Paulina Haide Becerril Palma
ENHANCING LEAN AND AGILE WAYS OF WORKING USING AN ASYMMETRIC MULTI-USER VIRTUAL REALITY SETUP
Faculty of Information Technology and Communication Sciences (ITC) M. Sc. Thesis
August 2021
ABSTRACT
Paulina Haide Becerril Palma: Enhancing Lean and Agile ways of working using an asymmetric multi-user virtual reality setup
M.Sc. Thesis Tampere University
Master’s Degree Programme in Computer Science - Human-Technology Interaction August 2021
At present, the relevance of studying VR to ease industry processes has increased in importance as several studies indicate that VR solutions might enhance future production models that use centralized digital data. However, there is lack of research on the specific use cases, stakeholders, and requirements the VR solutions should meet to fulfil the needs proper of the industry. Therefore, the goal of this research was to analyze a multi-user VR application to aid the maintenance and technical documentation process in the industry. Due to the COVID-19 restrictions and in adherence to the Lean and Agile principles, the multi-user VR evaluation was conducted using an asymmetric VR setting.
Asymmetric VR settings refer to the use of immersive and non-immersive platforms simultaneously, in this case HMD and desktop VR. The virtual environment VE used to in this user study is a Collaborative Virtual Reality Environment, COVE-VR. The COVE- VR is a software created by Tampere University in collaboration with KONE. The VE was designed to aid the collaboration in the industry with a focus on digital content creation. The user study took part in the elevators and escalators company KONE, with participants from sites in Finland, China, India, and USA. The participants of the study represented 20 end-users of the industry. The results of this study showed that as VR has the potential to optimize industrial tasks in a Lean and Agile manner, especially for geographically dispersed teams. Moreover, the introduction of asymmetric VR settings can enable the participation of wider audiences, easing rapid 3D visualization, avoiding travelling costs, and reducing hardware and software cost in comparison to HMD platforms. Lastly, the study demonstrate that VR promotes remote collaboration and sense of co-presence even for desktop VR users.
Keywords: Human-computer interaction, Virtual Reality, Multi-user Virtual Reality, User-centered Design, Experimental Research, User Experience, Virtual Maintenance, Usability test, Asymmetric Virtual Reality.
The originality of this thesis has been checked using the Turnitin Originality.
Acknowledgements
I would like to acknowledge everyone who help me finish this thesis work.
Thanks to the KONE and Business Finland co-founded project HUMOR, for providing the means for me to conduct this research.
I would like to thank my professors and all personnel at Tampere University that welcomed me to Finland with open arms and gave me an opportunity to pursue my graduate studies. Thanks to Simo Ahonen for outlifting me on my moments of doubt during my studies. To Prof. Markku Turunen for procuring industry-academia research projects and encourage students to actively learn from both contexts. In particular, I am grateful to my mentor and thesis supervisor Dr. Jaakko Hakulinen for believing in me and guiding me on my learning process since I started my graduate studies.
Thanks to the team at KONE that supported me and encouraged me to conduct my research. To Hanna Heinonen, Viveka Opas and Phong Truong for sharing your expertise in support of this research. My sincere gratitude to Dr. Sanni Siltanen for your advice, guidance and for being an inspiration for young female researchers like me.
To Pierre Olivier for giving me strength and caring support, thanks for being by my side and for giving me a hopeful future. My heartfelt appreciation to my family and friends, especially to my dad and siblings that have always supported me to pursue my dreams and aspirations.
Lastly my utmost gratitude and recognition is to my mom. You were the most extraordinary, loving, and strong mom and woman. All I am and all I will be is because of you. I will miss you in every step I make, you will always live in my heart.
¡Mi más sincero y profundo agradecimiento!
Contents
1 Introduction ... 7 2 Related work ... 10
2.1 Human-computer interaction 10
2.2 Methods in human-computer interaction 13
2.3 Virtual Reality 16
2.4 Multi-user virtual reality 20
2.5 Virtual reality in the product development process 23
3 The users and context ... 29
3.1 Lean and Agile context 29
3.2 Users, stakeholders, and tasks 32
3.3 Virtual reality at KONE 36
4 Virtual reality in the KONE process ... 37
4.1 Software 41
4.2 Hardware 43
5 User study ... 44
5.1 Selection of methods 44
5.2 Procedure 46
5.2.1 Introduction 49
5.2.2 Video training 50
5.2.3 Hands-on training 50
5.2.4 Multi-user usability test 51 5.2.5 Semi-structured interview 52
5.2.6 Online survey 53
5.3 Data analysis 53
5.4 COVID-19 Protocols 54
6 Results ... 56
6.1 User experience 56
6.2 Multi-user VR enhances teamwork 60
6.3 Lean and Agile working process 62
7 Discussion ... 66
7.1 Multi-user asymmetric VR in the industry 66
7.2 Multi-user VR and the Lean and Agile principles 68
7.3 Asymmetric VR settings: use cases in the industry 70
7.4 User Experience elements of COVE-VR 73
8 Conclusions ... 76 9 Appendix ... 92
9.1 Appendix A: Current ways of working questionnarie 92 9.2 Appendix B: COVE-VR single-user case questionnarie 92 9.3 Appendix C: COVE-VR asymmetric multi-user setting questionnaire 93
9.4 Appendix D: User study tasks list 94
9.5 Appendix E: Online Survey 97
9.6 Appendix E: COVID-19 protocols 98
List of Acronyms
AR Augmented reality
CAVE Cave automatic virtual environment CVE Collaborative virtual environment
COVE-VR Collaborative virtual reality environment CAD Computer aided design
xR Extended reality GVM Global virtual models HMD Head-mounted display
HUMOR Human optimized extended reality HCD Human-centered design
HCI Human-computer interaction KTI KONE Technology and Innovation MVP Minimum viable product
MR Mixed reality PLC Product life cycle RV Reality-virtuality
TPS Toyota production system UX User experience
UCD User-centered design VE Virtual environment VM Virtual maintenance VR Virtual reality
About KONE
This research was conducted inside KONE Technology and Innovation unit, where the author had a thesis worker contract. The aim of this study was to examine the implications of multi-user virtual reality application in the product development process through a use case at KONE Corporation (later referred as KONE). The mission of KONE is to improve the flow of urban life. As a global leader in the elevator and escalator industry, KONE provides elevators, escalators, and automatic building doors, as well as solutions for maintenance and modernization to add value to buildings throughout their life cycle. Through more effective People Flow®, KONE makes people's journeys safe, convenient, and reliable, in taller, smarter buildings. In 2020, KONE had annual sales of EUR 9.9 billion, and at the end of the year over 60,000 employees. KONE class B shares are listed on the Nasdaq Helsinki Ltd. in Finland.
KONE is continuously advancing efforts to continue its innovation leadership in the market, as such the company has been actively involved in extended reality (XR) research projects. Starting in the first quarter of 2020, KONE and Tampere University researchers started to work collaboratively in the Human Optimized Extended Reality (HUMOR) project. HUMOR is a co-funded Business Finland project which aims to research the impact of XR in the industry and academia, among others. In particular, KONE’s use case aims to study and evaluate the usefulness of virtual reality throughout the product development process. This thesis was conducted at KONE with the support of the maintenance method development and technical documentation teams in Finland, India, China, and USA.
1 Introduction
The feasibility and importance of adopting virtual reality (VR) as part of the product development process has been widely studied since VR became mainstream in the industry [Kovar et al., 2016; Navas et al., 2020; Ottosson, 2002; Wolfartsberger et al., 2020]. As such, VR has been studied as a tool to aid ergonomic evaluations [Di Gironimo et al., 2013; Dong et al., 2013; Gomes de Sá and Zachmann, 1999; Guo et al., 2018], supplement the product design review [Choi et al., 2015; Freitas et al., 2020;
Wolfartsberger, 2019; Wolfartsberger et al., 2020], and even to enhance cross- disciplinary teamwork [Çöltekin et al., 2020; Du et al., 2018; Gaoliang et al., 2010; Li et al., 2003; Rand et al., 2005; Steed et al., 2012; Wolfartsberger et al., 2020]. At present, the relevance of studying VR to ease industry processes has increased in importance as several studies indicate that VR solutions might enhance future production models that use centralized digital data [Kovar et al., 2016; Navas et al., 2020; Wen and Gheisari, 2020]. At present, virtual reality has even become a suitable alternative to cope with the COVID-19 pandemic remote work recommendations [Domb et al., 2021; Sasikumar et al., 2021]. However, it remains unclear how VR applications could be integrated as part of the industry processes and milestones. Moreover, there is lack of research on the specific use cases, stakeholders, and requirements the VR solutions should meet to fulfill the needs proper of the industry.
Figure 1.1: Example of VR application to support the product development process
The goal of this research was to analyze a multi-user VR application to aid the maintenance and technical documentation process at KONE. Due to the COVID-19 restrictions the multi-user VR evaluation was conducted using an asymmetric VR setting.
Asymmetric VR refers to the use of immersive and non-immersive platforms simultaneously, in this case head-mounted display (HMD) and desktop VR [Horst et al., 2020; Ouverson and Gilbert, 2021; Steed et al., 2012]. The virtual environment (VE) used to in this study is a Collaborative Virtual Reality Environment (later referred as COVE- VR). The COVE-VR system was designed to enhance collaboration in a VE, focusing on digital content creation [to appear Burova et al. 2021]. The findings of this research aim to analyze COVE-VR application as a tool to strengthen the Lean and Agile principles that prevail in industry. Based on the aforementioned goals, the research questions of this study aim to investigate the potential use cases in the process, effects from the Lean and Agile perspective, users and stakeholders, and elements of the user experience of the COVE-VR system. The research questions this study looks at are:
• Question 1: What are the effects of implementing multi-user collaborative virtual reality applications in the maintenance method development and technical documentation process?
• Question 2. How do the ways of working change when multi-user collaborative virtual reality applications are introduced into the workflow following a Lean and Agile point principles?
• Question 3. Who will be the key users of the application throughout the process?
• Question 4. What are the elements of the subject matter expert’s user experience?
The first research question investigates the current working process to subsequently analyze how the COVE-VR application influence internal processes. The second question focuses on analyzing VR as a tool to enhance Lean and Agile management philosophies.
The third question investigates the stakeholders and roles of users as VR is adopted in the internal process. Lastly, the fourth question investigates the COVE-VR user experience from the KONE employees’ perspective. In order to do it so, this study uses the user experience (UX) model proposed by Hassenzahl [2018] as a reference to have a consistent terminology of UX. For the purpose of this thesis, I will use the human-centered design (HCD) approach as a framework to answer my research questions. According to the ISO on HCD [ISO, 2019] any design process that follows a HCD approach should follow six design principles detailed in the Table 1.1.
Table 1.1 The six principles of the human-centered design process
This thesis was part of the Business Finland co-innovation project, HUMan Optimized extended Reality (HUMOR). The HUMOR project included 7 Finnish universities and +20 companies. This study was a collaborative research project between of KONE and Tampere University. In this thesis, I analyzed the COVE-VR application as a tool to enhance the maintenance methods development and technical documentation process at KONE. The study and sections of this thesis respond to the HCD principles. The overview of the HUMOR research activities and their relevance to this thesis are visualized in the Figure 1.2. In this study first, I will provide an overview of past relevant studies the (Chapter 2. Related work). In Chapter 3 (The users and context) I will describe the users, current work tasks, and the environment. Thirdly, in Chapter 4 (Hypothesis of virtual reality in the process) I will describe the hypothesis to formulate the user study, as well as the hardware and software used in the evaluation. In Chapter 5 (Evaluation methods) I will describe the methods selected for the evaluation and the evaluation procedure. In the Chapter 6 (Results) I will provide an overview of the results of the evaluation.
Figure 1.2: HUMOR research activities and the thesis structure
Design based upon explicit understanding of users, tasks, and environments Users are involved throughout the design and development
The design is driven and refined by user-centered evaluations The process is iterative
The design addresses the whole user experience
The design team includes multidisciplinary skills and perspectives
2 Related work
In this chapter I will present a set of previous studies and trends on virtual reality, design and evaluation methods on virtual reality, and virtual reality applications in industrial settings. Firstly, I will introduce the terms human-computer interaction, human- centered design, and user experience. Subsequently, I will provide an overview of virtual reality VR, and multi-user VR solutions. Lastly, I will introduce previous work on virtual reality in the product development process.
2.1 Human-computer interaction
In an age where technology keeps increasing in complexity, new fields of study evolved to improve computer-based solutions. Human-Computer Interaction (HCI) is a field of study that aims to ease the user’s understanding and usage of intrinsically complex novel technology. Moreover, HCI aims to ensure creators develop meaningful and useful solutions. Due to the complex nature of HCI, researchers define it as the cumulative study of different fields at once. For instance, Holzinger and Pasi [2013] argue that HCI consists of the study of human characteristics such as comprehension or understanding of information, computer science, and design. The authors also argue that HCI analyzes the interplay of information that occurs between a computer-based solution and the user.
Although historically HCI studied productivity applications and was mainly oriented to construct usable experiences [Jacko, 2012], nowadays the field has expanded to encompass visualization, information architecture, and collaborative systems among others.
Figure 2.1: Fields influencing human-computer interaction [Human-computer interaction.2018]
Since HCI is studying novel solutions from the human perspective [Jacko, 2012], researchers have proposed methods to design and evaluate HCI solutions from a holistic perspective. For instance, researchers have proposed different methods to support the human-computer interaction design process [Maguire, 2001; Maloney-Krichmar et al., 2004]. Maguire [2001] concluded that, within human-computer interaction, the human- centered design (HCD) approach complemented the software development by including the user needs and requirements throughout the development of a computer-based solution. On the other hand, Norman [2005] defined human-centered design where the goal is that technology adapts to the person and as such new products should have proposed different methods to support the human-computer interaction design process focus on the user’s requirements and capabilities. Currently, the ISO 9241-210 [ISO, 2019] on human-centered design for interactive systems defines HCD as “approach to systems design and development that aims to make interactive systems more usable by focusing on the use of the system and applying human factors/ergonomics and usability knowledge and techniques”. In addition to HCD methods, Norman [1986] and Maloney- Krichmar et al. [2004] described subsets of human-centered design methodologies called user-centered design (UCD). According to the authors the aim of user-centered design is to involve the user throughout the overall design process by a variety of participatory methods. In everyday work both HCD and UCD are used similarly, however according to several authors [ISO, 2019; Gondomar and Mor, 2020; Gasson, 2003] they differ so that HCD affords a wider scope of study that includes the participation of stakeholders and end-users similarly. The inclusion of stakeholders and users in the design process complements the study of systems that were not designed to be actively used but are used by consequence such as pervasive systems [Howard, 2012]. Pervasive systems are systems that can appear in any format, anytime, and everywhere, an example can be found in tracking applications.
In addition to design methods in HCI there are subsets within the field that aim to study specific aspects of HCI. The term user experience (UX) was adopted to describe the study of pragmatic and hedonic aspects users experience when using novel technology [Hassenzahl and Tractinsky, 2006]. Hornbæk and Hertzum [2017] and van Schaik and Ling [2011] argue user experience is focused on the study of pragmatic and hedonic mechanisms while other models such as the technology acceptance model (TAM) [Venkatesh and Davis, 2000] aim to analyze the solutions solely from a utilitarian perspective (easiness of use and usefulness). On the other hand, the ISO 9241-210 [ISO,
2019] describes UX as “a person’s perceptions and responses that result from the use or anticipated use of a product, system or service”. Since UX encompass a wide range of characteristics from usability to cognitive and even affective aspects, several authors have argued that one the challenges of UX evaluation starts by defining what UX entails.
The lack of a formal UX definition hinders the selection of methods to evaluate the aspects of the UX that the analysis aims to answer [Law and van Schaik, 2010].
Consequently, several authors [Hassenzahl, 2018; Hornbæk and Hertzum, 2017; Law and van Schaik, 2010; Minge and Thüring, 2018a] have claimed the best way to analyze UX is by creating systematic models to define what UX entails.
There are several benefits of UX models as they provide a better understanding for researchers to analyze how users perceive and value the novel solutions, thus offering a common ground on the key elements to evaluate [Hassenzahl, 2018]. At the moment, there are different UX models available that aim to provide different perspectives on how to analyze user experience. For instance, Mahlke and Thüring [2018a] proposed a model to integrate the interaction characteristics, qualities, and overall judgements and reactions from the system. Other authors have claimed UX is dynamic and fluctuates over time, for this reason it should be studied throughout the usage lifecycle. As such, Karapanos et al.
[2009] proposed a long-term UX model that aims to evaluate UX throughout the product’s life cycle. Their model studies the UX in four stages: orientation, incorporation, identification, and anticipation. Although this type of construction of UX over time could provide further insights on how to evaluate any solution it also provides limitations to the evaluation as they need to be conducted consistently throughout the usage.
In this thesis, I used the User Experience model proposed by Hassenzahl [2018] as a reference to have a consistent terminology of UX. This model aims to define they core elements of UX and their functional relationships from the designer and the user perspective (Figure 2.2: Hassenzahl’s user experience model). Firstly, the model describes how single users construct a particular perception of the product or apparent product character. Secondly, it describes the apparent product character as pragmatic or utilitarian and hedonic or emotional attributes. Lastly, the model refers to the fact that the perception of the product is moderated by the situation of use and lead to consequences such as appeal, pleasure, or satisfaction.
Figure 2.2: Hassenzahl’s user experience model [2018]
2.2 Methods in human-computer interaction
Currently, human-computer interaction HCI aims to study the characteristics that ultimately define how individuals interact and experience computer-based solutions.
Within HCI there exist different frameworks that aim to research users or evaluate the attributes of novel technological solutions, such as virtual reality, visual interfaces, robots, and other aspects of technology. The use of evaluation frameworks enables researchers to analyze and measure how successfully the hedonic and pragmatic characteristics (UX elements) of the product accomplish their intended purpose. The evaluation of UX can be carried out in two different ways: formative evaluation or a summative evaluation [Greenstein, 2010]. The first one aims to find the areas of improvement to inform the iterative design process, while the latter one aims to determine the overall quality of the UX of a finished product. At present, there exist a range of research methods to define the UX characteristics of the products [Lazar et al., 2017;
Nunnally and Farkas, 2016]. Consequently, the methods can be subdivided into several categories that vary depending on their aim and type of evaluation.
For the purpose of this thesis, I will describe the evaluation methods I used to answer my research questions based on the data collection type: observation-based, and opinion-based. The first group of evaluations are the observation-based evaluation methods. The observation-based methods aim to collect data about the user’s behaviors [Rohrer, 2014]. According to Rohrer [2014] and Nielsen [1994] these evaluations involve
researchers who analyze the behavior of the user in the context of use. Ethnography studies and usability tests are examples of observation-based methods [Rohrer, 2014]. In general, a usability test involves a group of users that represent the user population and a set of tasks that represent real-life situations [Lazar et al., 2017]. The usability testing as a research method can be used to analyze and learn how users interact and understand the product without prior experience, hence focused on the intuitiveness of the product tested [Nielsen, 1994]. According to Lazar et al. [2017] the range of usability evaluations is quite broad as the approach can be adapted depending on the purpose of the study (See Figure 2.3: Example of VR usability testing setup). For instance, usability test can be done in-person and remotely. Remote usability evaluations support conditions where the evaluators and the users are in different locations [Andreasen et al., 2007; Chalil Madathil and Greenstein, 2011; Dray and Siegel, 2004a; Lazar et al., 2017]. Remote testing can be conducted in synchronous or asynchronous formats. The synchronous format involves real-time communication between the user and the evaluator while in the asynchronous both parties work separately. Despite the characteristics of the study, in general usability testing aims to improve the quality of an interface by finding the flaws that need improvement [Lazar et al., 2017]. Regardless of the type of usability study conducted, often times usability evaluators introduce introspection protocols to ease the analysis of the user’s sensations and thoughts [Matlin, 2002]. In addition, introspective protocols also help facilitators acquire information about the user’s mental model or internal representation of how things are or seem to be. In usability testing the think-aloud methodology [Ericsson and Simon, 1993] is commonly used to foster and procure the user’s introspective exercises.
Lastly, the opinion-based evaluation methods are characterized by providing information about the user’s way of thinking or attitude of users [Rohrer, 2014]. This type of methods can help determine which features to implement. Some examples of user’s feedback evaluation methods are surveys, interviews and focus groups. For instance, the semi-structured interviews are a type of interview that combines a mix of predefined and fixed questions and open questions throughout the course of the interview [Wilson, 2013].
Semi-structured interviews can help the researcher gather information systematically following a central topic. However, it gives freedom for further exploration when there are issues, or some other relevant topics emerge. For instance, focus group interviews allow individuals discuss while being guided by a moderator [Garmer et al., 2004; Lazar et al., 2017; Shah and Robinson, 2007]. The discussion can be conducted to inquiry the
users’ previous experiences or as a reflective exercise after a user test. The reflective exercises occur usually in retrospect; thus, it is likely to include group interviews after the usability testing [Garmer et al., 2004]. Likewise, other opinion-based research methods widely accepted in HCI are surveys. A survey is a set of predefined questions which an individual is asked to respond to [Lazar et al., 2017]. Surveys are usually self- administered and are not monitored by the researcher. Is important to note that within HCI there exist standardized surveys in the form of questionnaires that researchers have designed, analyzed, and proven to produce comparable findings in different UX evaluation contexts. The standard questionnaires in HCI aim to assess different aspects of the UX experience, such as immersion in virtual reality, user interfaces, and the appearance, among many others. In addition, standard questionnaires often use scales to ease the quantitative evaluation of items in the questionnaire. One of the most commonly used scales in UX questionnaires is the Likert scale [Norman, G., 2010]. This scale helps to measure attitudes based on statements that the researchers define. The participants then rate the strength of the agreement towards the statements.
Figure 2.3: Example of VR usability testing setup
There are strengths and weaknesses to all the different types of research methods of UX and one of the goals of this work is to find the most appropriate ways to evaluate the system so that to the method match the goal of the study. For instance, the effectiveness of usability tests to discover issues and improvements has been widely accepted in the HCI community [Andreasen et al., 2007; Chalil Madathil and Greenstein, 2011; Lazar et al., 2017; Maguire, 2001; Nielsen, 1994]. However, authors [Garmer et al., 2004] have
debated the use of usability tastings as isolated tools to discover usability issues.
Moreover, it is a common practice to complement usability evaluations with another research method depending on the aim of the evaluation. For instance, Garmer et al.
[2004] argued that group interviews provide an opportunity for participants to discuss issues discovered in the usability test. Alternatively, Lazar et al. [2017] stated that one of the strengths of the surveys is their ability to provide a general perspective of attitudes of geographically dispersed participants. In addition, the researchers argued that survey data could provide quantitative data in a fast manner. Moreover, Lazar et al. [2017] concluded that one of the benefits of standardized and validated measurements is that nowadays there exist a wide range of validated measurements to validate a wide range of phenomena such as virtual reality. However, according to Lazar et al. [2017] the data collected through surveys might lack the deepness that other methods, such as ethnography studies could provide. On the other hand, Wilson [2013] argues that one of the weaknesses of interviews is that the experience of the interviewer affects to data and that it requires consistency among the interviews in the study. Nevertheless, Wilson [2013] claims semi- structured interviews are ideal to address complex topics as they enable clarification and probing.
2.3 Virtual Reality
In the “The ultimate display” Sutherland [1965] pioneered concepts of immersive technologies that later helped frame what we know today as immersive technologies, among them virtual reality. In his work, the author argued computer interfaces should serve as many senses as possible, thus following multimodal interaction protocols.
Multimodal interaction refers to the stimulation of more than one sense at the time. In addition, Sutherland [1965] stated that multimodal computer-generated objects should not necessarily follow the physical reality limitations we know.
Most recently, technology advancements have made it possible for developers to design immersive displays that serve users’ senses in a natural way [Fuchs et al., 2011;
Gutierrez et al., 2008; Hale and Stanney, 2014; Milgram et al., 1995; Schroeder, 2006;
Sherman and Craig, 2003]. One of the most commonly accepted taxonomies of immersive technology was proposed by Milgram et al. [1995]. In their research, authors framed mixed reality (MR) as real and virtual objects presented together in a display within a spectrum of reality to virtuality that they defined as the reality-virtuality (RV) continuum (Figure 2.2 The reality-virtuality continuum). As part of the reality-virtuality RV
continuum, Milgram et al. [1995] described augmented reality (AR) and virtual reality VR. On one hand, they described AR as added stimuli to an environment that is mainly real. On the other hand, they defined VR as a completely synthetic world which may or might not mimic the properties of a real-world environment.
Figure 2.4: The reality-virtuality continuum [Milgram et al., 1995]
It was until the late 80s when Jaron Lanier coined the term virtual reality in an attempt to define a theoretical framework to examine the relationship of telepresence mediums [Steuer, 1992]. However, at present there is no formal definition of virtual reality, as it is a technology that is constantly evolving [Fuchs et al., 2011; Hale and Stanney, 2014;
Sherman and Craig, 2003]. As such, several authors have attempted to define what virtual reality entails. For instance, Gutierrez et al. [2008] described VR as the ability of users to interact and navigate in realistic 3D settings. According to Gutierrez et al. [2008]
interaction as the capability to choose and control entities in the setting, and navigation as the ability to freely explore the 3D scene. On the other hand, Hale and Stanney [2014]
argued VR offers solutions that provide convincing, life-like experiences to the user’s senses. Similarly, Sherman and Craig [2003] described virtual reality as a technology through which we can experience an imagined reality with our physical senses. Authors also claimed there are four elements to experience virtual reality: virtual world, immersion, sensory feedback, and interactivity. In their definition, virtual world is an imaginary space and/or the description of objects in a space and the rules that govern such objects. Thus, it can exist in the mind of its originator or be displayed through a virtual reality system. Immersion is the sensation of being in an environment and is composed of physical (sensory) and mental immersion elements. Physical immersion is the use of a VR system to replace or augment the stimulus to the user’s senses. While mental immersion is the sense of presence within an environment. Thus, physical immersion is a characteristic of VR, while mental immersion could be the goal of the VR systems.
Sensory feedback is the ability of VR to provide feedback to the senses based on the actions the users have in the virtual environment. The VR system must be capable of tracking the user’s physical location to provide sensory output. Interactivity is the ability of system to respond to the user’s actions. It also refers to the ability that allows users change they viewpoint within a world. The interactivity allows users to move physically within the world and gain new perspectives through their navigation in the environment.
Although Sherman & Craig [2003] definition encompass some key elements of VR, it lacks to highlight other relevant elements such as the naturality of the immersion and the system immediate responses. Conversely, Fuchs et al. [2011] definition of VR includes the four elements to experience virtual reality proposed by Sherman and Craig [2003]
plus elements to describe the response time and natural ways of interaction. Therefore, in order to keep consistency in this study, this research utilizes the definition of VR proposed by Fuchs et al. [2011] as follows:“…uses computer science and behavioral interfaces to simulate a virtual world the behavior of 3D entities, which interact in real time with each other and with one or more users in pseudo-natural immersion via sensorimotor channels.”
In their definition computer science refers to the hardware and software requirements that support the creation and interactivity with the virtual environment. Behavioral interfaces encompass motor and sensorial interfaces; motor interfaces inform the VR system about the user status, while sensorial interfaces simultaneously inform users about the VR system status through the senses. The term real-time interaction is used to describe when the user does not perceive time lag between their actions and the sensorial response. Lastly, the term pseudo-natural immersion is included to clarify that the immersion cannot be natural as users learned to experience reality in the real world, thus the dynamics in VR are intrinsically not natural to users.
Nowadays, users can experience immersive content such as virtual reality through a variety of hardware solutions. For instance, Cruz-Neira et al. [1992] described a room- size space where images are projected in the walls so users can get a sense of an immersive VR environment by being surrounded by the projections. Authors called this type of immersive visualization as cave automatic virtual environment (CAVE) (See Figure 2.5 CAVE users at the Idaho National Labs (INL). Photograph courtesy of INL).
Other platforms include hardware similar to common helmets where users can experience virtual reality controlling the interaction by the movements of their head. According to Chen et al. [2013] interacting VR systems using movements is the best method to control
VR. These systems are commonly referred as head-mounted display (HMD). The HMD hardware positions a display in front of the user’s eyes by adjusting the helmet-like to the user’s head. The VR HMDs aim to occlude and conceal the visual field of the user to increase the immersion of the computer-generated content [Hua, 2017].
Figure 2.5 CAVE users at the Idaho National Labs (INL). Photograph courtesy of INL
At present, some examples of VR HMD in the market are HTC Vive1, Oculus Quest 22, Varjo VR-33, HoloLens 24 among others. User’s interaction with the VE varies depending on the type of head-mounted display they operate [Qian and Teather, 2017]. Some HMD provide interaction with the VR environment through the controllers, as such the controllers are used to gather the actions of the user (input) to transmit to the VR system.
This is the case for systems such as HTC Vive or the Oculus Quest family (See figure 2.6). On the other hand, HoloLens 2 operate primarily by using the user’s head movement.
In addition, there exist other forms of HMD interactions with the system. For example, HMD VR systems can be controlled through the user’s movements of the eyes. This selection method is called eye-tracking and it is controlled depending on how long the user stares at a specific object [Hou and Chen, 2021]. The staring time to trigger time is
1 https://www.vive.com/eu/product/
2 https://www.oculus.com/quest-2/
3 https://varjo.com/products/vr-3/
4 https://www.microsoft.com/en-us/hololens
called dwelling time. Generally speaking, this last technique is getting more popularity in the commercial VR systems as they become more stable and precise.
Figure 2.6 Oculus Quest 2 Headset with Controllers [Oculus quest 2.2021]
In addition to immersive VR platforms, there exist non-stereoscopic platforms that allow users experience virtual reality worlds using non-immersive virtual reality displays. An example of the non-stereoscopic platform is desktop VR. According to Robertson, Czerwinski et al. [1997] desktop VR is the use of interactive 3D graphics to visualize virtual worlds using a desktop display without headtracking. While desktop VR platforms lack some immersive features proper of VR, they might improve the user experience in other ways [Makransky et al., 2019]. For instance, Lee and Wong [2014] argues that desktop VR enhances the user experience of users with lower spatial capabilities. In addition, adopting a desktop VR solution might potentially be a cheaper option in comparison to HMD or CAVE VR environments [Li et al., 2003]. Nowadays, it is common that VR applications are available in different platforms alike: HMD, desktop, and/or CAVE.
2.4 Multi-user virtual reality
Single-user VR systems described in the previous section focus on multimodal immersive interaction for one user at a time. However, researchers have argued VR systems should aim to enhance multimodal interaction communication, and teamwork of multiple users simultaneously [Churchill and Snowdon, 1998; Lee, J. et al., 2020]. As such several researchers have analyzed the features and protocols of multi-user VR systems such as multimodality, sense of presence, among others [Churchill and Snowdon, 1998]. Research on multi-user virtual reality have led to the study of collaborative virtual environments (CVE), virtual environments where multiple users can collaborate under same conditions. CVE was described by Churchill and Snowdon [1998] as distributed
VR systems which allow users to interact with other users or environments. Similarly, Benford et al. [1996] described collaborative virtual environments CVE as a networked virtual world to support group work. They argued three elements should be present in the system for it to be considered a CVE: the users need to be represented to one another in a graphical form, users should be able control their own viewpoints, and users should be able to interact with each other. According to Benford et al. [1996] the aim of CVEs is to provide an integrated, explicit, and persistent context for operation which combines both users and the users’ information into a common display system. In a similar way, Schroeder [2006] proposed three dimensions to which users could experience virtual environments: presence – “being there”, copresence – “being there together”, and connected presence – “the extent to which presence and copresence are mediated through a medium”. Most recently Mestre [2015] broaden the definition of immersion and clarified that immersion referred to the sensations induced by the technology building the VR experience, while presence referred to the experience of having left the real world and being presence in the virtual world. Thereafter, the inclusion of collaborative and social elements helped shaped what we know as multi-user VR as the simultaneous interaction and copresence of users and elements in a shared virtual environment [Benford et al., 1996; Schroeder, 2006].
Initially researchers focused on studying multi-user VR systems as distributed systems, in other words the VR system’s memory shared over a network [Carlsson and Hagsand, 1993] where multiple users could join in. As CVE studies progressed researchers concluded CVEs provided symmetric or equal immersion to the users.
However, some authors argued CVE limited or hindered the immersion of users not wearing HMDs, also called non-HMD users [Duval and Fleury, 2009; Gugenheimer et al., 2018; Horst et al., 2020; Jansen et al., 2020; Lee, J. et al., 2020; Steed et al., 2012;
Wolfartsberger et al., 2020; Xu et al., 2019]. For this reason, later research on multi-user VR applications focused on analyzing different conditions in which users could experience the virtual environment, including no-HMD participation (See Figure 2.7 Non-HMD user interacting with the VE using a tablet [Wolfartsberger et al., 2020]). In particular, researchers analyzed VR systems that do not offer all users the same interaction and/or roles in the virtual scene. The subclass of VR systems that provide different types of immersion or roles differentiation was later described as asymmetric virtual reality [Horst et al., 2020; Lee, J. et al., 2020; Ouverson and Gilbert, 2021]. There are several examples of asymmetric VR and AR applications in the literature. For
instance, Steed et al. [2012] proposed a VR system to ensure technical symmetry so all sensory cues were available to (HMD and non-HMD) users, and to pursue social symmetry so users would have the same ability to interact with each other. Steed et al.
[2012] solution consisted of a CAVE-based VR system [Cruz-Neira et al., 1992] and an avatar embodiment for the remote participants in the remote location. On the other hand, Jansen et al. [2020] developed an asymmetric augmented reality solution to combat the exclusion of collocated non-HMD users and foster social interaction between HMD and non-HMD users. Their solution provided the HMD user with augmentations through the HMD at the same time that the HMD projected augmented content onto a planar surface in support of the non-HMD user.Lee et al. [2020] analyzed an asymmetric VR HMD solution to optimize the roles for non-HMD and HMD users. Their experiment consisted of a collaborative game where non-HMD users had three different types of roles:
omnipresent /immersive role, assistant, or audience role. In their research, Lee et al.
[2020] confirmed the different roles and levels of immersion provided a satisfactory immersive experience for both HMD and non-HMD users. Lastly, Wolfartsberger et al.
[2020] described an application that allowed non-HMD users interact with the VE using a tablet. When non-HMD users pointed an item in the VE, all users in the session would see the item highlighted in the VE (See Figure 2.7).
Figure 2.7 Non-HMD user interacting with the VE using a tablet [Wolfartsberger et al., 2020]
Moreover, previous research have showed benefits in bringing multiple users to the same VR environment, including easing the communication among teams [Benford et al.,
1996; Çöltekin et al., 2020; Du et al., 2018; Geszten et al., 2015; Ottosson, 2002;
Schroeder, 2006; Wolfartsberger et al., 2020]. Benford et al. [1996] suggested that computer-supported cooperative work could define altogether the context in which distributed work takes place. Similarly, Ottosson’s [2002] study concluded there are three main applications where VR contributes to improve the product development process:
simulation, skills training, and communication among distributed teams. On the other hand, Schroeder [2006] considered immersive technology as an end state in which synthetic and multiuser environments that are displayed to the senses cannot be developed further. Hence, argued multi-user immersive solutions will increase the social symmetry in asymmetric settings. In recent studies, researchers haveconcluded multi-user VR has achieved a state where it can enhance co-prescience [Choi et al., 2015; Çöltekin et al., 2020; Du et al., 2018; Guo et al., 2018; Wolfartsberger, 2019], even in asymmetric settings [Horst et al., 2020; Jansen et al., 2020; Lee, J. et al., 2020; Wolfartsberger et al., 2020; Xu et al., 2019]. Aforementioned studies argue the sense of co-presence was achieved by sharing a spatial context, in this case VR, and being able to see the others’
actions.
During the past couple of decades, researchers have analyzed VR as tool to enhance teamwork and to create the sense of co-presence. However, there is little research on multi-user applications that are not fully immersive to all users. In contrast to immersive VR, asymmetric VR solutions can be cheaper to implement while still providing a sense of co-presence [Li et al., 2003; Rand et al., 2005; Wolfartsberger, 2019; Wolfartsberger et al., 2020]. Hence, asymmetric VR applications might be an attainable and affordable option for companies. Moreover, asymmetric applications might expand the accessibility of VR to users that do not have readily access to immersive VR hardware [Gaoliang et al., 2010; Rand et al., 2005]. Due to the availability of hardware and software at KONE at the time of the study and in order to study virtual reality from a multi-platform perspective, in this study, I analyzed the effects of virtual reality in the product development process from an asymmetric perspective using HMD and desktop VR of the COVE-VR application. Half of the participants evaluated the COVE-VR application using HMDs while the second half used the desktop VR version of the same application.
2.5 Virtual reality in the product development process
In recent years, new production models that aim to interlink production data became more popular in the industry. Lasi et al. [2014] referred to this new production model as
the industry 4.0. Subsequently, several authors described industry 4.0 as a factory intertwined to centralized product data in real-time [Kovar et al., 2016; Navas et al., 2020;
Wen and Gheisari, 2020]. This new production model aims to shorten the product development periods, facilitating fast decision-making, and increasing resource efficiency and collaboration. In order to advance industry 4.0’s centralized data, several researchers have proposed mixed reality platforms to rapid data visualization, such as augmented and virtual reality [Kovar et al., 2016; Navas et al., 2020; Wen and Gheisari, 2020]. For instance, Kovar et al. [2016] parted from the industry 4.0 model to describe Global Virtual Models (GVM) as production cells with access to real-time automated operations. Furthermore, their study outlined the benefits of pre-production tasks using GVMs as a common base of data while VR applications would enable communication among stakeholders, visualization of the product’s status, simulations, predictions, and other information for effective collaboration. On the other hand, Navas, et al. [2020]
considered industry 4.0 as intelligent software solutions to provide virtual process simulations. According to the authors the development of maintenance methods could be integrated to the industry 4.0 model in what they described as maintenance 4.0. In their opinion, the maintenance 4.0 model would work as a centralized point of information, running supporting maintenance simulations in real-time. Thus, all the diagnostics, predictions and preventive maintenance work could be done and managed using digital models that would mimic the real products, also called digital twins. In addition, Navas, et al. [2020] suggested to ease the visualization of detailed real-time simulations that occur in the digital twin by using VR applications. Consequently, the rapid visualization would also enhance Lean and Agile ways of production in the industry.
Virtual reality VR solutions have not only been studied as a way to visualize process following the industry 4.0 model [Kovar et al., 2016; Navas et al., 2020; Wen and Gheisari, 2020], but also to solve a variety of problems in the product development cycle.
As such, researchers have designed and tested VR solutions aiming to evaluate the feasibility and benefits of VR in the product development cycle [Kovar et al., 2016; Navas et al., 2020; Ottosson, 2002; Wolfartsberger et al., 2020], improve the product design review process [Choi et al., 2015; Freitas et al., 2020; Wolfartsberger, 2019;
Wolfartsberger et al., 2020], ergonomic tasks [Di Gironimo et al., 2013; Dong et al., 2013;
Gomes de Sá and Zachmann, 1999; Guo et al., 2018], and collaborative work [Çöltekin et al., 2020; Du et al., 2018; Wolfartsberger et al., 2020].
In the early 2000’s Ottosson [2002] conducted a study to analyze the potential use cases of VR in the product development process to face the increasing needs of rapid testing, technical content creation, and peculiarities of each product’s life cycle (PLC). The author concluded VR could enhance the product development process in support of geographically distributed teams, training, and simulation tasks. In addition, other researchers have aimed to analyze the benefits of VR applications as a tool to collaboratively evaluate the product design during the design review [Choi et al., 2015;
Freitas et al., 2020; Wolfartsberger, 2019; Wolfartsberger et al., 2020]. Freitas et al.
[2020] studied the implementation of VR design review applications and the use of virtual prototypes in the design review stage. Their study concluded that virtual reality applications eased the identification of design errors, bolstered agile design corrections, and provided cost reductions compared to the use of traditional industry methods, such as physical prototypes. In addition, Wolfartsberger [2019] presented a solution to enhance visualization during the design review. Their studies concluded their solution increased visibility of issues in the 3D model than in a CAD software on a PC screen. Similarly, Choi et al. [2015] concluded that the use of VR applications improved the understanding of complex scenarios by providing a rapid and immersive visualization experience. On the other hand, Wolfartsberger et al. [2020] presented a solution to enhance the human- human interaction in VR during the design review. Their application supplemented VR users with multimodal feedback triggered by non-HMD users (See Figure 2.8). The argued VR improved the communication among stakeholders during complex collaborative design review sessions. Although [Choi et al., 2015; Freitas et al., 2020;
Wolfartsberger, 2019; Wolfartsberger et al., 2020] analyzed different use cases of VR in the produce development process, the studies disregarded to describe how integrating VR in the process could bring agility in the process. In addition, the current research on VR applications for product development process provides limited multicultural or cross- disciplinary use cases studies.
Figure 2.8: Design review using 3D CAD (left) and asymmetric VR (right) [Wolfartsberger, 2019]
In addition, several authors studied VR as a way to conduct simulation tasks by analyzing the product in its context of use [Di Gironimo et al., 2013; Dong et al., 2013; Gomes de Sá and Zachmann, 1999; Guo et al., 2018]. Some examples of product development process simulation tasks that can be performed in VR are maintenance procedures, working postures, product assembly and disassembly, product usability, among others.
Dong et al. [2013] presented a product maintainability and maintenance time prediction VR application that aimed to respond to the lack of physical prototypes in the early design phases (See Figure 2.9). The authors concluded maintainability and time prediction tasks could be conducted prior to creating a physical prototype using a VR-based application.
They referred to the use of virtual maintenance simulation applications to replace real- life maintenance tasks as virtual maintenance VM. Additional research has been conducted on multiuser simulation applications that aimed to test maintenance development processes in VR. Guo et al. [2018] tested a multiuser maintenance simulation application where one user conducted the simulation guided by a supervisor throughout the entire duration of the test. The simulation test showed that the participants could have more explicit and practical understanding of the maintainability needs before the product required maintenance work.
Furthermore, other authors [Di Gironimo et al., 2013; Gomes de Sá and Zachmann, 1999]
analyzed potential industry cases where VR applications could work as a tool in cross- disciplinary product development processes. For example, Gomes de Sá and Zachmann [1999] described a VR application to simulate maintenance tasks using virtual prototypes to document such tasks simultaneously. The researchers recorded the actions directly from the VR scene for later use in support of service manuals creation. Similarly, Di Gironimo et al. [2013] developed and tested an application to ease maintenance analysis and multimedia manual development. Their experiment consisted of users simulating ergonomic maintenance tasks in VR whilst the sessions were being recorded to support documentation creation. The media output from the VR session was later edited to compile the information in a comprehensive way. As such Di Gironimo et al. [2013] and Gomes de Sá and Zachmann [1999] described ways to use VR as a tool to manage the documentation and maintenance process concurrently. Although both studies analyzed VR solutions with different approaches to document maintenance tasks, both failed to describe how VR applications could be integrated as part of the maintenance and documentation process and consequently lacked to assess applicability in the industrial tasks.
Figure 2.9 Virtual maintenance setup proposed by Dong et al. [2013]
On the other hand, researchers have analyzed solutions to improve users’
communication in VR to expand the adoption of multiuser VR in the industry [Çöltekin et al., 2020; Du et al., 2018; Wolfartsberger et al., 2020]. For instance, Du et al. [2018]
proposed a cloud-based multiuser VR headset system called collaborative virtual reality CoVR that facilitates interpersonal project communication in an interactive VR environment. The CoVR research aims to enhance human-human interactions with simulated face-face conversations in the virtual world. Du et al. [2018] conducted a usability test and an experimental evaluation to assess the comparative advantages of CoVR against traditional methods and single-user VR applications. The results from the experiments showed that the system improves interpersonal interaction in the immersive virtual environment and enhances communication in construction projects. Likewise, the study showed users performed better in building inspection task than those who were using a single-person VR system. In addition, Wolfartsberger et al. [2020] conducted a study to analyze communication in a asymmetric setting [Horst et al., 2020; Lee, J. et al., 2020] in the product development process. Their study focused on collocated HMD users and non-HMD users in the product design review. This study concluded that multimodal input from non-HMD users to HMD users increased the symmetry in the interaction with the system.
Research on VR in the product development process has led to immersive VR applications that diminish production times, improve safety and enable complex product simulations [Choi et al., 2015; Freitas et al., 2020; Wolfartsberger, 2019; Wolfartsberger
et al., 2020]. However, research on adopting VR seamlessly along the internal process as a tool to enhance cross-disciplinary collaboration remains limited. Therefore, this study focused on evaluating the effects of integrating COVE-VR application as a mean to support department-to-department collaboration in the industry. Firstly, this study aimed to analyzed what and how COVE-VR could support maintenance and technical documentation processes at KONE. Secondly, this thesis presents the implication hypothesis of implementing VR in support of cross-disciplinary and geographically distributed teams. Lastly, this study aimed to analyze the effects in the process from Lean and Agile approaches, as COVE-VR is integrated to an internal process following these (See Chapter 3, The users and context).
3 The users and context
In this chapter I will describe the research I conducted to understand the users, tasks, and environment by continuously involving the users in the process. Firstly, I will provide an overview of the use environment (Section 3.1. Lean and agile context) by describing company’s culture. Secondly, I will describe the users and tasks (Section 3.2.Users, stakeholders and tasks) by analyzing the current ways of working in the maintenance methods development and technical documentation. Lastly, section virtual reality at KONE (Section 3.3.Virtual reality at KONE) will broaden the understanding of the environment of use. In the section 3.3, I will provide an analysis of the user’s perception of the VR based on the COVE-VR single-use evaluation conducted in October 2020.
Thereafter the information presented in this chapter aim to understand the users and context of use based on the single-user evaluation as well as activities proper of this study (See Figure 3.1).
Figure 3.1: HUMOR research activities. Chapter 3 of this thesis in purple.
3.1 Lean and Agile context
The goal of this study was to investigate how VR can become a tool to enhance internal processes. Therefore, in this study I introduced VR as an instrument to enhance industrial process in accordance to the industry principles. In the industry, the management philosophies work as guidelines to enhance improvement throughout the product development process. For instance, they help reducing the number of iterations, reducing waste, or even reducing the amount resources need to. For the purpose of this thesis, I will describe two of the main management philosophies at KONE: Lean and Agile.
Lean manufacturing refers to the production using less compared to mass production systems. It is a management philosophy that can be traced back to the Toyota Production System (TPS) among other industry practices. Similarly, to the original Toyota’s Lean
“seven wastes” focuses is on reduction to provide customer satisfaction [Wang, 2010].
According to the TPS the seven wastes are: overproduction, excess in inventory, waiting, transportation, unnecessary motion, over processing and quality defects. The first of the
“seven wastes” is overproduction. According to the Lean principles, overproduction refers to producing things before they are required. Secondly, according to Wang [2010]
the excess of inventory is a waste as it signals potential improvements in the operating performance. The waste of waiting transportation occurs when goods are not moving or being processed. Conversely, unnecessary motion leads to excessive movement of goods increases the susceptibility of damaging them thus creating waste through transportation.
In the same way, unnecessary movement of people involved in the process can lead to waste. Over processing refers to use more expensive resources or adding more features than necessary. Lastly, the quality defects waste describes the quality defects that might result in rework or scrap. The TPS practices established some of the core principles of this management philosophy, however the Lean principles were described in subsequent years. The term Lean appeared in the 80s when Krafcik [1988] coined the term to describe a management philosophy. Nevertheless, the concept of Lean as we know it today was born in 1996 and was described by Womack and Jones [1996]. According to Womack and Jones [1996] Lean consists of five principles: specify value by specific product, map the steps in the value stream, make the value steps occur in a tight sequence, establish pull from customers, and pursue perfection. Likewise, the TPS practices, Lean is founded on the concept of continuous and incremental improvement on processes while reducing waste. Womack and Jones [1996] included the seven waste types proposed by the TPS, however, add several other types of waste to make the model more comprehensive [Veech, 2017]. The new types of waste are waste of skills, faulty goods, under-utilizing capabilities, delegating tasks with inadequate training, working with no metrics, not leveraging the contribution of workers, and wasting computer resources.
On the other hand, Agile was popularized by the introduction of the Manifesto for Agile Software Development. Agile software development is used as an umbrella term for frameworks, including Scrum, that are based on the Agile manifesto. The authors of the manifesto chose the term agile as it bestows the idea of adaptiveness and response to change [Cohn, 2010]. The Agile practices aim to discover requirements and developing solutions through the collaborative effort of cross-functional development teams, product users and stakeholders [Martin, 2020]. The Agile manifesto based the practices on four values:
• Individuals and interactions over processes and tools,
• Working software over comprehensive documentation,
• Costumer collaboration over contract negotiation, and
• Responding to change over following a plan.
The first principle responds to have a team working effectively. Software development over comprehensive documentation refers to focusing on the end-product rather than on the documentation of the process. The third principle aims to work closely with users and stakeholders directly. Lastly, the fourth principle supports rapid decision making amid fast-moving tech projects. Since the introduction of the Agile principles, adopters of Agile practices claim the practices lead to gains in productivity with corresponding decrease in cost.
This research project was conducted at KONE Technology and Innovation (KTI) unit. KTI is driven by Lean and Agile principles that ensure solutions maximize the value solutions deliver to internal and external customers by optimizing the efficiency and the resources following a customer-centric culture and strategy. Hence, the technical documentation development process follows the same principles that regulate KTI as part of the product solution that KONE delivers. KTI’s Lean and Agile principles focus on five different areas. Detailed definition of such areas will not be discussed in this chapter as it is not within the objectives of this thesis. Nevertheless, in this study will provide an overview of the relevant management principles that influence development and improvement of internal processes, based on KONE’s guidelines:
• Lean startup. Testing and iterating solutions with customers and users in order to gather up rapid feedback to suit the solutions to the market needs via actionable metrics to maximize learning. As well as the definition and use of MVP (Minimum Viable Products) to build business hypothesis before building final solutions.
• Design principles for real-life. To take into consideration real contexts of use to understand matters that will go unnoticed otherwise.
• Lean service creation. Validating ideas early on, solving real customer problems, and developing ideas in an inter-disciplinary and transparent way for all parties involved, encouraging collaboration.
• Feedback loops. Making intentional changes to the process, measuring their impact and learning from the results by drafting hypothesis with real customers, validating and tracking using metrics.
• Kaizen. Continuously improving processes by identifying waste and its root causes, brainstorming solutions, creating future state models, planning improvement actions, and communicating the control plans. Thus, eliminating targeted waste in the process.
3.2 Users, stakeholders, and tasks
To investigate and analyze the status of the maintenance methods development and its corresponding technical documentation process, I conducted a series of six semi- structured interviews with industry experts throughout December 2020. The focus of the interviews was to analyze the cross-department collaboration and perception of VR;
hence, the questions addressed the current processes and stakeholders to identify recurring issues. To collect meaningful insights, the six participants represented maintenance method developer and the technical documentation subject matter experts.
In this study, the term subject matter expert refers to users with expertise on a specific field. The participants were KONE employees at two sites where the maintenance and documentation teams have presence: KONE Finland and KONE India. The participants included two maintenance method development engineers from Finland and one from India, as well as two technical documentation experts from Finland and a technical documentation manager from India. Similarly, I grouped the different roles of the KONE technical documentation team within the term technical communication specialist. The interviews were conducted using Microsoft Teams videocalls and lasted on average 50 minutes. A detailed list of respondents per site can be found in Table 3.1.
Number / Site KONE Finland KONE India
Interview 1 Maintenance method developer Interview 2 Technical communication specialist Interview 3 Technical communication specialist
Interview 4 Technical communication specialist
Interview 5 Maintenance method developer
Interview 6 Maintenance method developer
Table 3.1 User research semi-structured interviews participants list
The participants of the semi-structured interview described ten actors in the maintenance and documentation process (See Table 3.2 Users, stakeholders and tasks in the KONE process). In this study, I considered the members of the technical documentation, maintenance development team and design engineers as the end-users (icon in pink), and the rest as stakeholders in the process. Further clarification of actors per task are visible in Figure 3.1, The maintenance methods development and technical documentation process.
User type Role name Tasks
1. Technical writer
(Technical communication specialist) Creates technical information 2. Technical illustrator
(Technical communication specialist) Creates technical illustrations in support of a product’s document
3. Information designer (Technical communication specialist)
Oversees technical documentation process, in particular resources, time, review, and approval process
4. Maintenance method developer Owns maintenance method development 5. Design engineer Designs the product and is the product’s CAD
owner
6. Test engineer Supports test procedures in the elevators 7. Risk assessment moderator Procures accurate safety documentation 8. Risk assessment approver Decides when the method is safe to proceed
with the documentation
9. Component owner Oversees product/component integrity throughout the development process 10. Project manager Oversees the overall project
Table 3.2 Users, stakeholders and tasks in the KONE process
Therefore, I analyzed the interview data using a visual technique called sequence modelling. Sequence modelling is a method used to visually represent the overall workflow, steps, actors, and issues in a process [Holtzblatt and Beyer, 2017a; Holtzblatt and Beyer, 2017c; Rex and Pardha, 2019]. The aim of the sequence models is to ease visualization of the steps in the process where the users struggle the most. For each interview I built a sequence model, the sum of all the sequence models was latter processed into a consolidated sequence model [Holtzblatt and Beyer, 2017a; Holtzblatt
Icon Definition End-users Stakeholders
and Beyer, 2017c; Rex and Pardha, 2019]. The aim of the consolidated model was to understand the status of the process to then adapt the COVE-VR evaluation to suit those needs. The comprehensive consolidated model can be found in Figure 3.1 The maintenance methods development and technical documentation process. In the figure, in left column there are descriptions of the issues per pain point, the pain points are illustrated as red lightings that break the flow in the process. Secondly the middle column illustrates the milestones. Lastly the right column indicates actors by milestone. The process steps include:
1) Initial inquiry loop
a) Project phase: initial notification of a new project
b) Project kick-off: the scope is reviewed to define tasks and the project is setup c) Outline creation: draft with written and visual content. It includes a theoretical
risk assessment.
d) Outline delivery: outline draft is shared
e) Cooperative expert exploration: ease understanding of the product using the documentation outline and any complementary information needed
f) Questions: conclusion of the initial inquiry loop to continue with the documentation draft
2) Documentation draft: collaborative creation of first documentation draft 3) Documentation update: update the instructions based on documentation draft 4) Documentation review: review of the documentation draft
5) Maintenance method assessment: the method is tested at the equipment location (this step relies and depends on the availability of a physical prototype)
6) Risk assessment: maintenance method is reviewed to ensure it is risk-free
7) Risk assessment review: review of the risk assessment process, this leads to request for further changes or approval of the risk assessment process
8) Documentation checking and approval: reviewers approve the content 9) Publishing: documentation is published in the internal platforms 10) Training: documentation is used to train local maintenance teams
The pain points are signaled with a red lightning and they signify there is a breaking point in the process. Figure 3.2 indicates that one of the most critical and challenging milestones in the developing process is the ability to have prompt and easy access to the product’s 3D information. Further analysis on the maintenance methods development and technical documentation process can be found in the Chapter 4.
Figure 3.2: The maintenance method development and technical documentation process
