception to sorting and
disposal, optimizing everyday operations. 1) MARKET INVESTIGATION
However, due to their automated sorting capabilities, The primary objective of market research is to gather cru-
machines like these require minimal human workers. Apart cial information about the product of interest from multiple
from loading the garbage into the machine, there may be sources, including questionnaires, social media, the internet,
only a few other areas in the sorting process that require competitors, and advertisements. It involves conducting a
human workers. The machine handles most of the sorting market survey to determine customer preferences and eval-
process on its own. The sorting system proposed in this uate the performance of competitor products.
research is designed to assist small-scale waste separation
plants. Although human workers are still needed for various 2) PRODUCT DESIGN SPECIFICATION
tasks, their roles have changed. Instead of sorting plastic The formal document called product design specification is
bottles one by one, they may rotate to perform waste man- developed by the product design team, which includes infor-
agement planning, marketing, maintenance, route planning mation gathered from market investigation. This document
transportation, or customer service, thereby avoiding direct serves as a guide for the design process and can be adjusted
contact with the waste. accordingly. The product design specification is developed
based on information obtained from various sources, includ-
B. PRODUCT DESIGN AND DEVELOPMENT (PDD) ing published literature and competitor’s products. It covers
Product design and development (PDD) is a systematic pro- various aspects such as performance, safety, maintenance,
cess that involves generating ideas or concepts, developing environment, and materials.

VOLUME 11, 2023 47621

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

3) CONCEPTUAL DESIGN purchases. As customers become more informed, demanding,

In the product development cycle, conceptual design plays and sophisticated, Kansei may be the deciding factor. Six
a pivotal role as it transforms the requirements of the target types of Kansei Engineering are explained as follows.
customer into a tangible object. Quality cost, manufactura-
bility, and product life cycle characteristics are factors that 1) TYPE I: CATEGORY CLASSIFICATION
can influence the process of conceptual design. Additionally, The Category Classification method involves breaking down
conceptual design is primarily concerned with the generation the Kansei category of a planned target into a tree diagram,
of ideas and is often referred to as conceptualization, which which is then used to determine the physical design details.
is closely related to creativity. These sub-concepts serve as parameters for product design,
and in some cases, questionnaires are used to assist manufac-
4) DETAILED DESIGN AND PROTOTYPING turers in finalizing the design.
The detailed design phase is critical as it ensures that the
new product meets all functional requirements, cost targets, 2) TYPE II: KANSEI ENGINEERING SYSTEM
and manufacturing constraints. The design team should work The Kansei Engineering Type II is a computer-assisted sys-
closely with the manufacturing team to ensure that the design tem that is related to Kansei Engineering. The Kansei Engi-
is optimized for efficient production. Prototyping and testing neering System (KES) is a computerized system that uses an
may be necessary during this phase to ensure that the design Expert System to facilitate the transfer of customer emotions
works as intended and meets all requirements. The use of and perceptions into physical product design. The KE sys-
virtual simulation can help to reduce the time and cost of tem is comprised of four databases and an inference engine,
physical prototyping, as it allows for testing and refining the as illustrated in Figure 3.
design before a physical prototype is created. Overall, the
detailed design phase is essential in ensuring that the final 3) TYPE III: KANSEI ENGINEERING MODELLING
product meets all requirements and is ready for manufacture Kansei Engineering Type III involves using a mathematical
and launch into the market. model in a computerized system, in addition to a rule base.
The mathematical model functions as a form of logic and
5) MANUFACTURE is applied to define fuzzy sets that express the degree of
In addition to material properties and manufacturing pro- desirability of a product for customers.
cesses, other factors to consider in this stage include
production volume, lead time, and environmental impact. The 4) TYPE IV: HYBRID KANSEI ENGINEERING
selected manufacturing process should be able to handle the Hybrid Kansei Engineering is a blend of Forward and Back-
desired production volume efficiently and within the desired ward Kansei Engineering techniques. Kansei Engineering
lead time. Additionally, the environmental impact of the Type II is a computerized Kansei Engineering System that
chosen manufacturing process should be considered, includ- enables customers to determine the product that best suits
ing the amount of waste generated and energy consump- their product feeling. In Type II, the flow of Kansei translation
tion. In recent years, sustainable manufacturing processes is from the customer’s Kansei to the design details. Designers
have become more popular, as they prioritize minimizing can also use this approach to determine the design of a Kansei
waste and reducing energy consumption, leading to a reduced product from the perspective of a ‘‘market-in strategy’’. The
carbon footprint. The material and manufacturing process designer first settles on the design image of the new product
selection stage is critical in determining the quality and and then identifies the design specifications of the product
cost-effectiveness of the final product. with the support of the KES. This direction is known as ‘‘For-
ward Kansei Engineering’’. However, designers sometimes
C. KANSEI ENGINEERING want to assess the degree of fit between their image and
Kansei Engineering (KE) originated in Japan during the design and Kansei. Backward Kansei Engineering provides
1970s as a tool to link customers’ emotional responses computerized suggestions to designers on how to improve
to the product design process by converting feelings into their design beyond what Type II can do. Figure 4 illustrates
measurable and physical design specifications [38], [39], both Kansei translation directions in Kansei Engineering.
[40], [41], [42], [43]. The word ‘‘Kansei’’ refers to the instinc- The Hybrid Kansei Engineering System allows designers
tual mental activity of a person who perceives an impression to obtain design specifications from Kansei words through
from an external stimulus. When customers plan to purchase a Forward Kansei Engineering. With the system’s output,
product, they already have a preconceived positive or negative designers can fuel their creativity and design a new product
image in their minds. KE starts by observing customer behav- based on their own ideas and the system’s suggestions, which
ior, capturing their emotions and affections through psycho- translate customer requirements. Next, the designer inputs
logical scaling, and analyzing the emotions using various their sketch into the system, and Backward Kansei Engi-
statistical methods to create product or system design specifi- neering recognizes the sketch through an image recognition
cations, aiming to maximize customer satisfaction with their system [44].

47622 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

FIGURE 5. Translation of Kansei [44], [45]. Kansei Engineering Type I as

used in the Mazda case Nagamachi.

customers to design products that truly reflect their feelings

and preferences.

D. KANSEI ENGINEERING—CASE STUDY

The application of KE has been introduced to various types of
industries, one of the classic case studies of KE is shown in a
new sports car named ‘‘Miata’’ (Eunos Roadster in Japan) of
‘‘Mazda’’. This car was developed and derived from Kansei
FIGURE 3. Kansei engineering system [45]. Engineering [45]. The translation of Kansei (tree diagram)
into car physical traits in the case of ‘‘Miata’’ is shown in
Figure 5.
The key consideration in the car development case was
the ‘‘Human-Machine Unity (HMU) concept’’, which was
identified as the main heading at the first level of the tree
structure. HMU was represented by four sub-concepts: tight
feeling, direct feeling, speedy feeling, and communication
between the vehicle and the driver. To achieve the tight
FIGURE 4. Kansei translation directions in Kansei engineering [45].
feeling, the design team conducted a small Kansei Engineer-
ing experiment where participants voted for a body length
around 4 meters and only two seats. The gear shift lever was
5) TYPE V: VIRTUAL KANSEI ENGINEERING identified as the essential tool for the direct feeling, and a gear
Virtual Kansei Engineering is a technique that combines shift lever of 9 cm in length was selected as the best fit for the
Kansei Engineering and virtual reality technology. The sys- intended Kansei. To achieve the speedy feeling, the power
tem creates a virtual environment with a new product, which train team worked to shorten the time gap between pressing
is determined by Kansei Engineering, and invites customers the accelerator and feeling the car accelerating. Other parts
to test and provide feedback on the product within the vir- evaluated were the design of the engine, exterior and interior
tual space. This feedback is then shared on social media details, and the exhaust pipe. Overall, Kansei Engineering
platforms. The use of virtual reality technology in Kansei Type II using statistical data in combination with experiments
Engineering provides a cost-effective and efficient way to was applied to identify and translate the customer’s Kansei
gather customer feedback and enhance the product design into concrete design parameters and characteristics for the
process. car.

6) TYPE VI: COLLABORATIVE KANSEI ENGINEERING III. KEY CONSIDERATION OF THE PROPOSED RESEARCH
DESIGNING The focus of this study is on developing a conveyor belt that
Collaborative Kansei Engineering Designing or Internet can efficiently separate plastic bottles of both opaque and
Kansei Designing System is a Kansei Engineering Sys- transparent surface types. The design team aims to achieve
tem supported by the internet. With the increasing reliance this by applying an easy-to-access concept that will also
on the internet, this system facilitates the collaboration support maintenance activities. The success of the prototype
and exchange of ideas between the customer and designer, will be measured through reliability values to ensure that
resulting in faster product development. In this system, it meets the requirements of the company and the workers
customers can input their own Kansei words, and design- who will be handling the bottles for 8-hour tasks. The design
ers can use them to analyze and generate product designs must not only meet customer requirements but also consider
based on the customer’s input. This approach empowers cost, mixed-use development, and versatile application. To 6

VOLUME 11, 2023 47623

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

achieve this, the concepts of universal and accessible designs the machine and bottles, physical properties of the bottles,
will be applied. The subsequent sections will elaborate on conveyor system, controlling platform, sensor capabilities,
the addressed issues, which directly affect the design pattern, motor, and conveyor. The primary objective of the study was
function, and size of the separating system. The ultimate goal to develop a system that could help plastic recycling com-
is to develop a conveyor belt that not only efficiently separates panies and waste separation plants to quickly sort different
plastic bottles but also reduces injuries and diseases that may types of plastic bottles using this technology. The system was
arise from prolonged handling of the bottles. designed to include all necessary features and remain cost-
effective. The scope and limitations of the key considerations
A. MECHANISM are explained in the following topics.
The product’s mechanism should be capable of handling a
large volume of plastic bottles efficiently. The system must 1) CONTROLLER-PROGRAMMABLE LOGIC
operate seamlessly and ensure rapid separation of the bottles. CONTROLLER (PLC)
Cost-effectiveness is also a key consideration, ensuring that Since the primary aim of this study was to explore the feasi-
the product is affordable for small and medium-sized enter- bility of integrating different types of conveyors to perform
prises (SMEs) to use in their operations. a sorting activity. The researchers chose a PLC-controlling
platform as the reference system, which is commonly used
B. PHYSICAL CHARACTERISTICS OF THE BOTTLES by waste sorting plants as their existing conveyor system.
The proposed design focuses on two key issues: the trans- To achieve the goal, an additional conveyor system controlled
parency and size of the plastic bottles. To effectively sort and by Arduino was integrated with the existing system. The
separate the bottles, the design will utilize a light sensor to study sought to identify the factors necessary to successfully
differentiate between opaque and transparent plastic bottles, complete the task and create guidelines for further research.
with the opaque bottles being sorted into a separate location. By investigating the integration of different conveyor types,
The size of the bottle will also be taken into consideration, the study aimed to provide insights into the design and
with standard sizes such as 350-ml, 600-ml, 1-liter, and implementation of conveyor systems in waste sorting plants.
1.5-liter bottles being accounted for in the design process. The application of PLC (Programmable Logic Controller)
has become increasingly popular in automation, providing
C. MATERIAL OF THE CONVEYOR various common support functions, from small to large areas,
It is important that the material used in the conveyor for detecting, reporting, and recording data. Compared to the
system has sufficient strength to handle the weight of a Arduino, which requires each function to be written from
large volume of bottles and the movement of the mech- scratch [55], [56], the PLC offers a more comprehensive
anism. Selecting the appropriate type of belt conveyor solution. However, the bare hardware cost for the controller
is crucial to ensure safe and productive operation, with and I/O is considered, the Arduino is a more cost-effective
easy maintenance and minimal impact on the environ- option than the PLC.
ment [46], [47], [48], [49], [50], [51], [52], [53]. To ensure
that the product is cost-effective and of good quality, it is 2) CONTROLLER-ARDUINO
recommended to estimate the cost of using various materials The controller in the developed machine will work with
in the design before starting the manufacturing process. devices such as sensors and motors to complete the function
of the conveyor system. This study utilized the versatile
D. SIZE OF THE CONVEYOR Arduino platform, which offers a wide range of processors,
Conveyor length and width (size) are determined based on sizes, and connectivity features. The Arduino software serves
product size and weight, and the load capacity of the system as an Integrated Development Environment (IDE) for all
must be sufficient to handle the weight of the products with- boards. The board is equipped with sets of digital and analog
out malfunction. Space is important as the system needs to fit input/output (I/O) pins that can interface with various expan-
comfortably in the designated area and be easily adjustable. sion boards, breadboards, and other circuits. The boards also
The height and size of the conveyor belt should be suitable feature serial communication interfaces, including Universal
for worker ergonomics and not exceed the standard height Serial Bus (USB) on some models, for loading programs from
of Asian individuals [54]. The sensor’s placement should be personal computers. The microcontrollers are programmed
strategically positioned to ensure accurate detection of the using a dialect of features from the programming languages
bottles regardless of their orientation. C and C++ [57], [58], [59], [60]. The Arduino Mega 2560,
based on the ATmega2560 microcontroller, is used in this
E. SCOPES AND LIMITATIONS study. It has 54 digital I/O pins (15 of which can be used
During the initial design phase of the study aimed at creat- as PWM outputs), 16 analog inputs, 4 UARTs (hardware
ing an efficient conveyor and automation system for sorting serial ports), a 16 MHz crystal oscillator, a USB connection,
plastic bottles, the researchers established certain limita- a power jack, an ICSP header, and a reset button. It contains
tions and boundaries. These limitations included the size of everything needed to support the microcontroller, and it can

47624 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

be connected to a computer with a USB cable or powered with TABLE 1. Types of sortation conveyor [63].
an AC-to-DC adapter or battery to get started.

3) ARDUINO FOR THE STUDY

The Arduino board is cost-effective and suitable for a variety
of microprocessors and controllers. The Arduino hardware is
user-friendly and available at varying prices depending on the
model. The Arduino platform is widely used and popular in
education, electronics automatic control, and robotics. It can
be quickly and easily learned, and users can apply it to var-
ious experiments and applications to demonstrate concepts
learned in class. The Arduino also provides interesting teach-
ing tools and an easy-to-learn platform for code writing and
control concepts. The Arduino is suitable for small businesses
where equipment cost is the main concern.

4) SENSOR
The sensor used in this system will detect the type of bottle by
utilizing the light intensity properties. Proper adjustment of
the distance between the bottle and the sensor is crucial, as the
sensor will check the light intensity after exposing it directly
to the body of the bottle [61], [62], [63]. In cases where the
sensor cannot detect the light intensity, the rejector will not
be able to separate opaque plastic bottles from transparent
plastic bottles.

5) MOTOR
The motor’s speed, position, and acceleration are interrelated,
as the speed is the derivative of position, and acceleration is
the derivative of speed. Even if only one of them is known,
it is possible to determine all three factors [63]. In a movement
system that separates the correct type of bottle, the motor’s
speed is the main parameter. To minimize errors during task
performance, the controller’s program input should be cor-
rectly and concisely formatted. In this study, a DC motor is
used due to its simpler installation and maintenance, high
startup power and torque, fast response times to starting, stop-
ping, and acceleration, and availability in several standard
voltages. DC motors generally have higher efficiency and
make better use of their input energy. Using a microcontroller
like Arduino to control the DC motor of the conveyor belt
system is a cost-effective solution compared to other meth-
ods. However, AC motors are generally more powerful than
DC motors because they can generate higher torque by using a
more powerful current. Both AC and DC motors are available
in various sizes and strengths that can meet any industry’s
power requirements.

6) CONVEYOR FIGURE 6. Conveyor sortation systems [66].

Conveyors are mechanical devices used for the easy move-

ment of objects, items, or packages. They typically consist of
frames that support rollers, wheels, and belts, which can be systems that use continuous belts to convey objects. Effi-
powered by motors or operated manually [64], [65]. High- cient and effective material handling requires the appropriate
friction belts can be applied to prevent product slippage. selection of belt type and shape for belt conveyors. Loca-
Belt conveyors are commonly used as material handling tion and positioning of the conveyor within the production

VOLUME 11, 2023 47625

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

FIGURE 7. Seven steps are required to accomplish the task.

facility must also be considered during the design process. and prevent breakdowns. However, the benefits of using a
AC or DC motors can be used to power belt conveyors, conveyor system, such as increased efficiency, reduced labor
directly or through reduction gears, chains, or sprockets, and costs, and improved safety, often outweigh the initial and
control systems can range from simple on/off switches to ongoing expenses. It is important for businesses considering
advanced variable frequency drives that regulate motor speed the installation of a conveyor system to carefully evaluate
and acceleration. their needs and budget to determine whether the benefits
Table 1 and Figure 6 show the types of sortation convey- justify the costs. It may also be helpful to consult with experts
ors used for merging, identifying, inducting, and separating in the field to ensure that the system is designed and installed
products that are conveyed to specific destinations. properly and that maintenance costs are minimized over the
long term.
7) CONVEYOR SELECTION
The belt conveyor’s crucial specifications [67], [68], [69] 8) SORTATION CONVEYORS
pertain to the type and shape of the belt, conveyor size (length Sortation conveyors are designed to separate and divert prod-
and width), load capacity, location, and electrical supply. ucts to different lanes based on size, weight, and other cri-
These requirements must be carefully considered and met teria, and are commonly used in shipping, palletizing, and
to design an alternative belt direction. The selection of a packing stations. The specific design of a sortation system
conveyor depends on various factors, such as product type, will depend on the needs of the production facility, including
speed, elevation change, and industry focus. Belt conveyors the type of products being transported and the desired level
come in various sizes, ranging from foot-long units used for of sorting accuracy. To sort and divert products based on size,
packaging lines to mile-long systems for mining operations. weight, and other criteria, sortation conveyors are an excellent
Typically, AC and DC motors power them, either directly or solution. They are frequently used in shipping and palletizing
through reduction gears, chains, or sprockets. Certain convey- operations, as well as packing stations [71], [72]. They have
ors are designed to locate products precisely between manu- been used as the initial platforms for model analysis in sub-
facturing operations. The controller system for the conveyor sequent processes. Typically, two types of surfaces are used
can be a simple on/off switch or a variable frequency drive to make bottles: transparent and opaque.
that regulates the motor’s speed and acceleration. Several
additional control options are available, and long belt con- IV. RESEARCH METHODS
veyors transport has been introduced [70]. To implement the proposed approach, seven main steps are
Conveyor systems are typically used in large-scale indus- necessary, which integrate three methods: Product Design
trial operations, and they can be costly to install and maintain. and Development (PDD), Kansei Engineering (KE), and
The cost of equipment and materials required for construct- Maintenance Engineering (Failure Analysis). These steps are
ing conveyors can vary widely depending on the type of illustrated in Figure 7, and each of them is accompanied by a
conveyor, the length and complexity of the system, and brief description.
other factors such as the materials used in the construction. A market survey was conducted to develop a conveyor for
Maintenance costs for conveyor systems can also be signif- separating types of plastic bottles in order to meet customer
icant, as regular cleaning, lubrication, and replacement of requirements. Data was collected using a questionnaire,
parts are necessary to keep the system running smoothly which was distributed through online, offline, and interview

47626 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

FIGURE 8. Conceptual model of 5 types of conveyor (a) One-Way splitter (b) Two-Way splitter (c) Falling bridge splitter (d) Robot arm (e) Gravity conveyor.

TABLE 2. Descriptions of conceptual models.

methods. The questionnaire was designed to determine development of an innovative separating conveyor system
parameters such as the desired material, size, and price of that could be marketed to recycling SMEs. To this end, a mar-
the system that would satisfy customer requirements. A total ket survey was conducted to identify customer requirements,
of 46 participants completed the questionnaire. The results and existing products on the market were reviewed. After
showed that 100% of the participants considered the bot- conducting the necessary research, the researchers were able
tle separation machine to be important for improving their to create five conceptual designs for the proposed separating
business. conveyor system. These designs include the One-Way splitter,
Two-Way splitter, Falling bridge splitter, Robot arm, and
Gravity conveyor. Figure 8 illustrates the designs visually,
A. STEP 1: CONCEPT DEVELOPMENT
and details on each design are provided in Table 2.
The researchers undertook a comprehensive analysis of the
challenges faced by recycling SMEs in separating clear and
turbid plastic bottles, with the aim of generating ideas for B. STEP 2: KANSEI ENGINEERING
the proposed separating conveyor system. Currently, these In this stage, the Kansei Engineering method is utilized to
companies spend significant time and resources on this meet the customer’s needs. Before beginning the KE process,
task. To address this issue, the researchers proposed the the approach’s scope is defined, and adjective words are

VOLUME 11, 2023 47627

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

identified. The KE method is sensitive to the fact that people TABLE 3. Product characteristics and customer’s requirements of ‘‘The
Input Phase’’.
from different cultures and ages think differently, and the
designs created by western or European platforms may not
be suitable for Asian characteristics. The KE results will pro-
vide a conceptual design that meets customer requirements.
After analyzing the questionnaire, the 3F’s (Form, Fit, and
Function) were used to generate the conceptual design. The
conveyor belt kit must be capable of automatically separating
transparent and opaque plastic bottles and accommodating
bottles of any size. It should also be suitable for placement
in a recycling plant. The KE process is employed to create
the conceptual design that meets customer requirements by
first generating a set of questionnaires and then distributing
them to the target customers. The obtained results are valu-
able for factor analysis in the subsequent process, where the
draft design of a new product can be created systematically.
The processes required to accomplish the KE task can be
expressed as:

1) PLAN AND DESIGN QUESTIONNAIRE

This research targeted small recycling companies and aimed
to improve their bottle recycling process. The study utilized a
questionnaire completed by 46 participants who were directly
involved in the process of recycling bottles. The questionnaire
was divided into three main parts: input, processing, and
output. The study focused on developing a conveyor to assist
the bottle-separation unit and considered three main phases:
input, processing, and output. To design the conveyor, the
product characteristics of each phase were described using
adjective words obtained from the questionnaire responses.
These adjective words were used to draft the conveyor design be provided to elevate workers to the appropriate height.
in the subsequent process. Alternatively, the parts could be delivered to the conveyor at a
height that is more convenient for workers to load or unload.
2) IDENTIFY THE ADJECTIVE WORDS It is crucial to ensure that the conveyor’s height is appropriate
To identify suitable adjectives that align with customer for the specific application and the workers who will be using
requirements, a brainstorming session and customer inter- it. Doing so can help improve efficiency and safety in the
views were conducted. The list of potential adjectives was workplace.
then categorized into the three main phases previously men-
tioned, and can be found in Tables 3 to 5. To facilitate the 3) ADJECTIVE WORDS WITH 7-POINTS SEMANTIC
systematic transfer of details, the researchers attempted to DIFFERENTIAL SCALE
translate these keywords into a technical-design platform. The researchers utilized a questionnaire that combined the
This process allowed for a comprehensive understanding 7-point semantic differential scale with adjective words to
of the desired product characteristics for each phase of the directly measure customer requirements based on their opin-
recycling process and informed the subsequent design of the ions. An example of the 7-point semantic differential scale
conveyor system. Furthermore, the technical-design platform for system adjectives is depicted in Figure 9.
ensured that the details obtained from the questionnaire and
interviews were accurately incorporated into the conveyor’s 4) ANALYZE ADJECTIVE WORDS WITH 7-POINTS SEMANTIC
design. DIFFERENTIAL SCALE
The design stage of the project requires careful considera- The responses of the 46 participants were analyzed using the
tion of various factors, such as the height of the conveyor for SPSS program. The analysis revealed three main factors that
input parts. If the conveyor’s height is below 80 cm, it may needed to be taken into consideration.
create challenges for loading or unloading the parts, causing
ergonomic issues for workers who have to bend or crouch a: KAISER-MEYER-OLKIN MEASURE OF SAMPLING
down to work with the conveyor. To address this issue, the ADEQUACY
conveyor’s height could be adjusted to a suitable level. If the To ensure the validity of the data for Factor Analysis, the
conveyor cannot be adjusted, a platform or step stool could Kaiser-Meyer-Olkin Measure of Sampling Adequacy was

47628 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

TABLE 4. Product Characteristics and Customer’s Requirements of ‘‘The TABLE 5. Product Characteristics and Customer’s Requirements of ‘‘The
Processing Phase’’. Output Phase’’.

FIGURE 9. The example of 7-points semantic differential scale of system

adjective.

employed. A value of 0.5 or greater indicates that the data

is reliable and suitable for Factor Analysis. The results of this
test are shown in Figure 10.

b: BARLETT’s TEST OF SPHERICITY

FIGURE 10. Kaiser-Meyer-Olkin Measure of Sampling Adequacy and
Barlett’s Test of Sphericity is a statistical test used to deter- significant results.
mine whether a correlation matrix is significantly different
from an identity matrix, indicating that the variables included
in the analysis are interrelated and can be used in factor From Barlett’s Test of Sphericity result, the significance
analysis. Barlett’s Test of Sphericity result is applied to deal value is less than 0.05, so H0 is rejected. This can be con-
with the hypothesis test. For this test the hypothesis is shown cluded that the Correlation Matrix ̸= 1.
as:
c: ROTATED COMPONENT MATRIX
H0 : Correlation Matrix = 1 In this study, the rotated component matrix was used to
H1 : Correlation Matrix ̸ = 1 determine which factor should be grouped together. This

VOLUME 11, 2023 47629

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

5) CUSTOMER SURVEY
To understand the needs and preferences of target customers,
a questionnaire was designed that included five different
conveyor designs. The respondents were initially uncomfort-
able with answering technical questions, so the design team
provided them with pictures and descriptions of the designs
to aid their understanding. The customers expressed concerns
about integrating new technology with their traditional way
of working and requested an easy-to-use platform that could
operate the entire system with the press of a single button,
similar to a washing machine. This requirement was identi-
fied as a key design consideration and was incorporated into
the system-level design stage by breaking it down into main
and sub-components.
FIGURE 11. Scree plot. To minimize costs and make the new system more user-
friendly, the researchers integrated the new platform with the
existing PLC controlling platform that was already in use at
was determined based on the factor loading of each factor, the plant. Based on the survey results presented in Table 6,
where the factor with the highest loading was placed in its the one-way splitter and two-way splitter designs were the
respective group. The scree plot was used to display the top two choices among respondents. To create the proposed
Eigenvalue on the y-axis and the component number on separation system, these two designs were combined using
the x-axis (Figure 11). It is important to note that when the Kansei Engineering to determine specific details.
Eigenvalue is greater than 1, the corresponding component
number is considered and recorded. 6) GENERATE CONCEPTUAL DESIGN (3F’S)
A scree plot is a graphical representation of eigenvalues The 3F’s method, including form, fit, and function, was used
(y-axis) against the number of factors (x-axis). It displays a to develop the conceptual design based on the results of the
downward curve, and the point where the curve’s slope flat- customer survey. The proposed design of the conveyor belt
tens indicates the optimal number of factors to be generated system was developed by applying the concepts of the waste
by the analysis. However, both criteria used to determine the conveyor and baggage scanner, as shown in Figure 13.
number of factors can sometimes result in an unreasonably
high number of factors. In this study, the scree plot suggested
three or five factors due to the two points where the slope lev- a: FORM
els off. The aim of factor analysis is to reduce many variables This refers to the physical appearance of the system. This
that describe a complex concept into a few interpretable latent includes considerations such as the overall shape and size of
factors that explain the maximum amount of variability in the system, the color and finish of the materials used, and
the data. In this study, the design of the innovative separating how the system fits into the environment where it will be
conveyor for different types of plastic bottles was separated installed. When applying the ‘‘Form’’ concept in real-world
into four groups, as shown in Figure 12. design, the goal is to create systems that not only function
Group 1: Function efficiently but also have an attractive appearance that blends
• The output part height should be lower than the input
well with their surroundings. This is especially important in
part height. public spaces like waste management facilities, where the
• The height of the input part must be lower than the
appearance of the equipment can impact people’s perception
human waist, which is typically around 80 cm. of the facility and its activities. To achieve this, the design
• The length of the conveyor must not be adjustable.
team has prioritized choosing a color scheme that aligns with
• The function of the system is limited to only separating
the facility’s branding or environment. By carefully selecting
plastic bottle types. colors that complement the surroundings, the equipment can
blend seamlessly into its environment, creating a cohesive
Group 2: Physical Characteristic
and harmonious aesthetic. Furthermore, the team has incor-
•The machine should be lightweight.
porated design elements that enhance the system’s visual
•The conveyor’s width should be broad.
appeal and user-friendliness, such as illuminated panels or
Group 3: Usability
user interfaces for operators. These elements not only add to
• The button should be used to control the machine.
the system’s aesthetic appeal but also improve its function-
• Additionally, the machine should possess a high capac-
ality and ease of use. By focusing on the ‘‘Form’’ concept,
ity to handle weight. designers can create systems that are both functional and
Group 4: Separation Method aesthetically pleasing, making a positive impact on the user’s
• The separation method should be horizontal. experience and perception of the facility.

47630 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

FIGURE 12. Rotated component matrix.

b: FIT
In designing a new conveyor belt for a waste separating
plant, the ‘‘Fit’’ concept is of utmost importance. The con-
veyor belt kit must be designed to seamlessly fit into the
plant’s working area. It is essential that the conveyor belt can
be easily adjusted for height and length to accommodate a
range of plant sizes. Figure 14 depicts a real working space
where the separating conveyor system is installed, empha-
sizing the importance of proper fitting. By prioritizing the
‘‘Fit’’ concept, the design team can ensure that the conveyor
belt fits perfectly into the plant’s working area, enabling
efficient waste separation and maximizing the plant’s over-
all performance. A well-fitted conveyor belt system reduces FIGURE 13. The drafted design of the proposed system.

the risk of jams or malfunctions, enhancing the system’s

overall reliability and reducing downtime. In conclusion,
the ‘‘Fit’’ concept plays a crucial role in designing a con- on their varying light intensities. This is based on the principle
veyor belt system for a waste separating plant. By ensuring that as light passes through a material, such as a plastic bottle,
proper fitting and adjustability, the design team can opti- its intensity decreases. To test and experiment with light
mize the system’s performance and reliability, ultimately intensity, a controlled environment box was designed. Three
leading to a more efficient and effective waste separation light sensors (LDR) were connected to a microcontroller
process. to collect light intensity data in three different situations:
clear plastic bottles, turbid plastic bottles, and opaque plastic
bottles (as shown in Figure 15). Two types of control modules
c: FUNCTION were utilized: a programmable logic controller (PLC) for the
The machine’s effectiveness relies on its ability to accurately ‘‘opaque lane’’ and a microcontroller for the ‘‘transparent
and efficiently separate different types of plastic bottles based lane.’’ The advantages of these two control modules were

VOLUME 11, 2023 47631

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

TABLE 6. The result from questionnaire.

FIGURE 16. Conveyor belt connected to PLC [79].

C. STEP 3: SYSTEM-LEVEL DESIGN

The conveyor belt kit comprises three main modules or
components: the delivering module, controlling module, and
detecting module. These modules and their respective com-
ponents are illustrated in Figure 17.
The proposed design aims to develop a waste separating
conveyor for plastic bottles. As such, preliminary studies
were conducted to explore ways to separate different types of
bottles. The main modules of the conveyor will be discussed,
including the use of light sensors to identify transparent
and opaque bottles. The input task involves placing empty
bottles onto the conveyor platform in a single-line sequence.
However, this platform cannot support a large volume of
bottles being thrown onto the collection area simultaneously.
To address this issue, ‘‘an extra shaking bed’’ may be
required at the initial stage of ‘‘the delivering module’’ to
FIGURE 14. Examples of the areas for setting-up conveyor: multi-bended organize a large quantity of bottles into a specific order,
style (left), and straight pattern (right) [77], [78].
ensuring they are arranged in sequence before being sorted
by the light sensors.
The proposed conveyor belt kit consists of three main
modules:
• The delivering module is responsible for the delivery of
plastic bottles (see Table 7 ).
• The detecting module is responsible for detecting and
separating the types of plastic bottles (see Table 8 ).
• The controlling module is responsible for controlling the
conveyor belt kit’s system components, such as motors,
rejectors, and sensors (see Table 9 ).
FIGURE 15. Function’’ of the developed system with 3 types of bottles.
Efficient and accurate separation of plastic bottles accord-
ing to their type is achieved through the seamless integration
documented and analyzed to provide guidance for waste of the three modules. The modules work together to ensure
separating plant manufacturers to identify and address issues the optimal performance of the separation process, resulting
without resorting to trial-and-error processes. The prototype in high-quality plastic that can be recycled efficiently.
created can assist in reducing potential failures during the The delivering module is responsible for organizing and
separating process in a real situation. Figure 16 displays delivering the bottles onto the conveyor platform, while the
an example of a commercial conveyor belt kit with counter controlling module is responsible for separating the bottles
sensors controlled by a PLC. based on their type using light sensors and controlling the

47632 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

FIGURE 17. The module and its components.

TABLE 7. Delivering module, its component and definitions. TABLE 8. Controlling module, its component and definitions.

speed of the conveyor belt. The detecting module is respon-

sible for detecting and counting the separated bottles.
The detecting module was designed to detect the separated TABLE 9. Detecting module, its component and definitions.
bottles and count them. The light sensors in the detecting
module are responsible for measuring the light intensity pass-
ing through the plastic bottle. The microcontroller processes
the data from the light sensors and determines the type of
plastic bottle. The signals from the microcontroller are sent
to the PLC in the controlling module to adjust the operation
of the separating platform. Counter sensors were used to
count the number of bottles passing through each lane. The
counting information was sent to the controlling module for
further analysis and processing.
The controlling module was designed to control the speed
of the conveyor belt based on the output signals from the light
sensors. The programmable logic controller (PLC) was used
to control the opaque lane, which is the lane where opaque
bottles are separated. The PLC receives input signals from
the light sensors and determines the type of bottle passing
through the lane. The microcontroller was used to control The cost of all the necessary components for all modules
the transparent lane, which is the lane where transparent amounts to roughly 14,784 THB or 428.74 USD.
bottles are separated. The microcontroller receives input sig-
nals from the light sensors and controls the speed of the D. STEP 4: DETAILED DESIGN
conveyor belt based on the type of bottle passing through the Motor plays a crucial role in the operation of the conveyor
lane. system, providing the necessary power to move the belt

VOLUME 11, 2023 47633

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

and transport products along the production line. A motor between clear and opaque plastic bottles based on the dif-
is an electrical device that converts electrical energy into ference in light intensity passing through each type. Clear
mechanical energy. In this case, a motor is being used to drive plastic bottles allow more light to pass through them, while
the belt conveyor. A 24 VDC (Volts Direct Current) motor opaque bottles block more light. By measuring the amount
has been selected over a 12 VDC motor, as it is capable of of light that passes through the bottle using the LDR, the
providing more power to drive the conveyor. By selecting a system can differentiate between the two types of bottles.
more powerful motor, the system is able to drive heavier loads This type of sensor is commonly used in manufacturing and
on the conveyor. This is because a higher power motor is able packaging processes, where it is necessary to sort objects
to provide more torque, or rotational force, to the conveyor based on their properties. The LDR can be used in con-
belt, allowing it to move heavier objects along the production junction with other sensors and automated systems to sort
line. The selection of a motor with a specific voltage rating is objects based on their transparency or other physical charac-
dependent on the power requirements of the system and the teristics, helping to improve efficiency and accuracy in these
load that needs to be moved. In this case, a 24 VDC motor was processes.
chosen as it was deemed to have sufficient power to drive the A rejector is a mechanical device used in manufacturing
conveyor with heavier loads. and packaging processes to separate or remove defective
Conveyor belt is a modern piece of transportation and or unwanted products from a conveyor or production line.
handling equipment that is widely used in various industries. A rejector is being used to push clear or turbid plastic from
It consists of a continuous loop of material, typically made the conveyor to separate these two types of plastic. The
of rubber or plastic, that moves along a series of rollers or rejecter is typically equipped with an actuator that generates a
pulleys. The conveyor belt is being used to transport materials mechanical force to push the defective or unwanted products
from one place to another, helping to automate the process off the conveyor or production line. In this case, the rejecter
and save time. The use of conveyor belts in manufacturing is being used to separate clear and turbid plastic by pushing
and distribution facilities has become increasingly common the turbid plastic off the conveyor while allowing the clear
as it allows for the efficient movement of materials and plastic to continue along the production line. The decision to
products without the need for manual labor. This can result reject or separate the turbid plastic is likely made by a sensor
in increased productivity and lower costs for businesses. or detector, such as the light dependent resistor (LDR) men-
Conveyor belts come in a variety of shapes and sizes and tioned earlier, that detects the level of light passing through
can be customized to suit the specific needs of the industry. the plastic bottle. If the plastic is determined to be turbid
They are used to transport a wide range of materials includ- and therefore unsuitable for the intended use, the rejecter is
ing food products, construction materials, and industrial activated to push it off the conveyor and out of the production
parts. Conveyor belts are a vital part of modern transporta- line. The rejector is an important tool in manufacturing and
tion and handling equipment, providing a safe and efficient packaging processes as it helps to ensure the quality and
means of transporting materials and products across various consistency of the final product by removing defective or
industries. unwanted items from the production line.
Proximity sensors are commonly used in automation and
manufacturing processes to detect the presence or absence E. STEP 5: PROTOTYPING
of objects in their proximity. There are different types of 1) FUNCTIONAL DESIGN
proximity sensors, but one common type is the photoelectric The functional design was developed based on the main
sensor, which uses a light source and a receiver to detect the requirements and preliminary study, which included the use
presence of an object. A photoelectric sensor is being used of a conveyor belt connected to a programmable logic con-
as a counter sensor. The sensor consists of a laser and a light troller (PLC) and counter sensors to monitor the motor’s
dependent resistor (LDR) which receives the light from the operation, the components of the conveyor belt kit, PLC pro-
laser. In its normal stage, the LDR receives the light from gramming, and electronic connections. The proposed model
the laser, and the sensor output indicates that there is no is designed to separate different types of plastic bottles based
object present. However, when an object passes by and blocks on their varying levels of light intensity, as explained earlier.
the light from the laser, the LDR receives less light, caus- Figure 18 provides an overview of the functions of the pro-
ing a change in its resistance, and the sensor output detects posed model, which includes the following:
that there is an object present. This change in the sensor Bottle detection: The system detects the presence of a
output can be used to trigger a counter or other automated bottle on the conveyor belt.
process. Light intensity measurement: The system measures the
A light dependent resistor (LDR), also known as a pho- light intensity passing through the bottle.
toresistor, is a type of sensor that changes its resistance based Bottle separation: The system separates the bottles into
on the amount of light it receives. It is commonly used in a different lanes based on their light intensity levels.
variety of applications where the detection of light levels is Rejector: The system identifies and rejects any non-targeted
required. An LDR is being used to detect the level of light materials or bottles that do not meet the specified light
passing through a plastic bottle. This allows differentiation intensity level.

47634 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

FIGURE 20. Dimensions of the drafted conveyor system.

FIGURE 18. Functional design diagram.

FIGURE 19. The drafted design of conveyor system.

Data collection: The system collects and stores data related FIGURE 21. The prototype of the separating system proposed in this
to the bottle separation process for future analysis and opti- research.
mization.
The functional design of the proposed model incorporates
the knowledge gained from the main requirements and pre-
liminary study, with the aim of developing an efficient and
accurate bottle separation system.
The functional design consists of four steps: first, the user
activates the conveyor belt kit’s control panel; second, the
microcontroller sends a signal to the motor, which moves
the bottle to the sensors’ area; third, the sensors detect the
light intensity passing through the plastic bottle to determine
its type and send the data to the microcontroller; fourth, the FIGURE 22. Two lanes of the separating system.
microcontroller sends a signal to the rejector to accept or
reject the plastic bottle.
as well as clear labeling and instructions to make it simple
2) DRAFTED DESIGN for users to operate the system. The design draft prioritizes
In addition to the three main modules, the design draft also user-friendliness and ease of maintenance, which should help
includes adjustable legs that can be used to customize the to improve the system’s overall efficiency and effectiveness
height of the system to suit different users (as shown in in separating waste.
Figures 19 and 20). The design team made accessibility a
priority during the design process, ensuring that the sys- 3) FINAL PROTOTYPE
tem is easy to use and maintain. This includes features The physical prototype (Figure 21 to 23) developed in
such as easy access to internal components for maintenance, this research differs slightly from the drafted design, as its

VOLUME 11, 2023 47635

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

FIGURE 23. The opaque lane controlled by PLC. FIGURE 26. Position of proximity sensors and light sensor.

FIGURE 24. Separating different types of bottles: (a) Truly transparent,

(b) Partly transparent, and (c) Opaque.

FIGURE 27. Online mode with open access (online) platform.

FIGURE 25. Loading the bottle to the conveyor and starting the
separating process.

purpose was to demonstrate the functionality of the separating

system and not to create a full-scale version. As a result, some FIGURE 28. Adjust the alarm (based on the number of bottles) to activate
the notification light.
parts or components, such as the conveyor belts, lighting hall,
and collecting tray, were modified to fit the smaller scale of
the platform. sensors, the system can accurately count the number of bot-
tles passing through, enabling efficient separation of plastic
F. STEP 6: TESTING AND REFINEMENT bottles according to their type.
1) WORKING SYSTEM For the key consideration: the working system is designed
The machine operates in two main processes: the separating to be reliable, efficient, and easy to use.
process and the counting process. To initiate the system, Additionally, in order to count the bottles in the transparent
a bottle is placed onto the conveyor belt, and the start switch is lane, two modes were implemented: online and offline modes,
pressed, as demonstrated in Figure 24. During the counting which work simultaneously.
process, three proximity sensors are used to detect bottles. The online mode utilizes a free online tool called
Proximity sensors No. 1 and No. 2 are installed on the trans- www.anto.io. This tool allows the user to command and
parent lane, while proximity sensor No. 3 is installed on the assign specific parameters, such as the number of bottles
opaque lane, as shown in Figure 25. to be counted or an alarm to trigger a notification light.
Figure 26 illustrates the layout plan for the working space. To access this tool, the user must enter a username and
The transparent lane and the opaque lane are clearly demar- password and select the dashboard to assign the desired task,
cated, and the proximity sensors are strategically positioned as shown in Figure 27. The controlling platform is created
to ensure accurate detection of the bottles. By using proximity (Figure 28).

47636 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

FIGURE 29. Screen of developed system.

FIGURE 32. Graph of light intensity versus time (millisecond).

of plastic bottles with a high degree of accuracy. Figure 32

presents a graph that clearly illustrates the difference in light
intensity readings between the three types of plastic bottles.
The controlling platform displayed the graph, which was
plotted based on the recorded data.
FIGURE 30. Three light intensity sensors for separating bottles.
A turbid bottle is one through which light can pass from the
top to the bottom, but not uniformly. The surface of the bottle
is not completely transparent, which results in some areas
being less transparent than others. In contrast, a dirty bottle
may have areas where light can pass through uneventfully,
but there may be other areas where the surface is partially or
completely obscured by dirt, dust, or other materials.
An opaque bottle is one that does not allow any light to
FIGURE 31. Three experiment settings for separating transparent (left), pass through its surface. Unlike turbid and dirty bottles, which
turbid (middle), and opaque (right) bottles. allow some light to pass through, an opaque bottle completely
blocks the transmission or reflection of light.
On the other hand, in the offline mode, the counting num-
ber is continuously displayed on the screen (Figure 29) while 3) FAILURE RATE
the machine is running. Two versions of the conveyor belt To ensure that a developed system is operational and ready
for separating types of plastic bottles were developed due to for use, a crucial aspect to consider is failure analysis. Fail-
the bottleneck of the Arduino UNO’s capabilities. The second ure is defined as the inability to meet an expectation or a
version, which uses the Arduino MEGA, is equipped with two lack of success. Failure analysis is performed for various
delivering data boards (RS485) for more stable data delivery reasons and factors, such as determining the failure rate,
to the ESP32 board, which is used to send the data to the which can help calculate the system’s reliability using an
server for real-time online counting. exponentially decaying probability function. Additionally,
system availability can be affected by planned and unplanned
2) LIGHT INTENSITY TEST downtimes, such as equipment failure, which results in a loss
To separate bottles based on their type, three light inten- of availability.
sity sensors were employed to measure the amount of light These factors can all contribute to the overall issue of bottle
passing through three different types of bottles: transpar- sorting errors. To address these issues, it may be necessary
ent, turbid, and opaque. These sensors were strategically to conduct a thorough analysis of the system to identify
placed such that sensor S1 was positioned at the leftmost the root cause of the problem. This may involve examining
part of the bottle, sensor S2 was situated in the middle, and the components and systems involved in the sorting process,
sensor S3 was placed at the rightmost part. The positions including the rejector, LDR sensor, LED light bulb, motors,
of these sensors are depicted in Figure 30, while the exper- conveyor belts, power supplies, and controlling system. Once
iment settings for the three types of bottles are illustrated in the root cause of the problem has been identified, appro-
Figure 31. priate measures can be taken to fix the issue and improve
The readings from the 3 sensors were reported as three the overall efficiency and accuracy of the sorting process.
distinct lines, and the different types of plastic bottles exhib- This may involve repairing or replacing faulty components,
ited varying sensor readings. Therefore, based on the find- adjusting settings or configurations in the controlling system,
ings of the study, it was concluded that the sensors used or redesigning certain aspects of the conveyor system to
were effective in differentiating between the various types improve performance.

VOLUME 11, 2023 47637

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

TABLE 10. Steps and equations required for calculating ‘‘Failure Analysis’’ [85], [86], [87], [88].

Mean-time-to-repair (MTTR) is a useful maintenance met- machine by depicting the end events. The fault tree diagram
ric that can measure the average time required to troubleshoot is generated through experimentation.
and repair failed equipment, indicating how quickly an orga- The Fault Tree Analysis (FTA) simplifies complex system
nization can respond to breakdowns and fix them. failures into a single failure event. The reliability of the
Another important factor that affects system availability is machine is determined by combining the failures of multiple
maintainability, which refers to the speed at which techni- components, connected through ‘‘AND/OR’’ gates. These
cians can detect, locate, and restore asset functionality after gates help identify how the failure of a single or multiple
downtime. When building a system, considering availability components can cause the top event to fail, and have their
during all design and construction phases can help the design own formula for calculating the machine’s reliability.
team plan for component failures and how to bring failed The failure event that occurred during the experiment can
items back to normal. It is crucial to note that unrecoverable be separated into four main factors; inadequate separation
(hard) failures are critical compared to recoverable (soft) opaque and transparent bottle, poor conveyor design, inac-
failures. curate data obtained from controlling system, and inadequate
The reliability and failure rate of each component in a counting bottle. From the fault tree diagram, the reliability
machine can be determined by considering various factors and failure rate for each component of the machine can be
such as Mean Time to Repair (MTTR), Mean Time to Failure determined by calculating from Mean time between failure
(MTTF), and Mean Time Between Failures (MTBF). Math- (MTBF), Mean time to repair (MTTR) and Mean time to
ematical models and parameter considerations are typically failure (MTTF) (Table 10 to 12). After the failure rate and
used to analyze significant failure events and to show how reliability are calculated, the ‘‘AND/OR’’ gate is used to link
the components are interconnected using ‘‘AND’’ or ‘‘OR’’ every end event together to calculate the main failure event.
gates. Once the errors have been identified, efforts can be The gate calculation is very important because the calculation
made to enhance the machine’s performance to meet the between AND gate and OR gate is differing and each event
desired standards. also have differed gate too.
Causes that make the conveyer belt kit fail can be classified
into 4 cases. The cases were come from;
4) FAULT TREE ANALYSIS (FTA) Case 1 - Sub-system ①: The inadequate separation of
In order to analyze the reliability and safety of engineering opaque and transparent bottles can be attributed to rejector,
systems, Fault Tree Analysis (FTA) is utilized (as depicted LDR sensor, or LED.
in Figure 33). The diagram presents a fault tree with a spec- Case 2 - Sub-system ②: A poor conveyor design may result
ified fault event, which is used to determine the machine’s from issues with the motor and conveyor belt in both Lane 1
reliability. The diagram clearly displays the reliability of the (Set 1) and Lane 2 (Set 2).

47638 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

TABLE 11. Failure rate and reliability.

FIGURE 33. Fault Tree Analysis (FTA) diagram with specified fault event.

Case 3 - Sub-system ③: Inaccurate data obtained from the 5) FAILURE ANALYSIS

controlling system can be traced back to power supply 1, The ‘‘Failure Analysis’’ task involved a 300-hour mission,
power supply 2, Arduino, or PLC. during which 600 tests were conducted. Each test took
Case 4 - Sub-system ④: Insufficient counting of bottles approximately 30 minutes to complete. The tests were con-
may be caused by problems with either all three proximity ducted at a rate of 30 per week or 120 per month, and the
sensors failing or with the ESP32 board. entire mission was completed in five months. The exhaustive

VOLUME 11, 2023 47639

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

TABLE 12. Failure rate and combined reliability of 4 main types of failure rate.

testing regimen allowed for a thorough analysis of the sys- demonstration of the proposed approach to provide guidelines
tem’s performance and helped identify potential failures or for target customers to easily understand how to improve
weaknesses. By conducting a comprehensive failure analysis, their business by identifying the items’ conditions that need
the design team was able to make the necessary improve- improvement.
ments to ensure the system’s reliability and robustness. The The design team dedicated significant effort towards devel-
investment of time and resources into the failure analysis task oping a conveyor system with a user-friendly platform and
paid off in the long run, as the resulting improvements helped high standards, while selecting reasonably priced compo-
create a more reliable and effective system. nents that cost less than 1,500 USD based on feedback from
Moreover, MTBF (in this study) means the total time spent target customers.
for uptime (MTTF) and downtime (MTTR). It refers to the
overall duration during which a system or equipment was
operational (uptime) or not operational (downtime – process G. STEP 7: PRODUCTION RAMP-UP
stops for maintenance purposes; repairing or inspecting). This While conveyor systems are well-suited for industrial-scale
includes all instances of planned or unplanned maintenance, plants, the costs associated with them can be significant.
repairs, and other activities that may affect the system’s avail- The initial investment required for equipment and materials
ability. The total time spent for uptime and downtime is an needed to construct conveyors can be high, as can ongoing
important metric for evaluating the performance and reliabil- maintenance costs. These expenses can vary depending on
ity of a system, as it reflects the system’s overall efficiency factors such as the type and complexity of the system, as well
and availability. Since the reliability and failure rate of the as the materials used in its construction. Despite the costs,
conveyor system were tested by running the system contin- however, conveyor systems can offer significant benefits
uously for a fixed duration of 300 hours, during which the to businesses, including increased efficiency, reduced labor
tester observed its performance. The fault tree diagram con- costs, and improved safety.
sidered four factors, namely inadequate separation of opaque Careful evaluation of business needs and budget consid-
and transparent bottles, poor conveyor design, inaccurate erations can help determine whether the advantages of a
data obtained from the controlling system, and inadequate conveyor system outweigh the expenses. It is also important
counting of bottles. The failure rate of each equipment (end to work with experts in the field to ensure that the system
event of the diagram) was obtained, and the reliability of each is designed and installed properly, which can help minimize
component was calculated. maintenance costs over time. To launch this innovation to
The researchers attempted to forecast the worst-case sce- small and medium-sized enterprises (SMEs), two alternative
nario of failed items that could potentially cause direct effects ideas are presented and discussed.
on the main system (sorting conveyor). As a result, all items Idea - Case 1: Purchasing the complete system, including
are considered sensitive cases where an ‘‘OR gate’’ is applied a standard conveyor (2 lanes of 1-meter length and 160-mm
for all sub-branches of the tree. If any of the items fail within width), would cost approximately 900 USD. Alternatively,
a sub-system, the sub-system will fail. Similarly, if any of the SMEs can minimize costs by acquiring a second-hand con-
sub-systems fail, the main system (sorting conveyor) will also veyor with medium-to-high quality.
fail. While the reliability analysis of each situation with the Idea - Case 2: SMEs who already have an existing con-
current configuration is over 75%, the system reliability pro- veyor can purchase only the key algorithm and control unit for
duced only 51.46% when the sensitive reliability calculation approximately 200 USD. The purchased key algorithm and
was applied using the ‘‘OR gate’’. control platform, including the program and sensing devices,
Applying the parallel structure in reliability engineer- would be installed, mounted, and positioned appropriately
ing (i.e., standby items or functions should be available), on the existing conveyor system, similar to building a cus-
the system reliability can be increased as more than one tomized solution based on individual specifications.
set of function can support the system to perform tasks, Besides, Table 13 and 14 present the advantages of utilizing
as shown in standby mode with supportive similar equip- a microcontroller. The Arduino code is readily available on
ment. The in-house prototyping system was constructed as a the internet as an open-source platform, allowing beginners

47640 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

TABLE 13. The strong points of a microcontroller. associated with the innovation, or acquiring the entire busi-
ness that owns the innovation.
In summary, the cost analysis indicates that the design team
invested 14,784 THB (428.74 USD) [87] in developing the
in-house conveyor system prototype. This prototype can offer
insights into the future trend of object separation systems
with an easy-to-access, smaller-scale platform. The proposed
conveyor design has the potential to significantly reduce pro-
cessing time, cost, and manpower required for sorting waste,
from a day to just a few hours and with a 50% reduction in
cost and the need for only 2-3 workers. However, the current
capabilities of the sensor are limited to light intensity and
can only sort bottles based on their transparency. To further
improve the system’s performance, enhancing the sensor’s
functionality with pattern recognition capabilities to detect
colors and classify object shapes for waste separation could
TABLE 14. The objective to sell the innovation. be beneficial.

V. SYSTEM DESIGN OPTIMIZATION

A. PROTOTYPING
The prototype features an acrylic sheet housing in black color
to protect the electronics from external light. The conveyor
belt has been designed to split into two parts and is linked
to a separating conveyor. The system includes a rejector that
immediately rejects the opaque bottle upon receiving a signal
from the light sensor. The transparent lane sensor detects the
color of the plastic bottle, and the sensors at the end of each
or students to access and initiate electronic system control conveyor count the number of bottles. The micro-controller
activities easily and quickly. By leveraging Arduino, indi- receives commands through a smartphone or keypad and dis-
viduals can construct impressive projects that enhance their plays step-by-step instructions on an LCD screen. The motor
future experiences. The primary benefits of incorporating speed is optimized to prevent missing steps and vibrating of
Arduino in support of a developed system are as follows: the motor. The speed is carefully controlled to ensure it is
• Arduino’s smaller size allows it to be installed in com- neither too fast nor too slow, as either case can negatively
pact areas, making it an ideal alternative to PLC. impact the conveyor’s performance.
• Microcontroller maintenance is simplified due to the
‘‘plugs-and-plays’’ concept. In the event of damage, the B. MICROCONTROLLER-PLC INTEGRATION
microcontroller can be readily replaced. The purpose of this study was twofold: to demonstrate the
• Arduino programming is less complex and requires less capabilities of the proposed conveyor design as a stand-alone
effort compared to ladder logic, which takes more time system and as an assistive or additional platform that can be
to reach the destination than the widely-used C++ seamlessly integrated with the existing conveyor constructed
language for Arduino. Errors and bugs can be easily on a PLC controlling platform. One of the main difficulties
rectified. in designing a new platform is integrating it with an existing
• Despite not being the base model (Arduino Mega), the one. The developed sorting conveyor demonstrates how easy
microcontroller is significantly less expensive than PLC. it is to access and apply to the real world, allowing target
• Unlike PLC, microcontroller can control a wide range customers to use it alongside their old system without having
of moving mechanism applications, not just on-off to replace it. The two systems can work well together. The
systems. microcontroller was chosen for this research due to its open-
‘‘Buying a whole system’’ refers to purchasing a complete source nature, lower cost compared to PLCs, and suitability
set of components that work together to provide a particular for SMEs and start-up plants. Open-source applications are
functionality; purchasing a complete conveyor system that generally easier to command and adjust, require less effort,
includes all the necessary sub-components, such as the con- and can support various types of applications as an easy-to-
troller, sensors, and movement-mechanism platform. access platform.
‘‘Buying the innovation’’ refers to acquiring the rights or In contrast, PLCs are closed-source and require more time
ownership of a particular innovative product or technology. to create a ladder diagram with complex patterns of inputs
This involves purchasing the patent, copyright, or trademark and outputs. Users need to provide many numeric values and

VOLUME 11, 2023 47641

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

assigned parameters for some functions. When maintenance fields. Therefore, the choice to use Arduino in the study could
or remodification is needed, users must check each step of the offer several advantages. For example, it could provide a
ladder diagram to extract a command to activate the device(s). cost-effective and efficient means of prototyping electronic
Different PLC brands require different codes and formats, circuits and collecting data. Additionally, it could enable the
which must be carefully understood; otherwise, the com- study to be replicated by other researchers, as the Arduino
mands may not be sent to the machine functionally, and they platform is widely accessible and supported by a large com-
cannot be easily modified. The conceptual model and design munity of developers. The decision to use Arduino in the
of the conveyor system were based on translated customer study is a promising one, as it could open up new avenues
requirements obtained from Kansei Engineering (KE). for research and innovation in various fields.
Therefore, the proposed conveyor system for separating
different types of plastic bottles can support start-up compa- VIII. RECOMMENDATION
nies or small-scale businesses where the fundamental require-
A. IMPROVING THE SYSTEM
ments of customer expectations can be achieved in the proper
To improve the sorting conveyor system, several steps can be
direction. Additionally, the developed separating conveyor
taken. First, the controller program input should be optimized
system can support recycling SMEs with a reasonable price
to minimize errors during task performance. The program
while reducing the manpower required.
should be formatted correctly and concisely to ensure effi-
cient communication between the controller and the motor.
VI. CONCLUSION
Second, sensors can be implemented to detect the position
The newly developed system is highly effective in sorting
and speed of bottles on the conveyor belt. This information
and separating different types of plastic bottles according
can be used to adjust the motor speed and position accurately,
to specific requirements. Its automatic operation signifi-
resulting in better sorting accuracy. Third, regular mainte-
cantly reduces time, cost, and manpower, just as anticipated.
nance should be conducted on the conveyor belt and motor
Consequently, it has become a valuable addition to SMEs,
to prevent breakdowns and ensure optimal performance. The
recycling companies, and sorting plants. When it comes to
motor and belt should be checked for wear and tear and
the controlling platform, two designs were considered in
lubricated regularly. Fourth, upgrading to more advanced
terms of cost. The first option involves a microcontroller,
motor types, such as AC motors or servo motors, can improve
controller, sensors, and movement-mechanism platform sub-
the system’s performance and accuracy. Advanced motor
components, it is priced at around 200 USD. The second
types can provide higher torque, better speed control, and
option utilizes a PLC to manage the moving and separating
more precise positioning. Finally, data analysis can be used
mechanisms, which costs approximately 450 USD without
to identify areas for improvement in the sorting conveyor
the sensors. However, the current sensors used light intensity,
system. Data collected from sensors can be analyzed to iden-
which limited the system’s ability to sort bottles based solely
tify bottlenecks and other issues that can be addressed to
on transparency. To enhance the performance of this proposed
optimize the system’s performance. By implementing these
model, additional features such as pattern recognition, which
steps, the sorting conveyor system’s accuracy, reliability,
can detect the colors or shapes of objects, could be integrated.
and efficiency can be improved, resulting in better overall
For motor application, AC motors can be a suitable choice for
performance.
supporting a sorting conveyor system, especially when high
torque is required. AC motors can generate higher torque by
using a more powerful current, making them more powerful B. PERFORMING MAINTAINANCE TASKS
than DC motors. Additionally, AC motors are known for When a process stops for maintenance purposes, it means
their reliability and can operate for extended periods with- that the equipment or system is taken offline temporarily
out maintenance. However, they may require more complex for repair, inspection, or other maintenance activities. These
installation and control systems than DC motors. Therefore, activities may be planned in advance (such as routine inspec-
the choice of motor type ultimately depends on the specific tions or scheduled maintenance), or they may be unplanned
requirements and constraints of the sorting conveyor system. due to equipment failure or other issues. During maintenance,
Further study can be conducted to compare the performance the equipment or system is usually taken apart, inspected,
and cost-effectiveness of AC and DC motors in a sorting cleaned, and repaired or replaced as necessary. Once the
conveyor system to determine which motor type is the best maintenance is complete, the equipment or system is tested
fit. to ensure that it is operating correctly before it is put back
into service. Maintenance stops are necessary to ensure that
VII. CONTRIBUTION equipment and systems are operating safely and efficiently,
The use of Arduino in the study is a noteworthy point. and to prevent breakdowns and other issues that can lead to
Arduino is a popular open-source platform that can be used downtime and decreased productivity. By conducting regular
for a variety of applications, including prototyping, and devel- maintenance stops, companies can prevent equipment fail-
oping electronic devices. Its flexibility and ease of use make ures, extend the life of their equipment, and ensure that their
it an attractive option for researchers and students in various processes are operating at peak efficiency.

47642 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

ACKNOWLEDGMENT [19] M. Vaughan (Sep. 1, 1999). Conveyor Sortation Systems. [Online]. Avail-
The authors would like to thank the IE-Laboratory and Work- able: https://bit.ly/40lyzTR
[20] G. Mills, Understanding Conveyor Systems. New York, NY, USA: Thomas
shop, School of Manufacturing Systems and Mechanical Publishing Company, 2023. [Online]. Available: http://bit.ly/40jauwU
Engineering, Sirindhorn International Institute of Technol- [21] Dorner. (Jul. 30, 2020). How do Diverting and Sorting Conveyors
ogy, Thammasat University, for the provided instruments, Work? Take a Look. [Online]. Available: https://www.dornerconveyors.
com/asia/blog/diverting-and-sorting
machining operation, and greatly assisted to finish this article. [22] S. P. Gundupalli, S. Hait, and A. Thakur, ‘‘A review on automated sorting
This study was supported by Thammasat University Research of source-separated municipal solid waste for recycling,’’ Waste Manage.,
vol. 60, pp. 56–74, Feb. 2017, doi: 10.1016/j.wasman.2016.09.015.
Fund, Contract No. TUFT 29/2565. [23] K. Rousta, I. Ordoñez, K. Bolton, and L. Dahlén, ‘‘Support for designing
waste sorting systems: A mini review,’’ Waste Manage. Res., vol. 35, no. 11,
REFERENCES pp. 1099–1111, Nov. 2017, doi: 10.1177/0734242X17726164.
[24] B. Miller, J. Spertus, and C. Kamga, ‘‘Costs and benefits of pneumatic
[1] N. Ferronato and V. Torretta, ‘‘Waste mismanagement in developing coun-
collection in three specific new York city cases,’’ Waste Manage., vol. 34,
tries: A review of global issues,’’ Int. J. Environ. Res. Public Health,
no. 11, pp. 1957–1966, Nov. 2014, doi: 10.1016/j.wasman.2014.06.008.
vol. 16, no. 6, p. 1060, Mar. 2019, doi: 10.3390/ijerph16061060.
[2] A. A. Abdulrasoul and S. S. Bakari, ‘‘Challenges and problems of [25] N. Teerioja, K. Moliis, E. Kuvaja, M. Ollikainen, H. Punkkinen,
solid waste management in three main markets in Zanzibar,’’ Adv. Recy- and E. Merta, ‘‘Pneumatic vs. door-to-door waste collection systems
cling Waste Manage., vol. 1, no. 2, pp. 1–9, 2016. [Online]. Available: in existing urban areas: A comparison of economic performance,’’
https://goo.gl/r7EvNQ Waste Manage., vol. 32, no. 10, pp. 1782–1791, Oct. 2012, doi:
[3] B. Ruj, V. Pandey, P. Jash, and V. Srivastava, ‘‘Sorting of plastic waste for 10.1016/j.wasman.2012.05.027.
effective recycling,’’ Int. J. Appl. Sci. Eng. Res., vol. 4, no. 4, pp. 564–571, [26] H. Punkkinen, E. Merta, N. Teerioja, K. Moliis, and E. Kuvaja, ‘‘Envi-
Jan. 2015, doi: 10.6088/ijaser.04058. ronmental sustainability comparison of a hypothetical pneumatic waste
[4] N. Maulidati, A. D. N. Cahyo, D. W. Djamari, M. R. Fikri, and collection system and a door-to-door system,’’ Waste Manage., vol. 32,
F. Triawan, ‘‘Development of sorting system for plastic bottle waste man- no. 10, pp. 1775–1781, Oct. 2012, doi: 10.1016/j.wasman.2012.05.003.
agement,’’ IOP Conf. Ser., Mater. Sci. Eng., vol. 1098, no. 6, Mar. 2021, [27] I. Onderová, V. Čačko, Ĺ. Kolláth, Ĺ. Šooš, A. Smelík, and A. Varga,
Art. no. 062103. [Online]. Available: https://bit.ly/3Z05AE5 ‘‘Pneumatic systems for automatic sorting equipment,’’ Acta Polytech.,
[5] M. Vaughan. (Sep. 1999). Conveyor Sortation Systems. [Online]. vol. 59, no. 6, pp. 573–579, Dec. 2019, doi: 10.14311/AP.2019.59.0573.
Available: https://www.plantengineering.com/articles/conveyor-sortation- [28] E. Simran, S. Kamboj, and A. Diwan, ‘‘Development of automatic sorting
systems/ conveyor belt using PLC,’’ Int. J. Mech. Eng. Technol., vol. 10, no. 8,
[6] S. Chatterjee, D. Hazra, and Y.-C. Byun, ‘‘IncepX-ensemble: Performance pp. 109–118, Aug. 2019. [Online]. Available: http://bit.ly/3YWMZZA
enhancement based on data augmentation and hybrid learning for recycling [29] N. H. Khan, D. K. Sarkar, A. Siddique, M. M. Rahman, and S. Chakraborty,
transparent PET bottles,’’ IEEE Access, vol. 10, pp. 52280–52293, 2022. ‘‘Fabrication of a conveyor belt with object sorting and counting facility,’’
[7] Z. Wang, B. Peng, Y. Huang, and G. Sun, ‘‘Classification for plastic Int. J. Adv. Resea. Technol., vol. 6, no. 5, pp. 60–64, May 2017. [Online].
bottles recycling based on image recognition,’’ Waste Manage., vol. 88, Available: https://bit.ly/3n7vh8g
pp. 170–181, Apr. 2019, doi: 10.1016/J.WASMAN.2019.03.032. [30] T. Ji, H. Fang, R. Zhang, J. Yang, L. Fan, and J. Li, ‘‘Automatic sorting
[8] U. W. Kohaupt, ‘‘Innovative sorting: An essential for economic improve- of low-value recyclable waste: A comparative experimental study,’’ Clean
ments in waste handling,’’ WEKA Industrie Medien, Vienna, Austria, Technol. Environ. Policy, vol. 25, no. 3, pp. 949–961, Oct. 2022, doi:
Jul. 2009. [Online]. Available: http://bit.ly/2U1l9uv 10.1007/s10098-022-02418-7.
[9] S. T. Maruthuru and A. Thirumoorthy, ‘‘Automatic material separating [31] W. Lu and J. Chen, ‘‘Computer vision for solid waste sorting: A criti-
conveyor,’’ Int. J. Innov. Sci. Eng. Technol., vol. 3, no. 12, pp. 130–132, cal review of academic research,’’ Waste Manage., vol. 142, pp. 29–43,
2016. Apr. 2022, doi: 10.1016/j.wasman.2022.02.009.
[10] G. Lodewijks, ‘‘Two decades dynamics of belt conveyor systems,’’ Bulk [32] X. Wu, J. Li, L. Yao, and Z. Xu, ‘‘Auto-sorting commonly recovered plas-
Solids Handling, vol. 22, no. 2, pp. 124–132, 2002. tics from waste household appliances and electronics using near-infrared
[11] M. M. R. Komol and A. K. Podder, ‘‘Design and construction of prod- spectroscopy,’’ J. Cleaner Prod., vol. 246, Feb. 2020, Art. no. 118732, doi:
uct separating conveyor based on color,’’ in Proc. 3rd Int. Conf. Elect. 10.1016/j.jclepro.2019.118732.
Inf. Commun. Technol. (EICT), Khulna, Bangladesh, 2017, pp. 1–5, doi: [33] N. Kroell, X. Chen, B. Küppers, J. Lorenzo, A. Maghmoumi, M. Schlaak,
10.1109/EICT.2017.8275163. E. Thor, C. Nordmann, and K. Greiff, ‘‘Near-infrared-based determination
[12] T. Peršak, B. Viltužnik, J. Hernavs, and S. Klančnik, ‘‘Vision-based sorting of mass-based material flow compositions in mechanical recycling of
systems for transparent plastic granulate,’’ Appl. Sci., vol. 10, no. 12, post-consumer plastics: Technical feasibility enables novel applications,’’
pp. 1–14, 2020. Resour., Conservation Recycling, vol. 191, Apr. 2023, Art. no. 106873,
[13] S. Serranti and G. Bonifazi, ‘‘Techniques for separation of plas- doi: 10.1016/j.resconrec.2023.106873.
tic wastes,’’ in Use of Recycled Plastics in Eco-efficient Concrete [34] N. H. M. Yasin, N. F. Abdullah, and M. A. A. Aziz, ‘‘Optimization of a
(Woodhead Publishing Series in Civil and Structural Engineering), conveyor belt system for municipal solid waste separation,’’ J. Adv. Res.
Sawston, U.K.: Woodhead Publishing, 2019, ch. 2, pp. 9–37. [Online]. Dyn. Control Syst., vol. 13, no. 6, pp. 1316–1327, 2021.
Available: https://bit.ly/42ktj4J [35] X. Huang, L. Yang, Z. Li, M. Xu, and M. Liu, ‘‘Optimization design of
[14] A. Tilmatine, K. Medles, M. Younes, A. Bendaoud, and L. Dascalescu, automatic sorting system based on PLC control,’’ IEEE Access, vol. 8,
‘‘Roll-type versus free-fall electrostatic separation of tribocharged plastic pp. 131979–131991, 2020.
particles,’’ IEEE Trans. Ind. Appl., vol. 46, no. 4, pp. 1564–1569, Jul. 2010, [36] S. M. Sapuan, ‘‘Concurrent engineering, product design, and devel-
doi: 10.1109/TIA.2010.2049553. opment,’’ in Composite Materials Concurrent Engineering Approach.
[15] J.-P. Lange, ‘‘Managing plastic waste–sorting, recycling, disposal, Oxford, U.K.: Butterworth-Heinemann, 2017, ch. 2, pp. 29–56. [Online].
and product redesign,’’ ACS Sustain. Chem. Eng., vol. 9, no. 47, Available: https://bit.ly/3JuZohs
pp. 15722–15738, Nov. 2021. [37] S. K. Mazumdar, Composite Manufacturing: Materials, Product, and
[16] C. A. Velis and E. Cook, ‘‘Mismanagement of plastic waste through open Process Engineering. Boca Raton, FL, USA: CRC Press, 2002. [Online].
burning with emphasis on the global south: A systematic review of risks Available: http://bit.ly/3JjVng1
to occupational and public health,’’ Environ. Sci. Technol., vol. 55, no. 11, [38] N. Stanton, A. Hedge, K. Brookhuis, E. Salas, and H. Hendrick, Handbook
pp. 7186–7207, Jun. 2021, doi: 10.1021/acs.est.0c08536. of Human Factors and Ergonomics Methods. Boca Raton, FL, USA:
[17] N. F. M. Tohit, N. A. Hassan, M. F. Rusli, E. M. Aidid, and R. M. Rus, CRC Press, 2005. [Online]. Available: https://bit.ly/40bx22x
‘‘Solid waste: Its implication for health and risk of vector borne diseases,’’ [39] M. Nagamachi and A. Lokman, Innovations of Kansei Engineer-
J. Wastes Biomass Manage., vol. 1, no. 2, pp. 14–17, Dec. 2019, doi: ing. Boca Raton, FL, USA: CRC Press, 2010. [Online]. Available:
10.26480/jwbm.02.2019.14.17. http://bit.ly/3LCn9H4
[18] N. B. D. Thi, G. Kumar, and C.-Y. Lin, ‘‘An overview of food waste [40] D. P. Restuputri, A. Fridawati, and I. Masudin, ‘‘Customer perception on
management in developing countries: Current status and future per- last-mile delivery services using kansei engineering and conjoint analysis:
spective,’’ J. Environ. Manage., vol. 157, pp. 220–229, Jul. 2015, doi: A case study of Indonesian logistics providers,’’ Logistics, vol. 6, no. 2,
10.1016/j.jenvman.2015.04.022. p. 29, Apr. 2022, doi: 10.3390/logistics6020029.

VOLUME 11, 2023 47643

 S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

[41] H. Quan, S. Li, and J. Hu, ‘‘Product innovation design based on deep [66] CICMHE. (2019). Material Handling Equipment Taxonomy. [Online].
learning and Kansei engineering,’’ Appl. Sci., vol. 8, no. 12, p. 2397, Available: https://www.mhi.org/cicmhe/resources/taxonomy
Nov. 2018, doi: 10.3390/app8122397. [67] D. Xiao, X. Li, and K. He, ‘‘Power balance of starting process for
[42] Q. Fu, J. Lv, S. Tang, and Q. Xie, ‘‘Optimal design of virtual reality pipe belt conveyor based on master-slave control,’’ IEEE Access, vol. 6,
visualization interface based on kansei engineering image space research,’’ pp. 16924–16931, 2018, doi: 10.1109/ACCESS.2018.2810258.
Symmetry, vol. 12, no. 10, p. 1722, Oct. 2020, doi: 10.3390/sym12101722. [68] K. Semrád and K. Draganová, ‘‘Non-destructive testing of pipe conveyor
[43] S. Chanyachatchawan, H.-B. Yan, S. Sriboonchitta, and V.-N. Huynh, belts using glass-coated magnetic microwires,’’ Sustainability, vol. 14,
‘‘A linguistic representation based approach to modelling Kansei data no. 14, p. 8536, Jul. 2022, doi: 10.3390/su14148536.
and its application to consumer-oriented evaluation of traditional prod- [69] M. A. Alspaugh, ‘‘The evolution of intermediate driven belt conveyor
ucts,’’ Knowl.-Based Syst., vol. 138, pp. 124–133, Dec. 2017, doi: technology,’’ Bulk Solids Handling, vol. 23, no. 3, pp. 168–173, 2003.
10.1016/j.knosys.2017.09.037. [70] Thomas Publishing Company. (2019). Types of Conveyors, a ThomasNet
[44] M. Nagamachi, ‘‘Kansei engineering: The implication and applications to Buying Guide. [Online]. Available: http://bit.ly/2GQNakH
product development,’’ in Proc. IEEE SMC Conf., Oct. 1999, pp. 273–278. [71] Bastian Solutions. (2019). Increase Shipping Accuracy and Efficiency With
[Online]. Available: https://bit.ly/3YSZfu0 Sortation Systems. [Online]. Available: http://bit.ly/2GNJ5Oh
[45] M. Nagamachi, ‘‘Hybrid Kansei engineering system and design support,’’ [72] W. Zaher, A. W. Youssef, L. A. Shihata, E. Azab, and M. Mashaly,
Int. J. Ind. Ergon., vol. 19, no. 2, pp. 81–92, 1999, doi: 10.1016/S0169- ‘‘Omnidirectional-wheel conveyor path planning and sorting using rein-
8141(96)00005-4. forcement learning algorithms,’’ IEEE Access, vol. 10, pp. 27945–27959,
[46] L. Zhao, ‘‘Typical failure analysis and processing of belt conveyor,’’ Proc. 2022.
Eng., vol. 26, pp. 942–946, Jan. 2011. [73] D. Santos, ‘‘Online-survey-WiFi-login,’’ Aislelabs, Toronto, ON, Canada,
[47] Y. C. Guo, G. Cheng, K. Hu, and Z. F. Wang, ‘‘Research on the idler spacing 2019. [Online]. Available: http://bit.ly/2V933qw
of belt conveyor,’’ Appl. Mech. Mater., vol. 127, pp. 295–299, Oct. 2011. [74] A. Martin. There’s a ‘Best Way’ to Press a Button, Apparently—And
[48] I. Ristovic, P. Bindzar, and G. Fedorko, ‘‘Limitation defining the area We Could Use it to Make Buttons Great Again. London, U.K.: Dennis
for the implementation of continuous conveyance systems in mines with Publishing, 2019. [Online]. Available: http://bit.ly/2EnekOw
underground mining of metallic mineral raw,’’ in Proc. MAREN, Belgrade, [75] Lauyans & Company. (2013). Heavy-Duty Conveyor. [Online]. Available:
Serbia, 2006, pp. 121–127. [Online]. Available: https://bit.ly/3LCsG0i http://bit.ly/3mWNTIb
[49] D. Woźniak and M. Hardygóra, ‘‘Aspects of selecting appropriate con- [76] Trade India. (2019). Belt Conveyors For Ink Jet Printer. [Online]. Avail-
veyor belt strength,’’ Energies, vol. 14, no. 19, p. 6018, Sep. 2021, doi: able: https://bit.ly/40f7m5a
10.3390/en14196018. [77] Pradeep Bonde. (2009). The Momentum Trade Factory. [Online]. Avail-
[50] I. N. Mhlongo, G. U. Nnachi, A. F. Nnachi, and A. T. Adesola, ‘‘Modelling able: http://bit.ly/2VdlVot
and simulation of conveyor belt for energy efficiency studies,’’ in Proc. [78] Easy Systems. (2019). Conveyor Belts For Rising or Lowering. [Online].
IEEE PES/IAS PowerAfrica, Aug. 2020, pp. 1–6. [Online]. Available: Available: http://bit.ly/2SeO1h7
https://bit.ly/3yKxGIY [79] A. Bradley. (2019). Miniature Machine, Ultimate PLC Trainer, Conveyor
[51] K. Holmberg, P. Kivikytö-Reponen, P. Härkisaari, K. Valtonen, and Part Detection, Electrical Training, Industrial Concepts, Automation Sup-
A. Erdemir, ‘‘Global energy consumption due to friction and wear in the plies and PLC Tools Sitemap. [Online]. Available: http://bit.ly/2N38DYD
mining industry,’’ Tribol. Int., vol. 115, pp. 116–139, Nov. 2017, doi: [80] Motion Dynamic. United MY6812 24V DC 150W, 2750 RPM Electric
10.1016/j.triboint.2017.05.010. Motor (Belt). [Online]. Available: http://bit.ly/2SYfFU9
[52] L. Hrabovský and J. Fries, ‘‘Transport performance of a steeply situ- [81] India MART Inter MESH. Rubber Conveyor Belt. Accessed: Jul. 22, 2022.
ated belt conveyor,’’ Energies, vol. 14, no. 23, p. 7984, Nov. 2021, doi: [Online]. Available: http://bit.ly/2XaSkhd
10.3390/en14237984. [82] Amazon. SunRobotics LDR Light Sensor Module (Photosensitive)
[53] W. Kawalec, N. Suchorab, M. Konieczna-Fuławka, and R. Król, ‘‘Spe- Based on LM393 for Arduino and Other MCU. [Online]. Available:
cific energy consumption of a belt conveyor system in a continuous https://amzn.to/2TSTkV6
surface mine,’’ Energies, vol. 13, no. 19, p. 5214, Oct. 2020, doi: [83] Maker Lab Electronics. (2015). Inductive Proximity Sensor Switch—
10.3390/en13195214. LJ18A3-8-Z/BX. [Online]. Available: http://bit.ly/2IOuRyo
[54] G. Lu, Y. Hu, Z. Yang, Y. Zhang, S. Lu, S. Gong, T. Li, Y. Shen, S. Zhang, [84] (Feb. 18, 2019). Rejector Stepper Arduino V5, [Internet]. S2 Innovation.
and H. Zhuang, ‘‘Geographic latitude and human height—Statistical anal- [Online]. Available: http://bit.ly/2EjSHyz
ysis and case studies from China,’’ Arabian J. Geosci., vol. 15, no. 4, [85] B. W. Niebel, Engineering Maintenance Management, 2nd ed. Boca Raton,
pp. 1–13, Feb. 2022. FL, USA: CRC Press, 2014. [Online]. Available: http://bit.ly/3ZUiHYO
[55] D. Reneker. (Jul. 13, 2017). PLC vs. Arduino For Industrial Control, [86] B. S. Dhillon, Engineering Maintainability: How to Design for Reliability
Control Design. [Online]. Available: http://bit.ly/2PxF5mH and Easy Maintenance. Amsterdam, The Netherlands: Elsevier, 1999.
[56] M. Tupac-Yupanqui, C. Vidal-Silva, L. Pavesi-Farriol, A. S. Ortiz, [87] B. S. Dhillon, Maintainability, Maintenance, and Reliability for Engineers.
J. Cardenas-Cobo, and F. Pereira, ‘‘Exploiting Arduino features to develop New York, NY, USA: Taylor & Francis, 2006.
programming competencies,’’ IEEE Access, vol. 10, pp. 20602–20615, [88] R. K. Mobley, L. R. Higgins, and D. J. Wikoff, Maintenance Engineering
2022, doi: 10.1109/ACCESS.2022.3150101. Handbook, 7th ed. New York, NY, USA: McGraw-Hill, 2008.
[57] M. Crepaldi, A. Merello, and M. Di Salvo, ‘‘A multi-one instruction [89] The Bank of Thailand (BOT). Rates of Exchange of Commercial Banks.
set computer for microcontroller applications,’’ IEEE Access, vol. 9, Accessed: Nov. 11, 2022. [Online]. Available: https://bit.ly/2IEaI9G
pp. 113454–113474, 2021.
[58] Arduino. (2019). Wikipedia. [Online]. Available: http://bit.ly/2yyvQLC
[59] Arduino. (2019). ARDUINO UNO REV3. [Online]. Available:
https://store-usa.arduino.cc/products/arduino-uno-rev3
[60] Amazon. (2018). ARDUINO MEGA 2560 REV3 [A000067]. [Online].
Available: https://amzn.to/2vpCwuk
[61] Faranux Electronics. (2022). Photosensitive Sensor Module
Detection Photoresistor LDR Light Sensor. [Online]. Available:
https://bit.ly/3n6Vx2S SUCHADA RIANMORA received the D.Eng.
[62] DX DealExtreme. (2014). BH1750FVI Digital Light Intensity Sensor Mod- degree from the Asian Institute of Technology,
ule. [Online]. Available: http://bit.ly/2Sk9ftS Thailand. She is currently a Lecturer with the
[63] B. Schweber, ‘‘Options abound when selecting a sensor for motor feed-
School of Manufacturing Systems and Mechanical
back,’’ Mouser Electron., Mansfield, TX, USA, 2019. [Online]. Available:
http://bit.ly/2GZbsJd
Engineering (MSME), Sirindhorn International
[64] M. A. Alspaugh, ‘‘Latest developments in belt conveyor technology,’’ Institute of Technology, Thammasat University,
Presented at the MinExpo, Las Vegas, NV, USA, Sep. 2004. [Online]. Thailand. Her research interests include reverse
Available: https://bit.ly/40fMTgS engineering, rapid prototyping, image process-
[65] P. Kulinowski, Belt Conveyors for Bulk Materials, 5th ed. Naples, FL, ing, industrial instrument and controlling systems,
USA: Conveyor Equipment Manufacturers Association, 1997. [Online]. design and development, design for manufacture
Available: https://bit.ly/3mTdDFj and assembly, maintenance engineering, and manufacturing processes.

47644 VOLUME 11, 2023

S. Rianmora et al.: Design for an Intelligent Waste Classifying System: A Case Study of Plastic Bottles

POOMPACHARA PUNSAWAT received the YOSAKORN TONGTAN received the B.Eng.

B.Eng. degree from the Industrial Engineering degree from the Industrial Engineering Program,
Program, School of Manufacturing Systems and School of Manufacturing Systems and Mechanical
Mechanical Engineering (MSME), Sirindhorn Engineering (MSME), Sirindhorn International
International Institute of Technology, Thammasat Institute of Technology, Thammasat Univer-
University, Thailand. His research interests sity, Thailand. His research interests include
include machine design, industrial instrument and industrial instrument and controlling systems,
controlling systems, production planning, design manufacturing process design, industrial automa-
and development, maintenance engineering, and tion, design and development, computer-aided
design for manufacture and assembly. design, and design for manufacture and assembly.

CHARINYA YUTISAYANUWAT received the

B.Eng. degree from the Industrial Engineering
Program, School of Manufacturing Systems and
Mechanical Engineering (MSME), Sirindhorn
International Institute of Technology, Thammasat
University, Thailand. Her research interests
include computer-aided design, design for man-
ufacture and assembly, industrial instrument and
controlling systems, production planning, design
and development, computer-aided design, and
machine design.

VOLUME 11, 2023 47645