Investigating the Implementation of Augmented Reality in Logistics

Investigating the

Implementation of

Augmented Reality

in Logistics

PAPER WITHIN: Production Development and Management AUTHOR: Telbin Thomas Cheriakkara and Jestin Alex

TUTOR: Milad Ashour Pour JÖNKÖPING JUNE 2020

subject area Production system with a specialization in production development and management. The work is a part of the Master of Science program.

The authors take full responsibility for opinions, conclusions and findings presented.

Examiner: Gary Linnéusson

Supervisor: Milad Ashour Pour

Scope: 30 credits (second cycle)

Abstract

The rise of the new industrial era has brought about the need for adopting new technol-ogies, digitization, automation, and improving the industrial internet of things. With the introduction of Industry 4.0, there have been a lot of changes that transformed the way of understanding how technologies have evolved and helped to improve productivity in various sectors. Through this thesis, we try to discuss how Augmented Reality (AR), which is one the main pillar of Industry 4.0 plays a vital role in industrial innovation and development. It is important to simplify the execution of logistics operations and make them more reliable where human resources are involved to decrease the error rate and decision-making time. The report provides information about the technologies that are currently used in the logistic industries such as barcode and RFID. A brief descrip-tion of AR components, barriers, and its applicadescrip-tions are discussed in the theoretical background. It follows a case study to understand common problem areas in the pro-duction system and find out how AR can improve these conditions by comparing them with data gathered from literature and interviews conducted. The thesis was performed in collaboration with one of the main logistic firms particularly focusing on the topic “Investigating the Implementation of Augmented Reality in Logistics”. This report will help to find out the critical problems that are present in the current processes and how augmented reality can help the industries to improve their system.

Keywords

Contents

1

Introduction ... 1

1.1 BACKGROUND ... 1

1.2 PROBLEM DESCRIPTION ... 2

1.3 PURPOSE AND RESEARCH QUESTIONS ... 2

1.4 DELIMITATIONS ... 3 1.5 OUTLINE ... 3

2

Theoretical Background ... 4

2.3.2 Components of Augmented Reality ... 8

2.3.3 Process Flow Chart ...12

2.3.4 Augmented Reality Interface ...13

2.3.5 Application of Augmented Reality ...14

2.3.6 Barriers of Augmented Reality ...15

3

Method and Implementation... 17

3.1 SOURCE OF DATA ...17

3.1.1 Literature review ...17

3.1.2 Observation ...18

3.1.3 Questionnaire ...18

3.2 RESEARCH DESIGN ...19

3.2.1 Qualitative Data Analysis ...20

3.2.2 Reliability and Validity ...20

4

Findings and Analysis ... 21

4.1 CASE STUDY ...21

4.1.2 Problem area 2: Production Floor ...24

4.2 ANALYSIS ...24

4.2.1 Receiving and Storing ...25

4.2.2 Picking and Shipping...25

4.2.3 Optimization of operation ...26

4.2.4 Training and safety ...26

5

Discussion ... 27

5.1 DISCUSSION OF FINDINGS ...27

6

Conclusion ... 31

6.1 SCOPE FOR FUTURE RESEARCH ...31

7

References ... 32

8

Appendices ... 37

8.1 APPENDIX 1:INTERVIEW WITH THE CASE STUDY COMPANY MANAGER AND OPERATORS. ...37

List of Tables.

Table 1. Comparison of Barcode and RFID (White, et al., 2007) ... 7

Table 2. Barriers of using AR in warehouse operation (Stoltz, et al., 2017) ...16

Table 3. Database and articles used for reference ...17

Table 4. Problem area and challenges ...23

Table 5. Uses of AR in warehouse operations (Stoltz, et al., 2017) ...25

Table 6. Benefits of AR in logistics (Stoltz, et al., 2017) ...29

Table 7. Benefits of AR in Order Picking in the work floor (Stoltz, et al., 2017) ...30

Table of Figures.

Figure 2: Two-dimensional Barcode ... 4

Figure 3: AR component interaction (Masood & Egger, 2020) ... 8

Figure 4: Six Degree of Freedom ...10

Figure 5: AR Procedure Flow Chart (Bathiche, et al., 2014) ...12

Figure 6: Combining real-world with the AR system (Bathiche, et al., 2014) ...14

Figure 7: Research Design ...19

Figure 8: Process Flow Diagram ...22

1 Introduction

This chapter mainly explains the background of the project and problem description. The purpose, research question, scope, delimitation, and the outline of the report is described.

1.1 Background

The concept of Industry 4.0 incorporates production processes and the implementation of the latest technology including, but not limited to the use of cyber-physical systems as well as widespread internet use (Torback & Kijewska, 2019). It is widely acknowl-edged that, in order to maintain a long-term competitive advantage, firms need to build core capabilities (Ward, et al., 1996) and achieve the highest levels of performance along with dimensions such as quality, flexibility, delivery, and cost (Sarmiento, et al., 2007). AR technology has been used in a wide range of fields like tourism, marketing, manufacturing, maintenance, and others (Westerfield, et al., 2015).

Augmented reality: a medium in which digital content is superimposed over the physi-cal world, which is both spatial and temporal in the physiphysi-cal world, and which is inter-active in real-time (Craig, 2013)

AR is seen as one of the technologies that could bring the “next big wave of change” in the industry (Glockner, et al., 2014). AR, which is one of the main pillars of Industry 4.0 plays a role in industrial innovation and development. AR offers a prominent solu-tion to solve these problems by allowing users to make decisions based on computer-generated visualizations and 3D model projections. Its applications in the industry in-clude production assistance, designing, training, quality assurance, support of indoor navigation through AR-based guidance, and remote maintenance (Schwerdtfeger & Klinker, 2008).

The use of AR technology is becoming important in logistics, in support of the human operator. The evolution of AR technology has proven to be an effective key to problems related to simulation, support, and processes guidance (Nee & Ong, 2012). Through AR, we can visualize the data that the employees work with and provide it when re-quired immediately so they do not need to memorize too much about the operation. The system will automatically retrieve the necessary data and help the worker to make the required decisions clearly and effectively. This can also help in clearer and easy execu-tion of the processes without errors and reducing the risk/accidents that may occur in the warehouse, especially for the forklift driver thus making safety a high priority (Fraga-Lamas, et al., 2018).

AR includes the integration of the physical and digital world through a wearable device in real-time and has been recognized as an enabler for the cyber-physical industry (Mourtzisa, et al., 2019).

1.2 Problem Description

In recent years, logistics service providers have been prompted to investigate the po-tential of AR technology such as AR glasses. Some have even moved forward to im-plement AR systems in full operation. Its potential application ranges from order pick-ing, special training applications, and maintenance for transport systems and has helped in the visualization and optimization of the material flow (Williams, 2019). Early ap-plications were centered on inventory operations and product picking, but as technology advances and the number of goods on the market expands, so, there is some indication that the usefulness of this technology will go even further. Existing technologies in-crease the performance of identifying logistics items. However, there remain some er-rors that could cause substantial damage and loss. The number of possible erer-rors may be minimized by the additional checking of items using 3D visualization (Gintersa & Martin, 2013).

The biggest problems now are the ergonomics of using smart devices and the reliability of data processing. Moreover, this device would have to find its place among other pick-by systems that can deliver the same data to the worker (Stoltz, et al., 2017). Order picking is by far the most studied field, as it accounts for more than 50% of the ware-house costs (Giannikas, et al., 2016). Existing research concentrates on how to optimize a human operator's routing using AR, what is the most effective way to show a picker's storage position, and contrasts between voice-communicated pick lists, using lights, via a head-mounted display (Schwerdtfeger, et al., 2009). The thesis will be performed in collaboration with one of the main logistic firms particularly focusing on the topic “In-vestigating the Implementation of Augmented Reality in Logistics”. This will help to find out the critical problems that are present in the current process and how through augmented reality helps them to improve their system.

1.3 Purpose and Research Questions

The study focuses on the opportunities and limitations that Augmented reality could bring in logistics management. It also aims to investigate the implementation possibil-ities of AR (Augmented Reality) technology in the field of logistics

Research Question:

RQ1: How could the AR system implementation effectively improve the information transfer in the Logistic system?

The objective of this research question is to form a generalized understanding of the effect of replacing conventional barcoding technology with AR technology. Through this, we can understand more about how effective AR is in the logistics field and its limitations.

RQ2: What effects can AR technology bring about in Order picking of delivery goods? In logistics, products are stored in their storage places. These products are delivered only when the customer provides the order. Likewise, there will be different kinds of products placed in different storage locations. Presently these data are stored in the computer and hard copies of this information are used to find the allocated area of the

products. Using AR, we can have a visualized picture of where the product is stored and avoid the complexity of order picking. This will help investigate both positives and negatives of implementing AR technology on delivery goods.

1.4 Delimitations

While studying the relevance of implementing Augmented reality in logistics various challenges, such as time constrain and procuring relevant information were faced. In the logistics company, several types of commodities are handled with different shapes, sizes, and costs. This study focuses its emphasis on heavy bulk goods in the logistics market, where the expense of introducing AR can easily be negotiated in the form of protection for goods.

The study will only mention the software components of AR but will not be providing its explanation. Detailed information can be found from the reference provided. The thesis will not be providing cost estimation regarding the implementation instead focus on the relevance of adopting AR in logistics processes, advantages, disadvantages, and comparisons with other related technologies being used currently in this field.

1.5 Outline

This thesis is categorized into five chapters. The outline of these chapters is described below:

Chapter 1; mainly deals with the introduction part, which consists of the background, problem formulation, aim of the study, and the research questions. It also describes delimitations and the outline of each following chapter.

In Chapter 2; Theoretical Background, basics of Augmented reality and other related technologies such as Barcoding, RFID, pros, and cons are discussed. The various com-ponents, benefits, and applications are also elaborated in this chapter.

In Chapter 3; Methods and Implementations, research method and design of the study are explained with its various steps of implementation phases.

In Chapter 4; Findings and analysis, the data collected from the company are pre-sented, which includes the questionnaire results, their comparisons, intended benefits, and analyzing the needfulness of this technology.

In Chapter 5; Discussion and conclusion, the research questions are answered and explained how this topic is relevant to future studies. It summarizes the overall thesis and helps to understand implementation prospects in logistics companies.

2 Theoretical Background

This chapter starts with an introduction to the technologies that are used currently in Logistics, i.e., barcoding and RFID technology. This is followed by presenting the con-cept of augmented reality (AR) and describes what is meant by the term augmented reality.

2.1 Barcoding Technology

Barcode has been used in the service industry since the mid-1970s (Attaran, 2007). It's a line-of-view process, which means that a scanner needs to ' see ' the barcode to scan it. Barcode is not a modern technology; it was already in use in the 1940s (Rajesh, 2003). Barcodes are part of every package that we buy and have become the "universal standard for package recognition and monitoring". The European companies mentioned that barcoding technology has benefited an improved efficiency over the last few dec-ades. The reason why they are using were; it increases the accuracy of ordering and invoicing, cost reduction, and the fact that newer technology isn’t ready yet (White, et al., 2007)

Barcodes can be described as a self-contained message with information encoded in a sequence of black bars with varying widths and white spaces between them. It is a way to describe a sequence of numbers or letters. There are several kinds of barcodes, but most of them represent a mix of broad and narrow bars for each character. Barcodes provide easy detection by an optical system. They are read by a scanner that sends sig-nals to a machine that covers the number identified by the barcodes. Machine defines bars as O s and 1 s (zero and one) and white blanks as ' offs ' or 'on's’. Therefore, a barcode is a sequence of 0 s which Is representing characters or digits in a shape that can only be recognized by computers (Ramanathan, et al., January 2014).

According to (Huber, et al., 2007)Barcode does not provide any security capability. There are two types of barcodes: One and two-dimensional barcodes. Label data is stored in both horizontal and vertical dimensions and, as more data are stored, the size of the barcode will be expanded in both horizontal and vertical directions, thereby re-taining a compact form for effective scanning and labeling requirements. Two-dimen-sional barcodes are now being used for concert tickets by sending a barcode to a cell phone and then checking the image at the door with a laser gun (White, et al., 2007).

Figure 1: One Dimensional Bar Code

Figure 2: Two-dimensional Barcode Barcodes have found a range of uses in diverse areas, including libraries and infor-mation facilities. In developing nations, barcode patterns have become a common

emblem for the public due to their use in all personal goods, books, food and retail packaging, clothes, other documents, etc.

The barcode system has the following constraints:

• The barcode defines a category of objects, not a particular component of that kind.

• Tracking is not fast. For instance, to keep track of the products, each barcode must be checked on each element in the store.

• The barcode does not contain any detail other than the code of the category of code (Thornton & Sanghera, 2007).

Barcode scanning eliminates errors involved with manual data processing and provides insight to better control the supply chain (Fraza, 2000). Currently, barcode-based sys-tems are commonly used in the short shelf-life supply chain. There are different types of scanning devices, from fixed readers to finger scanners, and they are found in several phases of the delivery network. In addition, the adoption of new bar-coding standards enables the adding of sell-by dates to codes to maintain the validity of product turnover and thereby help to overcome the issue of contamination (Shulman, 2001).

2.2 RFID Technology

RFID (Radio Frequency Identification) is a portable and automatic method of data col-lection that is similar to the bar code technique. Any basic RFID system is a hybrid of three key elements, namely: an active, passive or semi-passive RFID tags that acts as an electronic database and can be connected to or inserted in an identifiable physical object; a reader and its antennas that interact with the tag without requesting a line of sight; And a middleware-equipped database server responsible for controlling the RFID network and communicating with an intra-and inter-organizational information system. RFID tags can be small stickers containing antennas that receive and respond to trans-missions from RFID transmitters. RFID tags can be used to classify and document an-ything about the product (Thornton & Sanghera, 2007). Depending on the degree of use, varying from 1 meter to multiple meters, the expense of RFID tags is often signif-icant (Attaran, 2007). These tags are inserted in the package, pallet, or container. Tags are used to store and relay information on a specific device. RFID readers are radio frequency transmitters and receivers, powered by a microprocessor or digital signal processor that interacts with tags (Wang, et al., 2010).

According to (Wyld, 2006), the RFID technology has five primary capabilities which make it different from barcodes: there is no need to read the line of sight, each tran-sponder may have a unique code, It is more robust, it can carry more data and it enables many transponders to be read almost simultaneously.

To use RFID, a transponder, a reader, and software is required to forward the content from the reader. You may connect the transponder to a component or link it directly to the flow. The reader connects over the radio wave to the transponder. Transponders may be passive, or active, usually operated by a battery. With passive tags, the reader transmits a low-power radio signal from an antenna that the tag collects from its own antenna to control an integrated chip. Using the energy from the signal, the tag con-verses briefly with the user for authentication and data exchange. When the data is ob-tained by the user, it can be transmitted to a monitoring computer and placed in a data-base for further processing and analysis. Active tags function in the same manner as passive tags; the distinction is that active tags constantly relay signals through their antenna (White, et al., 2007).

Passive and active tags function in quite different way and still contain the same con-tent, known as the Electronic Product Code (EPC). This is a product numbering scheme that uses an extra series of numbers like barcodes and assigns a unique product ID num-ber to each object made. Since the information is processed in just one location, it makes it much easier to adjust the information, such as the destination of the item and the number of items. This also means that the route of the product through the supply chain can be closely monitored and reviewed (White, et al., 2007)

Active transponders are bigger and have an extended read capacity with a reduced life cycle and are more costly than passive transponders. The reader provides passive tran-sponders with energy and therefore demands more effective readers (Kvarnström & Vanhatalo, 2010).

The transponder consists of four components: an antenna, an antenna-connected chip, an (if active) energy source, and a shield that covers it. In addition to not needing a line of sight, RFID tags can be used to communicate with the readers, unlike barcodes. This characteristic makes RFID a new development in distribution, replenishment of stock, warehouse management, quality monitoring, and supply chain management (Barjis & Wamba, 2010).

RFID is useful in the logistics field in enhancing the performance of surveillance, trans-portation, loading, and unloading, etc. Although the initial cost in this technology is higher than that of other related technologies (such as barcode), the extent of RFID coverage is much greater. Since RFID technology has been seen as a significant tech-nology capable of offering strategic and organizational advantages, it is crucial to con-sider what defines the implementation of RFID in the logistics industry (Ramanathan, et al., January 2014). According to (Want, 2006), the widespread use of RFID is being held back by three primary problems: cost, design, and acceptance. There are also ex-ternal factors that complicate the use of RFID in processes. For e.g., the presence of water and metal in the field of reading can cause the reading rate to dramatically get reduced (Wyld, 2006). The following Table 1 illustrates a comparison of Barcode tech-nology and RFID.

Table 1. Comparison of Barcode and RFID (White, et al., 2007)

Barcode RFID

Need a line of sight to be read It can be read without a line of sight Can only be read independently Many tags to be read continuously Cannot be read if broken or filthy Can manage hard or filthy conditions Can only recognize the type of object Can distinguish a specific object It Cannot be modified & need manual

moni-toring and are thus vulnerable to human er-ror.

New information can be overwritten & can be automatically monitored to avoid human er-ror.

2.3 Augmented Reality 2.3.1 Introduction

In the 1960s, Ivan Sutherland with the help of his students at the University of Utah created the first virtual reality and Augmented reality head-mounted display system (Sutherland, 1965). It was the first step towards understanding the potential of AR. The first application of AR in order to support an industrial process started in the early 1990s when Caudell and Mizell from Boeing, presented a head-up see-through display used to augment the visual field of an operator with information related to the task she/he was carrying out (Caudell & Mizell, 1992). Even though the first Industrial Augmented reality (IAR) systems were mostly experimental but in the recent decade many com-mercial projects were initiated. For example, in 2013 Google Glasses attracted the eye of a wide market unfamiliar with AR or IAR and received interest from certain indus-tries. Since then active development of AR products and prototypes has become a fast-growing market with huge prospects.

Augmented reality includes integrating the physical and the digital world through a wearable system in real-time (Glockner, et al., 2014) .AR technology helps a person to see a computer-generated virtual environment combined with the real world. The "real world" is the place an observer can see, hear, taste, smell, or feel using the very senses of the observer. The "virtual world" is described as an environment created which is stored in a storage medium using a processor. In an AR system, a tracking device rec-ords the virtual world to the physical world in order to combine virtual and actual in-formation in a way that the user can use. In simple terms an AR system should:

1) Operate in Real-time

2) Mix virtual elements with reality,

3) be integrated with a 3D environment (Azuma, et al., 2001)

By providing flexible real-time information and the possibility of obtaining hands-free AR information, it can provide a substantial efficiency benefit by reducing the error rate, such as picking or assembly errors, and provides easy ways to communicate with maintenance experts in industries. AR can be used either in training or in assembly operations or as a live operator guidance system. 'Pick-by-vision' is a prominent concept

in logistics that uses AR to indicate locations and total quantity to be picked (Hanson, et al., 2017). By using AR both barcode and RFID can be scanned for identification. Another logistics area where AR could be used is general warehouse operations. Other prominent application fields include quality assurance and maintenance. Figure 3 illus-trates the basic components of AR and their interaction with each other.

Figure 3 : AR component interaction (Masood & Egger, 2020) 2.3.2 Components of Augmented Reality

2.3.2.1 Hardware components of augmented reality

In general, there are two key things that need to happen for each stage in the implemen-tation of augmented reality. The two steps are as follows:

1. The program must assess the current state of the physical universe and the pre-sent state of the virtual world.

2. The program has to view virtual reality in synchronization with the real world in such a manner as to allow the participant(s) to experience the virtual world elements as part of his or her actual environment and then go back to step 1 and go onto the next step.

In order to be able to respond correctly to the physical world, the application of aug-mented reality must have real-time data about the real world (Craig, 2013).

The constituent components of an AR system are:

a) A component for capturing images (i.e. a Charge-Coupled Device (CCD), a ste-reo, or a sensor for sensing depth).

Many modern augmented reality systems use computer vision methods to assess the position and viewpoint of the user with respect to the physical world. Here the sensor seems to be a camera to support computer vision. The camera "sees" the natural world and, depending on what it "sees," will determine where it (the object) is positioned and whether it is associated with the situation. In order to perform a computer vision, the software is used to interpret the data captured by the camera to order to decide what the camera "sees." Depending on the detail, the algorithm determines where the camera has to be in order to see the image. There will also be environmental indicators that the camera may use as landmarks to effectively assess the position and direction of those landmarks. In order to make the computer vision problem easier, many AR systems use symbols that are digitally positioned in the world that are icons that the device can quickly remember. Images used expressly for this purpose are referred to as fiducial markers, and sometimes fiducial symbols (Craig, 2013).

b) A monitor to project virtual image knowledge that the capture feature acquires. Display technologies are generally classified into two types: video mixed dis-play and optical view-through disdis-plays (Benko, et al., 2015).

The visual and physical knowledge previously obtained with a camera is digitally com-bined in a video-mixed display and displayed on a monitor. In comparison, in an optical see-through display, virtual information is superimposed on the user’s field of view by means of an optical projection system (Craig, 2013).

The equipment used for both display systems can be categorized as: -

• Hand-Held Displays (HHD) is a screen that fits into the hand of a user (e.g., tablets, smartphones).

• Special display. They use digital projectors to represent real object information in graphic form. Such displays encourage collaboration activities among users, as they are not connected to a single user.

• Head-Mounted Displays (HMD) These are the screens used in gadgets such as smart glasses and smart helmets, which allow users to view the entire external world (Benko, et al., 2015).

c) A processing unit that sends out the visual simulated information

A processor that coordinates and analyses sensor inputs, stores and retrieves data, per-forms the tasks of the AR application program, and generates the appropriate signals to display is at the heart of any augmented reality system. In other words, every augmented reality system requires a kind of computer to function. Computing systems for AR can vary in complexity across networks, from basic mobile devices such as smartphones and tablets to notebooks, personal computers, and workstation level machines. A port-able device is in some cases in contact with a strong server that may be placed at a distance (Craig, 2013).

d) Activating elements

AR relies on being spatially recorded, there must be a method for deciding the location of the user, the physical world, and any AR tools. For that, it requires both location and orientation. To completely evaluate the location, information on the six degrees of free-dom of the object to be observed is required. Six degrees of freefree-dom in this situation include X position, Y position, Z direction, yaw, pitch, and roll. At a minimum, this detail will include the positioning and orientation of the participant in the physical world (Craig, 2013).

Figure 4: Six Degree of Freedom • GPS

GPS is a navigation system that uses a network of 24 space satellites. The receiver will calculate its position in X and Y whether it will receive 3 satellites by calculating the time it takes for the GPS signal to pass from the satellite to the receiver. By measuring the time signal takes from many different satellites, the position of the receiver can be determined within a few meters. If there are 4 or more satellite signals available, the altitude of the receiver can also be determined. AR systems may use the positioning information provided by the GPS receiver to extract information on its position in X, Y, and potentially Z (Craig, 2013).

• Gyroscopes, Accelerometers, and Other Types of Sensors

A variety of other types of cameras can be found in augmented reality applications. The purpose is to obtain information about the physical world and to use that information to inform the application. Some of the more common sensors used, especially in mobile AR applications, are installed with many handheld devices, including accelerometers, compasses, and gyroscopes.

Gyroscopes report may provide detail, including yaw, pitch, and roll. Gyroscopes do not have position information. They can be useful for calculating the orientation of a portable machine or an interface system. Compasses give information as to the direction in which they are pointed. Much like an old-fashioned compass, an electronic compass will indicate if you are facing north, south, east, west, or somewhere in between. Ac-celerometers are doing exactly what their name implies. They are recording accelera-tion. These can be used to assess the direction in which it is traveling and shifts in orientation, which may not rely on the availability of GPS signals. Accelerometers are low-cost and can produce valuable data for AR applications (Craig, 2013).

There are many other sensors that are less widely found in AR systems but at the time and place where the AR device is being used may provide data about the physical en-vironment. For example, in an AR application, temperature sensors, humidity sensors, and other sensors can be used to overlay atmospheric information about a scene (Craig, 2013).

• Sensors for Gathering User Input

The sensors previously mentioned are mainly passive sensors from the participant's per-spective on the AR environment. Clearly, as the participant moves, the sensors record shifting and adjusting the view accordingly, but consciously communicating with a sen-sor isn't a deliberate activity on the part of the participant. Many of the common sensen-sors collecting user feedback may include buttons, touchscreens, keyboards, and other standard tools for user interfaces. Many handheld devices such as smartphones and tab-lets have a range of sensors in the form of buttons and keyboards (whether physical or virtual) that can be used by an AR technology designer to provide a means for the user to communicate directly and control an application (Craig, 2013).

2.3.2.2 Software components of augmented reality

The software that is involved directly in the AR application includes the following func-tional components:

a) Environmental acquisition (sensors) b) Sensor integration

c) Application engine

2.3.3 Process Flow Chart

Figure 5: AR Procedure Flow Chart (Bathiche, et al., 2014)

Figure 5 illustrates the process flow which starts by monitoring the real-world object. Here, objects such as visual objects, auditory objects, etc. can be seen. It is understand-able and appreciated that factors like location, date/time, engaging activity can be used for finding which item is to be monitored. These factors can also be employed in AR with virtual information. These objects can be monitored by the way of pattern or Sound recognition. Thus, virtual information is collected which can be related to real-world objects. At the next level, the real-world objects and activities can be augmented with virtual information. In other terms, the information can be integrated into a cohesive simplified version of real-world perception. Thereafter the augmented experience can be rendered to the final user. Thus, the system background can be used to make the augmented interface effectively (Bathiche, et al., 2014).

AR technology essentially "connects" a human user to a database or to a simulated or partially virtual world using some combination of the visual or aural or kinesthetic senses of the human user. Alternatively, as observed by the human user, AR technology may be seen as connecting the partially virtual environment with the real environment. AR technology enables the human user to perceive and be guided by database infor-mation processed that is integrated with the real world. Therefore, AR technology can allow a person to interact in an event in a virtual world by transforming the actions or

START

MONITOR THE REAL WORLD OBJECTS

GATHER VIRTUAL INFORMATION

AUGMENT REAL-WORLD OBJECTS

RENDER AUGMENTED OBJECTS

behavior of the human user within a given real-world area or volume into the desired response in the virtual world (Meisner, et al., 2007).

Visual AR technology comprises technologies such as "video See Through" AR and "Optical See Through" AR. Video See-Through Using AR imaging, a camera takes real-world images and combines or incorporates synthetic images electronically to cre-ate the augmented image. In Optical See-Through, Virtual images are projected on a see-through monitor the AR technology, allowing a human user to see the projected image of the virtual object on the monitor and the representations of the objects in the physical world from the screen (Meisner, et al., 2007).

2.3.4 Augmented Reality Interface

An ideal AR system should allow a physical user to communicate with objects in both the real and the virtual world, conveniently and readily. A visual AR program will, therefore:

a) identify and monitor input data such as coordinate marks or fiduciaries easily and accurately.

b) evaluate the input data easily to assess the relative position and orientation between the user and the target objects and to record them and

c) combine virtual objects seamlessly with real-world objects either by displaying or projecting an image of virtual world objects over real-life objects or by integrating an image of virtual world objects with a captured image of real-world objects electroni-cally (Bathiche, et al., 2014).

To maintain a stable alignment of objects, AR systems must calculate the location and orientation of the simulated object in real-time (Lima, et al., 2017). This can be done by using markers in the real scenes which can be Identified by cameras and compared to predefined trends (Khan, et al., 2015)or by using markerless tracker like Natural Fea-ture Tracking (NFT) Simultaneous Localization and Mapping (SLAM)

In NFT techniques, the AR system detects certain characteristic points from the image in real-time, and the pose of virtual objects is calculated based on these points (Fraga-Lamas, et al., 2018).

To estimate the robot's real-time positions and orientations, the SLAM algorithms were first developed for robotics applications and then adapted for AR (Billinghurst, et al., 2015). This method consists of creating a probabilistic feature-based map that estimates the camera's real-time pose and the position of interesting features (Davison, et al., 2007).

One method of implementing SLAM is adopted for AR via Parallel Tracking and Map-ping (PTAM), where tracking and mapMap-ping are done separately. PTAM allows a robust tracking method to be used, executing both tasks in different processing cores, and re-ducing video frame mapping in frames where the camera tries to move (Klein & Murray, 2007). Figure 6 represents the AR interface system.

Figure 6: Combining real-world with the AR system (Bathiche, et al., 2014) 2.3.5 Application of Augmented Reality

AR technology has found its application in almost all fields from medicine, education, business, manufacturing, military, tourism, and many more. This wide acceptance is because of its visualization and easy to understand the feature. It enhances the percep-tion of reality and helps preview products and simulapercep-tion using 3D models (Julie, et al., 2011). Some of the applications are mentioned below.

▪ Industrial manufacturing: Here AR can be used to visualize standard operating procedures and assist the operator in production, thereby reducing errors and mental stress (Dimitris, et al., 2019). This also improves the safety of the work-station and provides the required information at the right time.

▪ Commerce: AR helps to integrate print and video marketing. Nowadays we can find advertisements that have trigger photos that activate the video version of the ad when scanned using an AR-enabled device. Many other promotional fea-tures like virtual dressing rooms and apps that provide product previews to view

products in their home location even before purchasing the product (Sharma, 2015).

▪ Real Estate: Since the Internet is the best location where sellers and tenants are searching for new houses and buildings, AR can be of a lot of use for real estate. This will help clients view homes better, sort the ones they want, and do not like and save both clients, and realtors some money. It is also a perfect resource for construction workers.

▪ Tourism and maps: There are a lot of applications for AR in tourism. It can enhance the experience of sightseeing by providing information to the tourist in their language, visualize the museum’s exhibitions and see the history of the artifact or find out how they looked in the past. AR enabled maps and apps that help navigate and explore cities that have revolutionized tourism.

▪ Education: The incorporation of AR technologies into a classroom will make teaching more exciting, immersive, and perhaps the most-dull subject a little enjoyable. Students may get more comprehensive knowledge and step-by-step guidance, and above all better understanding of the subject.

▪ Healthcare: Healthcare has been one of the sectors which have benefited from augmented reality. Helping patients properly define and explain their symptoms with software such that will, for example, display an effect of cataract on the human eye, and thereby help a patient recognize the symptoms. It can support nurses to find veins more quickly and aid surgeons during the operation proce-dure (Kovtun, 2019).

2.3.6 Barriers of Augmented Reality

When AR is implemented at the pilot stage, it generally never progresses to full deploy-ment due to various barriers. It happens when AR fails to gain traction due to subopti-mal planning, communication, and implementation (Boland, 2019). Organization adop-tion should include advanced features such as scalability across computers, customers, websites, reducing video lag, intuitive user interfaces, protection of information, user access, etc. Main hurdles for the adoption of AR in an organization are:

• User Privacy: This is the major implications of the constant monitoring and recording of user behaviors across social, cultural, and IP-related barrier • Technological: efficiency and technological problems in optics, AR

sys-tems, applications, 3-D monitoring, and view are found.

• Fragmented Markets: There are presently several variants in the AR market. Hardware, software, and frameworks, all are lacking uniform data pro-cessing, management, storage, and device integration requirements (Hashim, 2017).

While considering the implementation of AR in warehouse operations, the initial costs of setting up an AR system in place can increase the costs of the projects (Agarwal, 2016). An increased cost would cause a negative impact on the decision makers of the project. Other significant issues to consider are the size and weight of the devices. De-veloping user-friendly applications that abide by the right context awareness

and framework is a significant barrier to its implementation (Silverio, et al., 2017). These are mainly because AR is in a transition phase of advancement and yet to reach its highest potential. (Stoltz, et al., 2017) has summarized some of the AR barriers that were found in the warehouse operations mentioned in table 2.

Table 2. Barriers of using AR in warehouse operation (Stoltz, et al., 2017)

Type of Barriers Challenges

Hardware Limitations

Battery • Batteries are not designed for long working hours, so extra batteries need to be kept or must be recharged re-peatedly.

Wearable Devices • Not comfortable for long working hours. It may be be-cause of the design of the product which needs to be improved. Symptoms like eye tiredness, headaches, and heaviness due to the device itself have been seen in a few cases.

• Head-mounted devices feel much slower than using handheld devices.

Processors • Devices will overheat after continuous use for long hours.

Security

Privacy issues • Few individuals feel that it’s a violation against their pri-vacy to wear devices with cameras and mic all the time. • Confidentiality of personnel/company information can

be a concern. Cost

Ownership • The initial investment may be high.

• The Setup and maintenance of IT solutions may be costly.

3 Method and Implementation

In this chapter, the research method and design of the study are explained with its var-ious steps of implementation phases used, in order to answer research questions and provide detailed findings and analysis is presented.

3.1 Source of Data

The primary and secondary data will be collected using the following methods. 3.1.1 Literature review

This method provides the required current knowledge including substantive find-ings with theoretical contributions to a topic. This will help in getting in-depth knowledge about the sector, the models that govern the system and pattern of change.

A qualitative analysis of literature aims to include 1) the theoretical context for this thesis, and

2) the interpretation of keywords and topics applicable to our study (Dudovskiy, 2018).

A literature review has been performed by selecting papers from Google Scholar, Sco-pus, ResearchGate, and JU library databases. The authors selected these databases be-cause of their thorough coverage of the publications in this area. It provides search combinations using "and" and an alternative of looking for keywords. The analysis method for qualitative literature was composed of two parts: Firstly, an introspective and unstructured one with a variety of different origins; secondly, a more formal one concerning the search for information using search strings and dashboards. The strategy was to identify articles that included “Logistics 4.0”, “augmented reality”, “order pick-ing”, “barcoding” and “RFID” as keywords in all fields. Table 3 shows the list of data-bases and articles referred to for the report.

Table 3. Database and articles used for reference

Sources No. of articles referred

Google Scholar 8

Scopus 12

ResearchGate 15

JU library databases 7

Science Direct 18

The literature review was conducted from various sources mentioned in the above table. Most of the data collected were from journal articles which makes the base clear. For the support, various conference papers, books, and company websites were also re-ferred. This helped in understanding the basics of AR technology and components. RQ1

and RQ2 were more answered by reviewing various papers and compiling data dis-cussed by other authors researching similar topics.

3.1.2 Observation

This method helps in understanding the actual condition in which the changes in the business model affect the people and assets involved in the system. it provides a good insight and data gathering technique where dynamics of decision making, and power transfer can be seen and understood. It is a method in which data is collected through observation. The Observation data collection approach is classified as a participatory analysis since when taking notes and/or documenting, the researcher must immerse themselves in the environment where their respondents are.

Observation as a means of collecting data may be formal or unstructured. Data collec-tion is carried out using different variables in organized or systematic evaluacollec-tion, and according to a predefined timetable. On the other hand, unstructured research is per-formed in a transparent and autonomous way, in a context where no fixed factors or goals will arise.

The benefits of the observation data collection approaches include direct access to study phenomena, high implementation versatility, and a permanent archive of the phenome-non. At the same time, the observation process is weakened by longer time criteria, strong observer bias, and observer effect on primary data, in a way that the presence of an observer may influence the behavior of sample group elements. It is important to remember the approach of gathering data from observation can be related to other eth-ical problems. Completely informed consent of the participant(s) of study is one of the basic ethical principles that researchers will stick to. At the same time, the actions of study group participants could shift with negative effects on the validity of the research if they are notified about the presence of the observer (Dudovskiy, 2018).

A Case Study was conducted to understand where AR can be used in the logistics sector and utilize its capabilities for improvements according to the actual problems that need rectification. Research questions helped to get deep knowledge from the real-time in-teraction during the case study. Observations were noted and analyzed with reference to the company visit and referred papers. Through this approach, the study phenomenon was versatile and structured.

3.1.3 Questionnaire

Based on the complexity of the questions, questionnaires can be categorized as both a quantitative and a qualitative tool. In specific, the responses received by closed-ended questions with multiple-choice response choices are evaluated using quantitative tech-niques that can include pie-charts, bar-charts that percentages. Answers to open-ended questionnaire questions are analyzed using qualitative approaches, including dialogue and logical thinking without using statistics and estimates.

The benefits of questionnaires include improved data collection capacity, low to no cost criteria, and greater objectivity rates relative to other alternative primary data collection approaches.

Types of the questionnaire used in this study are:

1. Open question questionnaires: Open-ended questions vary from other types of questions used in questionnaires as they can yield unpredictable findings and can make the study more original and useful. Even so, as data is gathered from the questionnaire of open questions it is impossible to interpret the effects of the tests.

2. Multiple choice questions: They are scalable, insightful, and offer clear data that you can easily analyze. Because they have a setlist of answers, they send standardized sample answers to make it easy for the respondents to finish the sample. However, the data you get back is limited to the options you make (Dudovskiy, 2018).

The interview sessions were conducted in 2 phases for the data collection. A mix of open-ended and close-ended questions were prepared. In the first phase, one of the pro-duction managers who handle the night shift was selected and interviewed using these prepared questions. This section was conducted in different stages. In the second phase, a questionnaire was given to 5 operators who were working in the terminal with differ-ent roles (truck drivers, terminal operators). The results from the questionnaire were compiled and analyzed to find the problem areas.

3.2 Research Design

To answer the research questions and provide detailed findings and analysis on the use of AR in the logistic sector, literature review, observation, and interviews are con-ducted. Validation of the information collected will be done through theoretical analysis

Figure 7: Research Design

of various relevant research papers and comparing answers. By using these methods, the implementation possibilities of AR were explored. Figure 7 shows the various steps that are adopted in the thesis.

3.2.1 Qualitative Data Analysis

Qualitative data refers to non-numeric records, such as transcripts of interviews, notes, video and audio files, photographs, and text documents. Qualitative data analysis used in this study are:

• Content analysis: It refers to the method by which verbal or behavioral data are categorized to classify, interpret, and tabulate the data.

• Narrative analysis: This approach involves reformulating the stories told by the respondents considering the meaning of each situation and each respond-ent's various experiences. In other words, narrative analysis is a researcher-led revision of primary qualitative evidence (Dudovskiy, 2018).

The qualitative data analysis was done mainly on the theoretical review which was con-ducted on the bases of interviews, videos, photographs, and documents. Based on these content and narrative analysis, the collected data were tabulated in various sections. 3.2.2 Reliability and Validity

This refers to the degree to which the same responses can be obtained on more than one occasion using the same tools. This means, if the work is associated with high standards of reliability, then other researchers need to be able to produce the same findings under comparable circumstances, using the same testing methodology. It is noted that prob-lems concerning reliability occur in several ways. Reliability is a problem any time a single observer is the data source since we have no definite protection against the effect of the subjectivity of that observer (Babbie, 2010).

The Validity of study can be defined as the degree to which empirical testing method criteria have been fulfilled during the research finding generation process and main ones are specified as; content validity, criterion-related validity, construct validity, in-ternal validity, exin-ternal validity, concurrent validity and face validity (Williamson, 2002). Measures to ensure the validity of research include, but not limited to the fol-lowing points:

1. Appropriate time scale must be chosen for the study

2. A reasonable technique must be selected, taking into consideration research charac-teristics

3. The most suitable sample form must be chosen for the study

4. In no way will the respondents be coerced to make different options from the sets of responses.

It is important to note that while there will never be a complete absence of risks to study reliability and validity, researchers need to try to mitigate this challenge as far as pos-sible (Cohen, 2007). The sources for the study are reliable as the literature used are referred to multiple time in different articles and information from the interviews pro-vide similar output. Most of the data is collected from recent papers which improves the validity of the paper.

4 Findings and Analysis

This chapter describes the data that have been gathered through different data collec-tion methods and how these data have been analyzed in order to produce the outcome. These include case studies, challenges faced by the company, key benefit factors, and providing solutions for by answering the research questions.

4.1 Case Study

For the purpose of the thesis, a company visit was conducted in a Logistics company which is located at Jönköping (Sweden), one of the main logistic hubs in the country. The items are transported to all parts of Sweden. The items that the company mainly deals with are white goods (fridge, oven, induction stove) and Home delivery goods (furniture, TV, AC coolers). The HD products are coming from various parts of Sweden and the white goods from Denmark. The company currently uses barcode technology to identify and sort goods. The company is also involved and keen to find emerging technologies and approaches that will simplify current ways of operation. As a result, the organization has taken the first step towards Industry 4.0 by introducing RFID tech-nology to replace traditional bar code techtech-nology in the previous years. When the pro-ject was introduced to the company, they were interested to gain more understanding of AR technology and its benefits. The aim was to get more understanding of the pro-duction line and various functions of the company where AR technology can be useful. By using this AR technology, we aim to reduce the risk/accidents that may occur in the warehouse, especially for the forklift driver thus making safety a high priority.

The meeting with the Production Manager helped in understanding the different tasks performed on the work floor daily and where they need assistance. Various sections of the work floor including the terminals and the storage area were monitored. According to the manager, there are a lot of positive benefits from the implementation of AR in the warehouse. The main areas that he pointed out for change were for the Storage place, production workflow in the terminal, and in the way of security. The manager explained that the company has a special area for the storage of goods. In that area, the customers tell the company to hold (store) their products in the warehouse and deliver it when the end customer places an order. For this purpose, racks and shelves are ar-ranged and they are marked with barcodes to identify when they need to retrieve the products. Once the product is placed in the storage room, by means of a scanning tool they scan the barcode on the rack and store data into the system. Next time, when the products are needed for delivery, the operator has to go into the system and find the location where it is stored. Likewise, this process is repeated for all the details of other products. Once the locating part is done, he needs to take the hard copies of these data before the operator moves to the storage space and that is how they locate the product. In case of any queries, the operator again needs to go back to the office to use the sys-tem.

The other area is the shop floor, where a variety of items like TVs, fridges, and furniture are handled by the company from different customers. During the production process, the workers need to keep a separate count on each customer and for that, they need to know the details of how many products of different customers that they had handled each day. Right now, they keep the count by themselves (operator) when they scan the products. But sometimes, when a large amount of goods has to be handled, the occur-rence of human error is common. The manager said that they put extra time on the operator’s training when they recruit to make them learn about the production process on the floor. And as the floor is big it takes too much time for the operator to understand all about these processes. The figure 8 below shows the various process that takes place:

Figure 8: Process Flow Diagram

The products that arrive at the terminal are sorted in designated areas. As the floor is big the operators must carry the product while sorting. The manager said that the com-pany has taken a lot of time on finding the best solution for the operators and making them feel comfortable during their work. All these changes take a lot of time to imple-ment and to understand whether they were useful or not for the production process. When we discussed the safety and security of the operators, sometimes it is harder for the forklift drivers to see the blind spots while they drive, especially while loading with and driving around to and fro. The accidents that occur are reported in those scenarios. Due to these conditions, the company is looking for better ideas to rectify these prob-lems. Table 4 below shows the various problem areas and their challenges.

Table 4.Problem area and challenges

Problem area Challenges

P1: Storage

• Traceability of the product.

• Retrieval of Information from the database. • Excessive paperwork.

• Time consumption for location and retrieval of the product.

• Excessive and unwanted human movement. P2: Production Floor

• Optimization of Sorting. • Operators safety. • Training for operators.

4.1.1 Problem area 1: Storage Unit a) Traceability of the product:

Traceability plays a vital role in logistics. It refers to a set of procedures and measures used to identify and track products from the production site to its final delivery point. Due to the increase in the information that is needed to trace the product nowadays, it has become really important to use technologies like RFID and barcodes. In the mentioned company parcels are scanned by the operators and then sorted according to their destination. They must keep a count of the number of goods from each customer and how many have been delivered out. Damage reports of the products have to be counted separately making the operator’s work very dif-ficult. Apart from this, lack of real-time information like the status of the product is hidden from the end-user.

b) Retrieval of Information from the database:

For obtaining the required information on the work floor, the operator needs to physically log in to the company database and search for information every time it is required.

c) Excessive paperwork:

Even after retrieving the required information, operators always have to keep hard copies of these data with them for locating the goods and all the other information related to the delivery of the product. This increases the difficulty of searching for products in the storage unit and adds to the wastage of paper.

d) Time consumption for location and retrieval of the product:

Sometimes even after the initial scanning of goods, information fails to be stored in the database and the operator has no way to know if the scanning was done properly or not. Thus, when the time comes to retrieve the product for delivery there are no other means but to find it without location information and taking a lot of time. Even

when the scanning is done properly, finding the exact location of products in the storage shelf is time-consuming and needs a lot of practice.

e) Excessive and unwanted human movement:

It takes a lot of effort from the operators to move around the work floor to get in-formation from the office then locate products and retrieve it for delivery. Reducing these movements to the minimum can be very useful for work as well as the health of the operator.

4.1.2 Problem area 2: Production Floor a) Optimization of Sorting:

When the volume of arrived products increases, the production floor becomes con-gested and it becomes very difficult to move around and sort. Even though floor design for production on the floor is decided accordingly, there is no way of know-ing if that plan works or is optimum for that volume of products. As there are many different types of products like furniture, electronic goods, and other household items that also have irregular shapes, makes sorting very difficult. Improvements can only be made through trial and error methods which are time-consuming. b) Operator safety:

The forklift drivers on the floor need to be really careful about their surroundings and the way they handle the products. When the products are irregular or too big, driving through the work floor becomes difficult due to blind spots which may lead to accidents. The drivers need to be very skillful as they do not have any navigation support. Finding the exact location for storage also makes the work very complex due to the lack of assistance.

c) Training for operators:

As the operators need to remember and execute a lot of tasks, training them takes a lot of time and effort. Many operations need to be repeated again and again just to train the new operators and prevent accidents. It would be very helpful if some as-sistance is provided while learning new tasks and help remember work procedures. 4.2 Analysis

This chapter shows how AR can help logistics companies in solving identified problem areas and improve their operations. Benefits were found by evaluating the use of AR that can be considered for its implementation. Table 5 illustrates the uses of AR across the warehouse operations which were compiled from various literature with similar problem areas.

Table 5.Uses of AR in warehouse operations (Stoltz, et al., 2017) Problem Area Operations Uses of AR P1: Storage Unit

Receiving & Picking

• AR guidance

• Indicating the allocated space. • Providing alternate solutions. • Assures scanning of all goods.

Storing & Shipping

• Display pictures and details about the product to be picked.

• Informs about errors and disruptions. • Show the best way to place picked items in

the pallet.

• Shows the right location for shipment. • Checking and counting of products.

P2: Production Floor Optimization of Opera-tions • Create simulations • Easier accessibility

• Anticipating bottlenecks & errors. • Technical advancement

Training & Safety

• Opportunity to learn new technology • A better understanding of operations. • AR assistance for forklift drivers.

• Receiving audio instructions besides vis-ual ones

4.2.1 Receiving and Storing

By the implementation of the AR, it helps the operator to show up where the items should be put and how to arrange them on the floor. The system will automatically check the received goods against the delivery report (Real & Marcelino, 2011). Once the product is received, the AR system will help to inform the operator about the allo-cated area to put the product to be stored. For that, they will receive the visual images on the display of the device. It will show up all the information about the product. For all these purposes, it will make a route plan to the storage location (Davarzani & Norrman, 2015). It also indicates the current state of the picker as well as the next step of the process. If once the storage space is filled, it will make an alternate solution and design a new route plan.

4.2.2 Picking and Shipping

Operators will get the allocated task that they need to perform in the storage space. Once the order from the end customer is made, the product must be delivered on the same day. The operator will get a display picture and detail of the item to be picked (Giannikas, et al., 2016). The storage location and the picking route can be visualized

on the device. It also provides or highlights the physical location of the item. Thus, they can avoid the unnecessary use of papers to find out the location. As the wearable device is always connected to the system, it provides real-time data to the operator as they need (Weaver, et al., 2010). The device will scan the item’s barcode to allocate to the product on each rack or shelf. It gives information on preventing congestion in gateways. Dur-ing the shipment process, it will show the best option to place the picked items in the allocated space. It indicates the right location/pallet for the shipment (Davarzani & Norrman, 2015). It shows where to put each order on pallets/ on trucks based on the type of order, destination, and fragility. The AR helps to check, count the products and orders to be loaded on the truck during shipment.

4.2.3 Optimization of operation

With the use of AR, it is possible to create simulations to formulate the floor plans and can visualize possible ways in which the work floor can be used optimally (Damiani, et al., 2016). Even when the products are irregular or big in size, they can be easily handled using the simulations created. All the information required for operations can be easily accessed from anywhere in the warehouse. It helps in anticipating bottlenecks and error in the process which can be removed for faster movement of work on the production floor (Stoltz, et al., 2017). By using this technology, the operator does not need to keep a track on each product on counting, because while the system is scanning, it itself keeps a separate count by differentiating each customer. And this can help reduce the work-load of the operators.

Many administrative and office tasks like conducting meetings, creating company presentations (visualizing graphs, bar diagrams) through AR will help to boost the com-pany’s efficiency and show its technical advancements (Panetto, et al., 2019). At the company level, AR can provide new talents with the information distributed throughout the organization. The use of AR technology also indicates that the company is adjusting to the latest innovations in the industry.

4.2.4 Training and safety

By giving operators, the opportunity to learn with new technology like AR, it will help them to plan, and help them to have a better understanding of the production operations before they have to face the real-time constraints. This allows new employees to explore and get information to require for immediate guidance on tasks. This will help reduce the time spent on training the employees and can be utilized for improving the produc-tivity of the process (Damiani, et al., 2018). For the forklift drivers, the AR assistance can be provided so that it can do tasks like scanning, navigating, and alert the driver if there are any obstacles. Employees can undergo simulations of possible emergencies that are lifelike. Instead of worrying about what to do if accidents like a fire happen, an individual could conduct the scenario in a secure 3D setting.

5 Discussion

This chapter is mainly divided into two sections where the first part deals with narrow-ing down the research from the data collected through findnarrow-ings and analyznarrow-ing it to an-swer the research question.

5.1 Discussion of Findings

Using the data collected from the findings and analysis, answers for the research ques-tions were compiled based on the research design. Validity and reliability of the infor-mation collected were done through theoretical analysis of various relevant research papers and comparing answers from interviews. By using these methods, benefits, and barriers of implementation possibilities of AR were understood and presented in the result. Figure 9 shows the sequence of approach to the thesis.

Figure 9: Sequence of Operation

Nowadays, the use of AR is more common through its ability to supplement real-world environments directly or indirectly with visualized characters obtained with special hardware, software, and accessories. The concept of AR refers to all activities whose main objective is to increase the real-world environment with virtual knowledge which enriches human senses and abilities. AR succeeds in combining virtual information with the real world. The AR framework alters digital information in the physical world. There are several devices available on the market to create a concrete AR system and use one technical solution or the other. Manufacturers forecast the future of AR prod-ucts year by year and endorse new models of hardware with improved features and results (Cirulis & Ginters, 2013).

Research Questions Literature Review Questionnaire/ Interview Findings Analysis Validation Final Result