Indoor positioning in
manufacturing environments

An investigation of system usage for
Industry 4.0 applications

POVEL CROONA
ANDREAS NIKLASSON

Department of Communication and Learning in Science
Chalmers University of Technology

Gothenburg, Sweden 2020





Indoor positioning in manufacturing environmentst
An investigation of system usage for Industry 4.0 applications

POVEL CROONA
ANDREAS NIKLASSON

© POVEL CROONA, ANDREAS NIKLASSON, 2020

Master’s Thesis 2020
Department of Communication and Learning in Science
Chalmers University of Technology
SE-412 96 Gothenburg
Telephone +46 31 772 1000

iii



Abstract
This study investigates how radio-based indoor positioning systems should be used
to generate user value. To do this, problems that indoor-positioning can solve and
potential system users must be found. The general approach is the design method design
thinking. This means that iterations are used to continuously increase understanding of
the user and the problems. To gather qualitative data 27 interviews (14 unstructured
& 13 semi-structured) were conducted. This data was then analyzed using the Affinity
Interralationship method do identify problems that indoor positioning could solve in
the manufacturing industry. Quantitative data gathered from a positioning system at
a manufacturing site was used for investigating possible system usages. Solution ideas
were evaluated with the identified main user and a focus group. Our analysis shows
that a radio-based indoor-positioning system can be used to solve problems with lack
of data and low-qualitative data. It also identifies the production engineer as the main-
user of such a system. User value is generated to this user by using the system to
increase safety and productivity; Supply data for manufacturing KPI calculations; and
enable manufacturing process analysis. The study concludes that indoor-positioning
has many relevant areas of usage in manufacturing environments. However, solution
ideas with the biggest impact are today limited to monitoring applications. We would,
therefore, suggest further research regarding integration between systems used in the
manufacturing environment and radio-based indoor positioning systems. This would
enable controlling activities as well.

Keywords: Indoor positioning, Design thinking, KPI, Smart factories, Industry 4.0,
Manufacturing

iv





List of Abbreviation

AGV |Automated guided vehicle
GDPR | General data protection regulation
RTLS | Real-time location system
UX | User experience
ROI | Return on investment
GUI | Graphical user interface
TCP | Transmition control protocol - Protocol for internet communication
KPI | Key performance indicator

vi



Acknowledgements
First of all we would like to express our sincere gratitude to our supervisors at Eric-
sson, Joakim Pontén and Kjell Berglin. Without your support, knowledge, and
continuous encouragement this master thesis would not have been possible to realize.

At Ericsson our deepest thanks are also directed to all the people on the dedicated
network team. Your skills and knowledge were only trumped by your helpfulness and
were invaluable resources during our first months.

We would also like to thank Henrik Westberg at SKF, for going the extra mile and
helping us both with academic and practical endeavours.

Our appreciation is also directed to the staff at SKF’s D-factory and especially Oliver
Carlen for helping us evaluate and improve our ideas.

Jan Wickenberg, our supervisor at Chalmers, have also been an important person
in the completion of this Master Thesis. Thank you for all feedback that helped us
improve our work.

Finally, all the people that participated in interviews, evaluation sessions, focus group
or in any other way aided us during this research deserves our greatest thanks.

Thank you!

Povel Croona & Andreas Niklasson, Gothenburg, May 2020

vii





Contents

1 Introduction 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Research Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2 Theory 5
2.1 The indoor positioning problem . . . . . . . . . . . . . . . . . . . . . . 5
2.2 The Manufacturing Industry and how it is measured . . . . . . . . . . 5

2.2.1 Industrial evolution . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2.2 Lean Prodcution . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2.3 Gemba . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2.4 KPI - Key Performance Indicators in the manufacturing industry 7

2.3 Organizational structures according to Mintzberg . . . . . . . . . . . . 8
2.4 Industry 4.0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.5 Design methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2.5.1 Design Thinking . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.5.2 Affinity Interrelationship Method -AIM . . . . . . . . . . . . . 12
2.5.3 Braindrawing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.6 Persona . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.7 Knowledge gaining and transfer . . . . . . . . . . . . . . . . . . . . . . 15

3 Method 17
3.1 Research design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

3.1.1 Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.2 Systematic combination . . . . . . . . . . . . . . . . . . . . . . 18

3.2 Litterature study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3 Data collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

3.3.1 Quantitative data . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3.2 Qualitative data . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

3.4 Identification of valuable system usage . . . . . . . . . . . . . . . . . . 23
3.4.1 Understand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

ix



3.4.2 Explore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.4.3 Evaluate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

4 Results 32
4.1 Identified problems that indoor positioning can solve . . . . . . . . . . 32

4.1.1 Missing data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.1.2 Misused data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

4.2 Identified users of an indoor positioning system . . . . . . . . . . . . . 33
4.3 Identified value adding system usages for main users . . . . . . . . . . . 35

4.3.1 Increase safety and productivity at the manufacturing site . . . 35
4.3.2 Supply data for manufacturing KPI calculations . . . . . . . . . 36
4.3.3 Manufacturing process analysis . . . . . . . . . . . . . . . . . . 38

5 Discussion 39
5.1 Result interpretation and implications . . . . . . . . . . . . . . . . . . 39
5.2 Ethical aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

6 Conclusion 43

References 44

A Personas i

B Use-case iii

x



1
Introduction

This introduction chapter aims to give an understanding of why this research is con-
ducted. Initially a background is presented to problematize the lack of indoor position-
ing in the manufacturing industry. This is followed by a description of the purpose and
limitations of the research. Finally, the research questions that this thesis will answer
are presented.

1.1 Background
Manufacturing industries are constantly trying to optimize production processes by
analyzing data from the production (Amrina, Firdaus, et al., 2018). This is often done
by identifying and reducing activities that do not add value to the product. (Liker
& Meier, 2006). Operations that occur in the production cells can be measured very
accurately. But mobile activities over greater distances, such transportation of material
or human activities on assembly lines, are harder to analyze. In outdoor environments
analysis of moving objects can be done using GPS. But because devices and satellites
requires free sight, positioning using GPS is not possible in indoor environments.

There are many different technologies making indoor positioning possible today. But
no complete solution have been created to function as an indoor counterpart to GPS.
Samama (2019) explains that this is not a technical problem but instead derives from
the inability to meet required customer constraints, such as ease of use, cost, and
performance. The inability to satisfy these parameters simultaneously is described as
the indoor positioning problem and have caused many positional projects to deliver
unsatisfying result. Because of this, investors are becoming increasingly reluctant to
provide funding for indoor positioning projects.

The solution of the indoor position problem must be based on the actual needs of the
industry (Samama, 2019). This requires an understanding of how industries operate
when implementing new technology, as well as the potential user of those systems. In-
dustries measures its operation using KPI’s (Key Performance Indicators) and relates
to parameters such as production output and overall equipment effectiveness (Amrina
et al., 2018). Because of this, it is important to show how investments in new systems

1



or technology can improve KPIs, especially when it comes to larger investments. Un-
derstanding this is key to present improvement suggestions that will be approved by
upper management. Commonly these suggestions are backed up with a business case
that explains how an investment will benefit the organization.

Successful implementation, however, does not only rely on correct business priorities.
Loranger (2014) describes that these priorities often lack the reality of the user. With
close user involvement, products are more likely to meet the user’s expectations and
requirements. This also increases the workforce’s acceptability of new systems, which
makes changes and new technology easier to establish (Benyon, 2014). Understanding
the user is because of this essential, especially when a product is a result of technological
advancement rather than market demand.

This study aims to find problems in the manufacturing industry that can be solved
with indoor positioning. Identified problems will be analyzed to find solutions that
generate value for system users in their work. The research is made possible by a pre-
views venture project between Ericsson - Private Networks and multiple other industrial
stakeholders. This project resulted in an indoor positioning system that can position
material handling equipment and will be used to gather position data. The next section
will give a brief overview of the system’s functionality and purpose.

Ericsson Private Networks

Figure 1.1: Ericsson Private Network setup

Ericsson offers cellular network Service Solutions, aiming to be a leader in 5G deploy-
ment. Ericsson radio technology makes it possible to set-up and manages small private
cellular networks, see figure 1.1. This means installing radio antennas on a closed-off
area. Cellular devices can then connect to this network. One of Ericsson’s goals is to
offer manufacturing industries this network solution. Ericsson is currently testing such
a system at the SKF manufacturing plant in Gothenburg.

2



Positioning with cellular networks

Figure 1.2 shows a setup at an industrial facility. The cellular network is distributed
via radio-dots (antennas) mounted in the facility. When several radio-dots covers an
area, you can use the response time from a device to each radio-dot to position the
device.

Figure 1.2: Overview of system setup with radio-based technology

1.2 Purpose
The purpose of this study is to investigate usages of radio-based indoor positioning
systems for Ericsson. The potential benefits of position data generated by the test
system installed at SKF will also be evaluated. The result of this evaluation will
support usage prioritizing. It will also be used to investigate the return of investment
(ROI) for radio-based indoor positioning systems. To do this manufacturing problems
and system users must be identified.

1.3 Limitations
This thesis is limited to investigating problems, system users and system usages in
manufacturing environments from January to June 2020. Because of this, variation on
an annual basis will not be shown in the qualitative or quantitative data. Positioning

3



data is gathered from a test system that is monitoring material handling equipment
at a manufacturing site. This means that no comparison between similar positioning
systems will be made. Finally, there will be no implementations of the findings. The
ending phase of the design process will instead focus on the evaluation of the content
and its value for the users.

1.4 Research Questions
To enable the best understanding of how an indoor positioning should be used in a man-
ufacturing environment problems must be identified. This will be done by answering
the question:

• What problems can radio based positioning systems solve in the manufacturing
industry?

Identified problems must then be linked to potential users of the indoor positioning
system. In particular a main user should be identified to enable the development of
problem solutions that are likely to be value adding for the manufacturing industry.
The second research question is, therefore, phrased as follows:

• Who are the main users of indoor positioning data in the manufacturing industry?

When problems and main users have been identified the best system usages will be
investigated. The result should be able to answer the third research question:

• For what should indoor positioning systems be used?

4



2
Theory

The purpose of this chapter is to give the reader a short introduction to how man-
ufacturing industries measures and improves production processes. This is done by
explaining lean production philosophies and manufacturing KPIs. This is followed by
a brief explanation of Industry 4.0. The chapter also describes a number of method-
ologies used for user centered design. These methods will later be used to analyze the
processes and system usages in manufacturing environments.

2.1 The indoor positioning problem
Indoor positioning is a technology that makes it possible to find the location of an
object in indoor environments. In the industrial section, indoor positioning could be
used to improve safety systems, such as collision avoidance; or automated monitoring
and quality control (Mautz, 2012). The main type of location that is discussed in this
thesis is physical locations and is expressed in the form of 2-D coordinates. Location
can also be expressed in symbolically with names like office or warehouse (Gu, Lo, &
Niemegeers, 2009). Samama (2019) discusses the potential of positioning in industrial
indoor environments and says that even though the economic outlook has been described
as exceptional there still does not exist an indoor positioning standard. The reason for
this is that the needs of the market are poorly formalized. Mautz (2012) agrees that
user requirements must be redefined to enable the development of indoor positioning
applications, but also emphasize in benchmarking of implemented systems.

2.2 The Manufacturing Industry and how it is mea-
sured

The manufacturing industry is sometimes described as the corner stone of the modern
economy. Continuous improvements of industrial processes have led to increasingly
better living standards. This chapter aims to give a brief overview of the manufac-
turing industry to better understand the challenges of creating change that improves
production processes.

5



2.2.1 Industrial evolution
Even though the rapid growth of the industry during the mid 19th centuary often is
refered to as the industrial revolution, many historians argues that the industrial ex-
pansion instead should be viewed as evolutionary (Basalla, 1988). Small improvements
of the coalmine steam engines made it efficient enough to be used in other contexts, like
the textile industry. This increased both the profit and complexity at the factory sites
which called for a better understanding of the working processes. By breaking down
longer working processes into small pieces and limiting variation, the work of a pro-
fessional could be assigned to less experienced worker. This enabled lower production
costs that made more products available to the common people e.g. the standardized
T-Ford.

2.2.2 Lean Prodcution
Lean production or simply Lean is a production philosophy that grew from the American
research program International Motor Vehicle Programme, conducted during 1985-1990.
The research aimed to compare the effectiveness of different automotive manufacturing
industries. The findings suggested that Toyota Production Systems (TPS), developed in
Japan after World War II, had the most effective production methods (Womack, Jones,
& Roos, 1990). Because of this many of the Lean production principles can be traced
to the structure of TPS during the 80’s (Börnfelt, 2009). Today Lean is widely used
in industries that focus on increasing quality, shorten lead time and decrease variation
in production (Liker & Meier, 2006). This is done by applying Lean principles and
eliminating waste in the production.

Lean principles

The foundation of Lean stems from applying its methods to create maximal value for the
customer and reach target organizational goals. Methodologies from Lean production
aims to, as the name suggests, slim down operations leaving only value-adding activates
to increase effectiveness (Börnfelt, 2009). Other fundamental concepts within Lean are
flow and resource efficiency. Flow efficiency is defined as the sum of value-adding
activities in relation to the throughput time. Resource efficiency instead focuses on
the maximal usage of assets and people (Modig & Åhlström, 2012). Manufacturing
industries traditionally have opted for resource instead of flow efficiency, even though
Lean generally is regarded as a method the prioritize flow (ibid.).

Waste within Lean

Lean production defines seven different types of waste that all can be placed into the
categories Muda, Muri and Mura. Liker and Choi (2004) describes that eliminating
waste is key to reduce the number of activities that not is value-adding. Muda represents
waste that stems from activities that does not increase value for the customer and is the

6



main focus within Lean production. Muri instead focuses on the waste that stems from
overloading employees or assets, resulting in injuries and mechanical wear. Finally,
Mura represents waste from fluctuation which can create uneven material flows and
overproduction (Womack et al., 1990; Liker & Choi, 2004). The seven types of waste
consists of transport, inventory, motion, waiting, overproduction, defects and unused
creativity.

2.2.3 Gemba
Gemba, a Japanese word for "the actual place" meaning the place where value is created,
for example, in a manufacturing process (Richardson & Richardson, 2017). Going to
Gemba, the place where value is created, is essential in many Lean tools. Toyota, the
source of several lean principals, has a bias for low-tech visual management (ibid.).
The company does not want its engineers and managers to sit at the office and monitor
the production via updated visuals on computer screens (Richardson & Richardson,
2017). Instead, they prefer the low-tech visualizations based on data provided by people
working in the process. More advanced monitoring systems can be implemented after
low-tech visualizations have proven to be sufficient. Otherwise, one is likely to be
trapped in a digital solution that does not fulfill the needs and is not able to change
due to investments already have been made.

2.2.4 KPI - Key Performance Indicators in the manufacturing
industry

The manufacturing industry measures different Key Performance Indicators to monitor
the production processes. These indicators rely on data provided by the factory systems
and employees and aims to describe the process current and historical state. By looking
at this data, production personnel tries to minimize the production losses and maximize
the utilization (Richardson & Richardson, 2017).

Stakeholders in the manufacturing industry need to prioritize which KPIs to monitor
(Amrina et al., 2018). There are variations in how many, and which, KPIs a manufac-
turing site chooses to monitor. Some manufacturing KPIs are not possible to calculate
with positioning data. The KPIs listed in table 2.1 are only manufacturing KPIs which
can be calculated with positioning data. The KPIs contains both human aspects as
well as technical matters.

7



Table 2.1: Manufacturing KPIs possible to calculate with positioning data

Acronym Name Formula of calculation

- Loading Time total available time - planned downtime

OEE Overall Equipment Effectiveness Availability rate ∗ Performance rate ∗ FPY

AR Availability Rate Operating time
Loading time

∗ 100%
PR Performance Rate P arts P roduced ∗ Ideal Cycle T ime

Operating time

FTY First Time Yield Σ Units Leaving the P rocess
Σ Units put into the P rocess

FPY First Pass Yield FTY 1 ∗ FTY 2 ∗ ... FTY n
VA Value adding time V alue adding operations

All operations
∗ 100%

DT Down time T otal Downtime
Loading T ime

- Idle non productive time
Loading T ime

ELR Employee Late Ratio Σ Employee Late
Σ T otal Employee

- On Time Delivery No. of products delivered on time
No. of total produced products

∗ 100%

2.3 Organizational structures according to Mintzberg
The organizational structure can be divided into five parts (Mintzberg, 1979). Depend-
ing on the organization, these parts can vary in size and scoop. The five parts consists
of Operative core, Middle management, Strategic Apex, Support staff, and Techno struc-
ture. Figure 2.1 shows an overview of these parts.

The foundation of an organizations is the operative core. People working in the opera-
tive core are responsible for the actual work that creates the product or services. This
also includes direct support, such as planning of production. Mintzberg (1979) means
that there exist a limited range of control that a manager is able to handle. Because of
this, larger organizations needs a middle management to oversee day-to-day activities
in the operative core. The middle management have also responsibilities to manage
external relationships and branches of the company. At the top of the organizational
structure the strategic apex resides. The strategic apex is responsible for setting up
and reaching organizational goals. This does not only include economical goals but also
goals regarding how the organization is perceived by media and customers.

Besides the core structure that has been presented, Mintzberg (1979) also identifies the
technostructure and support staff as key parts of organizations. The technocstrucure
consists of analysts, such as engineers, accountants and researchers. These roles are
not responsible for operational tasks, instead they evaluate the processes and try to
optimize them by using tools like work studies. The last key part is the support staff

8



Figure 2.1: The organizational key parts

that provide indirect services e.g. salary administrators and cleaning services.

2.4 Industry 4.0
In order to recapture manufacturing to Germany the German government formed a
strategy which is today is refereed to as Industry 4.0. The strategy rapidly spread to
other European countries, as well as US and Japan (Reinhard, 2016)

Herman, Pentek (2016) describes the following design principals of Industry 4.0:

• Interconnection is about using the possibilities of connected devices. Once
products, machinery, and other devices are connected to a network, ways of wire-
less data communication between them are possible. For the people controlling
and working with these systems the Interconnection blaze a trail for increased
efficiency and effectiveness.

• Information transparency. A fusion of the physical world and the virtual
world makes room for new ways of controlling processes. Drawings, electronic
documents, and simulation models are examples of the virtual world (Herman,
Pentek, 2016). An example of information from the physical world would be the
position of a tool. Sensor data from the real world must be transmitted to central
databases, accessible for all connected devices (Herman, Pentek, 2016). Also,
analyses of collected data must be accessible for both humans and devices in the
system.

• Decentralized decisions. A combination of interconnection and information
transparency enables decentralized decision making. Devices in an industry 4.0

9



manufacturing site will be as autonomous as possible, only when conflicts between
devices occur the decision should be escalated to a higher level (Herman, Pentek,
2016).

• Technical assistance. The human’s role in an industry 4.0 environment will be
strategic decision-makers. Decisions close in time are taken by the devices. This
puts requirements on visualization techniques (Herman, Pentek, 2016). Screens
and wearables will be more important since the collaboration between the humans
and the factory will be increased.

2.5 Design methods
User experience, UX, is the totality of a users experiences from a product (software,
hardware etc). According to ISO 9241-210 UX is defined as "a person’s perceptions and
responses that result from the use or anticipated use of a product, system or service".
The UX field consists of studying, designing and evaluating user experience. UX meth-
ods can be used as tools for evaluating existing systems, but also for designing new
ones. ISO 9241-210 clusters the following areas as the roots for user centered design:

• Positioning the user as a central concern in the design process

• Identifying the aspects of the design that are important to the target user group

• Developing the design iteratively and inviting users’ participation

• Collecting evidence of user-specific factors to assess a design

Poorly designed software costs millions of dollars (Hartson & Pyla, 2012) because the
product is difficult to use. This fact enhances the economic drivers for why good UX
is preferable. This chapter will describe the theory behind design thinking as well as
giving an introduction to methods that often is used in the designing process.

10



2.5.1 Design Thinking

Figure 2.2: Design thinking process by Nilesen & Norman Group (Gibbons, 2016)

In order to achieve user satisfaction, for each and every user or sub-user of the system,
the design thinking approach puts focus on the user perspective of the product. By
using this method, created by Don Norman, the chance for a successful delivery that
meets user expectation increases.

Design thinking means to intentionally focus the design around the concerns, interests,
and values of the users (Denning, 2013, s.31). The steps in the iterative design thinking
process (figure 2.1) starts with an understanding section, which contains a empathize
and a design phase. The empathize phase is about what the user says, feels, does
and talks about the product. You must know what motivates or discourages users to
be able to empathize with the user an in that way see things from their perspective.
Observations and unstructured interviews are suitable data collection methods for this
type of information gathering. While empathizing with the users, the developing team
should pay attention to what the user:

• Says What is actually said during the interview? Try to direct quote the inter-
viewee.

• Feels How does the user feel about the experience? What worries the user?

• Does What the user does during the observation/interview?

• Thinks What is on the users mind during the interview? What matters to the
user?

11



From a deeper knowledge of the user, opportunities for innovations is made possible.
The gathered data is used to find the user needs and problems. This process is in design
thinking referred to as define and together with emphasize it builds an understanding
of both the user and the problems. Curedale (2019) also means that during defining it
is important to identify the stakeholders.

Once you gain knowledge about the users you can enter the explore phase. The goal
is to generate ideas that solves the problems based on the user. This often requires
iterating back to previous phases to reinforce user knowledge and problem definitions
(Gibbons, 2016). There are many tools available for ideation, such as brain drawing
or brain storming. By aiming at quantity and range the target is to capture a wide
variation of ideas (Wikberg Nilsson, 2015). Based on the ideas, still in the explore
phase, you now create a prototype. The prototype shall, based on your ideas, solve
the defined problem. If you feel that you lack information in this phase you must go
back to the user and problems to expand your understanding of them.

Finally, materialize the prototype and expose it to the users. Test it, and use the
feedback to improve the prototype, but also redefine the problem and if necessary go
back and improve your prototype. When the prototype tests yield the desired result
the last and most important step start - Implementation. This requires integration in
current system or work processes and is often time consuming. Denning (2013) means
that the there is a difference between design and innovation. An innovation is concerned
with getting a community of people to adopt the concept rather than the design process
behind it. Often this constitutes 90% of the work, and the actually design the other
10%.

2.5.2 Affinity Interrelationship Method -AIM
Affinity Interrelationship Method, AIM, is a 10 step method for analysing qualitative
data sets (Alänge, 2009). It is a systematic approach to problem-solving and has been
used at companies and universities for over 30 years (Alänge, 2009). Yet the AIM
process is flexible. Depending on the specific purpose of the analysis the setup, can be
changed to fit the problem being solved. Group size, problem scope, and the amount
of data being analyzed are such circumstances that call for modification of the original
process setup. The AIM requires:

• Post-its

• A White-board

• Pencils

• An appointed leader

The AIM is performed in 10 chronological steps:

1. Formulate the question: The study question should not be to narrow neither

12



to broad. Therefore, the question must fit the research projects scope and have a
suitable detail level to support the project progress.

2. Warm-up: Each participant gets time to explain his/ her take on the formulated
question. Time available for the warm-up is 5 min, so in a group of 5 participants
every participant gets 1 min each. Criticizing is not allowed during warm-up.

3. Collect data: The target is to write 19-24 post-its in total.

• The answers should be "full sentence answers"

• Facts instead of opinions, use the interview data to formulate the sentence

• Ladder of abstraction - choose formulations which is actually said/done by
the interviees

• Use multi valued information

• Write a full sentence

4. Clarify the Meaning: All the answers on post-its must be clarified if needed.
Each and every post-it is clarified by the team, in order to reach consensus about
the meaning. The team is allowed to ask questions to the author, but it is not
allowed to argue with the author of the post-it weather the statement is correct
or not. The clarification of an answer will be written down on the original post-it.
After the clarification process the post-it is moved to a separate part of the white
board.

5. Grouping - affinity steps: The post-it’s are now corrected. The team must
now group post-it’s with a similar meaning closer together. This task is based
on grouping by searching for "affinities" between the post-its. The grouping is
preformed during silence, and the members are allowed to split an rearrange
groups. The grouping in this step is a consensus driven process, which ends when
there is a shared view of the grouping.

6. Higher level grouping:

• First level titles. The groups are summarized and the team put headings
on each group. The headings explains the meaning of the group. Singular
post-its can be leaved without headings.

• Second level grouping. Continue grouping the headings and the singular
post-its. This is the second level grouping. Some post-its, with no heading,
also known as "lone wolfs", are accepted.

• Third level grouping. After the second grouping, the team must put labels
on the new groups. By applying the same procedure as in the second level
grouping, new 3d level groups are constructed at the board.

13



Show connection interrelationships. In order to show if there is a connection
between the groups the team points arrows between such groups. The aim is to
go through the possible dependencies between all groups.

7. Final Layout When all the interrelations between the groups were pointed out
the final AIM analysis was concluded. The data has now been analyzed and
clarified and shows the answer to the question stated in the beginning of the
process. It is important to find:

• Outlines in first level groups

• Outlines in 2nd/3rd level groups

• and to draw final arrows

8. Evaluation With all the concluded answers the group ranked the answers de-
pending on the impact on the original question. The most important issue which
influenced the question analyzed was pointed out by each group member. Where
there was a different between the individual rankings done by the group members
voting was applied. The finalized rankings were stated on the AIM board when
finished.

9. Concluding The AIM session ended after the main problems were put in a
full sentence where the importance from the rankings decides in which order the
problems being listed.

2.5.3 Braindrawing
Braindrawing is a creative ideation technique that utilizes sketching as a way for the
participants to express their ideas (Wikberg Nilsson, 2015). There are four roles to
follow when braindrawing:

• Don’t criticize, neither yourself or others.

• Aim for wild and crazy ideas.

• Combine ideas and seek improvements.

• Quantity before quality.

By using sketches instead of talking and writing when expressing the ideas the partici-
pants might find new perspectives. A collaborative approach will contribute to common
ownership of the ideas, which prevents a strong personal connection to specific ideas
(Wikberg Nilsson, 2015). Brandrawing is executed as follows:

1. Place papers and pencils on a table

2. Define the theme

3. Prepare the team and explain the theme

14



4. Sketch ideas for 5 minutes

5. Rotate the sketches within the team members

6. Continue drawing on the newly provided sketch

7. Stop when all participants have sketched on all papers

After the idea sketching round the team discuss and categorize the ideas. The most
important findings and ideas will be noted and presented as the result of the brain-
drawing.

2.6 Persona
When developing solutions for users, it is suitable to have a specific, fictive but yet
realistic, user to elaborate around. The persona will act as a real user for the develop-
ment team, and the team can use the persona when describing suggested solutions in
order to enhance the benefits of, or reject ideas (Hartson & Pyla, 2012). The number
of personas must vary depending on the size of the user group (Hartson & Pyla, 2012).
Since it is an bottom up process the exact number of personas is not possible to tell
until the creation process is executed. In figure 3.6 the constructed persona template
is shown.

2.7 Knowledge gaining and transfer
Double-loop learning is the meta-process of changing the main process based on feed-
back (Argyris, 2002). It is about questioning why you solved a problem in a specific
way. To explain double-loop learning you must first be familiar with single-loop learn-
ing. Single-loop learning is where the learner performs an action based on an action
strategy and then corrects its action strategy based on the consequences.

Double-loop learning occurs when the learner reflects over why he/she picks the specific
action strategy. The learner reflects over "why we do what we do", compared to single-
loop learning, where the learner corrects "what we do".

Knowledge gained during research projects is essentisal for the result. The majority
of organizations have some sort of knowledge management system (Werr, 1998). The
gained knowledge will be accessible for the organization first when the learner has shared
with others. There are two established methods for an organization to treat knowledge,
either knowledge is regarded as something that can be objectified and organized or
its regarded as something that is socially embedded (Werr, 1998). Depending on how
knowledge is treated in the organization one can construct different knowledge sharing
methods.

15



Figure 2.3: Visualization of the stages in single and double loop learning

16



3
Method

This chapter will describe the methodology that was followed during this Master The-
sis. Initially the overall research design will be presented, followed by descriptions of
the literature study and the collection of quantitative and qualitative data. This also
includes a short summary of methods used for both data gathering and analysis. The
remaining part of the methods describes the design thinking process which focus on
identifying user and what users value about a positioning system.

3.1 Research design
When writing a scientific report, the choice of research design will greatly impact the
study’s results. Therefore, it is important to evaluate and choose a method approach
suitable for the scope of the study. The first question that needs to be addressed is what
type of data that is most likely to answer the research question. In this case the problem
revolves around the supposed needs of positioning in the manufacturing industry. Since
the study will be conducted at only a few factory sites, the sample size will be much
too small to make any quantitative conclusions. The positioning data that is available
is however extensive enough to enable analysis of the factory processes. Because of this
quantitative data will still influence the end result, even though the study is considered
qualitative.

The report takes a bottom-up approach in its attempts to answer the research questions.
This goes very well together with the Design Thinking methodology that emphasize
in understanding the actual users. Defining problems from this perspective gives a
better chance of solutions that aligns with current needs and increases success rate of
implementations. The iterative process described in Gibbons (2016) have because of
this served as the main influencer of the research design.

3.1.1 Case Study
Case studies are an approach that through in-depth examinations of a particular case
can bring understanding regarding a complex issue (Lichtman, 2014; Dooley, 2002).
Most commonly it is used in qualitative research projects but has found uses in the

17



quantitative field as well. The purpose of case studies is described by Dooley (2002) as
a way of answering questions of how and why. Clardy (1997) expands this further by
including questions of who, what, where and when targeted at a specific situation.

When conducting case studies of groups or organizations one of the main goals is to find
key informants giving access to the needed knowledge. In research these individuals
are also named Gatekeepers because of their ability to give access to information that
otherwise would be unavailable (Given, 2008). This is also the biggest weakness with
case studies, since the research results heavily depends on finding theses informants
(Lichtman, 2014). Case studies are also criticised by Yin (2014) that claims that they
are unable to tell something about general matters. Dubois and Gadde (2002) replies
by stating that this should not be the aim of a case study. Instead case studies should
be used to connect phenomenons to their context. This can be done by using what
Dubois and Gadde (2002) refers to as systematic combination.

3.1.2 Systematic combination
Dubois and Gadde (2002) describes systematic combination as a nonlinear process that
matches reality with theory by evolving theoretical framework, empirical fieldwork, and
case analysis simultaneously. This process aims to strengthen case studies by increasing
the reliance on theory. By doing this the data can provide better support for theory
building or expansion. Systematic combination bears resemblance to the abductive
research approach, which aims to investigate the reletionship of everyday language and
concepts (Kirkeby, 1990). The main focus is to allow for the study to make a "best
prediction" by acknowledging that observations always will be incomplete. This makes
it different from the inductive reasoning that uses specific observations to draw general
conclusions (Dudovskiy, 2016). Abduction and induction also differs in its theoretical
objective. The inductive approach is described as a method for theory generation,
whereas systematic combining builds on the refinement of existing theories (Dubois &
Gadde, 2002, p.559).

Theory building by case studies generally requires an overlap of data analysis and
data collection. The juxtaposition between these two activities can create creative
insights that force the reframing of case perceptions (Eisenhardt, 1989). Systematic
combinations uses these learnings to reconsider and change the framework. This allows
for a more stringent relationship between framework and research results, making it
easier for the reader to follow the process (Dubois & Gadde, 2002). The learnings taking
place in the interplay between search discovery are, however, interesting at a meta-level.
Dubois and Gadde (2002) means that understanding the researchers learning process
would benefit learning in the research society. In this report the learnings will, therefore,
be discussed as a parallel result to the research answers.

18



3.2 Litterature study
As a first step in finding an answer to the given research question, a literature study
were conducted. The aim was to get a better understanding of the field both regarding
the technical aspects but also look in to what indoor positioning system usages that
already existed. To limit the amount of search results different key words were used
and sources that were considered relevant were stored in an excel sheet. Sources were
sorted by title and theme followed by a short explanation and a ranking of its perceived
usefulness. The reason for this was to make it easier to review the sources at later stages
of the writing process. Example of key words were RFID, UWB, Visualization and User
Interfaces. The databases used for conducting the searches was mainly Scopus because
of the Field-Weighted Citation Impact parameter linked to each source. Values higher
than 1 suggests that the report or article have been citations than similar sources.
The approach limited the searches and increased the over-all quality of the consumed
literature.

3.3 Data collection
This section describes the different data collections that will be used in this research.
The qualitative data will mainly consist of interviews. There will also be quantitative
data collection, that will be used to enable visualization of the positioning system.

3.3.1 Quantitative data
Positioning data from the test site will be gathered over via a API communication with
the system server (RTLS). The data contains information of:

• Position (X,Y,Z)

• Device ID

• Time

• Confidence interval

Positioning data was gathered from the test site using a built-in API that could commu-
nicate with the positioning system server. The data was received in a continuous flow
of text strings containing the position (x, y, z-coordinates) linked to a specific Tag-ID
and a timestamp. In Table 3.1 an example of the data-string is shown.

Table 3.1: Structure of positioning data

Tag-ID x y z Timestamp
0x001337 16.09 13.85 1.76 1575936000.05

19



The data could either be used to show real-time positions of forklifts in the production
or be saved to enable analysis. Data-storage was done with the help of a Linux service
set up to run a Python program. The service acted as a safeguard for potential problems
like power failure that would cause the computer to restart. To increase usability the
data was sorted in timestamp intervals spanning 24 hours and saved in files named after
the corresponding date. As a first step a heat-map, see Figure 3.1, over the test facility
was created. The heat-map showed how the data were distributed, and if the collected
data was reliable. When the reliability of the data had been confirmed a database was
constructed to allow searches in the extensive data. Using SQL quarries could isolate
this information from the 3218828 lines of data to visualize routes of specific forklifts.
In figure 3.2 a query and the result is shown.

Figure 3.1: Heatmap over data ploted on the map

20



Figure 3.2: Example of Query search in database

To further push how quantitative data could be used to better understand it a web-
based GUI was created. Real-time data from the RTLS-API was visualized on a local
host. This required a server setup that connected with the RTLS through a TCP-socket,
and a client connected to this server. The client could then use the data to visualize
forklifts icons in real time on a custom map built using the JavaScript library Leaflet.
Figure 3.3 shows a screen capture of the running browser GUI.

Figure 3.3: Real-time applications running on a JavaScript tcp-socket

21



3.3.2 Qualitative data
The qualitative data mainly consisted of interviews but focus groups and testing ses-
sions were also used to gather data. Information was also gathered using unplanned
conversations with manufacturing employees.

Interviews

Both semi-structed and unstructured interviews were conducted. Semi-structed inter-
views uses pre-structed questions that are more formal than questions in a normal
conversation. Unlike structured interviews questions can be added or removed during
the sessions. It is also allowed to change the order in which the questions are asked.
This requires that the interviewer have a good understanding of the subject, so that
the added questions does not confuse the interviewee (Phillips & Stawarski, 2008).
Unstructured interviews is freer and allows the interviewee to interpret the asked ques-
tions. The focus is instead to use follow-up questions to gain a deeper knowledge of a
the subject (Trost, 2007).

Since the reserach aimed to have a bottom-up structure, the target of the interviews
mainly consisted of people that would use a positioning system in their work. This was
also a way to build a network of people to contact for future testing of system usages.
A few interviews was also conducted with professions responsible of investments in
factory settings to better understand the path of decisions. The interviews started with
a brief description of the study and its aims to generate a better understanding of what
information that was sought. The interviewees were encouraged to speak as freely as
possible to lower the risk of forced answers. Because manufacturing industries are very
protective when it comes to information regarding production cycles and processes, a
choice to not use recording equipment was made. This allowed the interviewees to
talk more freely without feeling monitored. In table 3.2 the full list of interviews is
presented.

22



Table 3.2: List of interviews

Interviewees Unstructured
interviews

Semi-structured
interviews

Logistic engineer 0 2
Plant manager 1 0
Production engineers 1 3
Production planner 0 1
I4.0 Transformation manager 1 1
Logistic and digitization manager 1 1
Logistic Strategist 0 1
Digital and solutions officer 1 1
Math Scientists 0 1
Production Engineering Phd. 0 1
Production System Phd. 0 1
Fork lift operator 1 0
Production staff 7 0
Sum 14 13

Focus groups

As part of the qualitative data gathering focus group sessions was conducted. Focus
groups in early stages of the research gives a better understanding of the subject and
helps with defining the research question. At later stages focus groups can be used as
a tool for interpretation (Esaiasson, Gilljam, Oscarsson, Towns, & Wängnerud, 2017).
The preferred size and setting of focus groups is 6-8 people from similar social and
cultural environments. (Liamputtong, 2011). But smaller focus groups can also be
used.

A successful focus group relies heavily on a non-threatening and open environment to
enable the participants to speak their minds (ibid.). Because of this focus groups have
a moderator that monitors the discussions, which also make sure that the group does
not stray too far from the subject. To enhance the data that is gathered from focus
group studies, two moderators can be used instead of just one. The advantage of this
is that one of the moderators can focus on the questions while the other can attend to
the answers. It is also easier to make sure that the session stays on track and hold up
to the time plan.

3.4 Identification of valuable system usage
The design process is not linear, it is iterative. Therefore when reading this section it
should not be interpreted as chronological. To increase the readability of the report the

23



process although is explained as chronological. The design process consisted of three
main phases, Understand, Explore and Evaluate. Iteration occurred mostly within
a phase, but also between the phases when the project team lacked knowledge.

3.4.1 Understand
According to the user centered design approach the data must be gathered from all
kinds of users which might interact with the system. The following stakeholders were
identified and interviewed during the understanding phase:

• Logistic engineer

• Plant manager

• Production engineer - medium size company

• Production engineer - global company

• Production planner

• Industry 4.0 Transformation manager

• Logistic and digitization manager

• Logistic Strategist

• Digital and solutions officer

• Math Scientists

• Production Engineering Phd.

• Production System Phd.

• Fork lift operator

• Production staff

At the test facility automation of the material handling fleet was recently preformed.
This system had similarities with the radio positioning system regarding data visual-
ization. The project responsible for the new material handling system, a production
engineer, had interesting takes on positioning systems and data usage from such sys-
tems. The team interviewed him and by combining his experiences with interview data
from interviews with the responsible person for the installation of the test system the
team concluded potential work roles at companies that generally had a stake in posi-
tioning data usage. This approach led to interviews with the corporate functions listed
above. By searching for companies within the research project scope at a reachable
distance from our location we manage to get in touch with both medium sized and
global production responsible.

24



Defining problems with AIM

Interview data were analyzed via the Affinity Interrelationship Method, AIM. Since
the research group consisted of fewer participants than suggested by the AIM method
the appointed task leader also took part in the analysing process. In order to analyze
the outcome from the qualitative data collection a study question were chosen. Each
member of the AIM session had 5 minutes to summarize the interview session with their
own words. The other member of the session were not allowed to criticize nor comment
while listening to the summary. Each participant formulated answers to the problem.
The answers were written on post-its, one answer per post-it. The team then followed
the AIM process and concluded the analysis. Figure 3.5 shows the board during the
AIM process and figure 3.4 shows a summarized result.

Figure 3.4: AIM outcome and its levels

25



Figure 3.5: One AIM board during analyze

Persona development

In figure 3.6 the constructed persona template is shown. This template is based on
the collected qualitative data the interviewees goals, pain points, motivators and back-
grounds were sorted out. The process was of bottom up character, where the many
interviewees attributes was sift in to a few distinct different personas.

26



Figure 3.6: Persona Template

Using this template, three unique personas was identified. Their roles consisted of Plant
Manager, Production Engineer and Production planner. The personas can be found in
Appendix A.

3.4.2 Explore
Based on the findings from the understanding activities ideas and concepts were to be
made. Problem descriptions from the AIM process second grouping session were used
as themes for the braindrawing activities. Even though the themes were based on the
AIM outcome, they were discussed and reworked. Since the purpose of the research
was to investigate potential usages for radio-based positioning systems in Industry 4.0
environments future applications needed to be considered. The AIM, however, only
focused on current problems. Because of this, a "future applications" theme was added
to enable discussion of industry 4.0 solutions. By stating the problem descriptions from
the AIM session on board including the future theme, the team combined and or split
the statements until six themes were brought up. The themes were the following:

• Driver support

• Safety

• Historical visualization

• Material carrier

• Work Studies

27



• Future applications

The selected themes were used in braindrawing sessions, where the team members based
on the themes draw ideas of how to use radio-based positioning data to solve the found
problems from the Understanding phase. Figure 3.7 shows an example of how the ideas
were presented. The braindrawing process suggests the drawn ideas were formed by all
team members. This makes it easier to scrap ideas.

Figure 3.7: The result of one braindrawing session

A total number of 60 ideas were produced during the braindrawing sessions. The ideas
were listed in a spreadsheet. Each idea was explained in words by the team. The
solution ideas were listed as described in table 3.3. To get a better understanding of
who would use the a specific solution three more columns were added, representing the
three personas. If it was likely that a persona would use the idea the box was checked.
This made analysis of a potential main-user possible by counting how many of the ideas
that had a fit with a specific persona. The idea-list was also used to build idea-clusters.
The clusters were ideas that had similar user and area of usage. From the 60 ideas 10
usage clusters was built:

1. A system usage based on zones, such as geo-fencing in manufacturing environ-
ments.

28



2. A system usage focusing on safety-features and possibilities of using radio posi-
tioning to increase safety in manufacturing environments.

3. A system usage that would visualize position data for material handling operators.

4. A route planning tool with live updated driver information (like indoor GPS).

5. A system usage focusing on object tracking.

6. A system usage facilitating FIFO operations.

7. A system usage which visualize Manufacturing KPI’s whith position data.

8. A system usage based on adaptive production setups.

9. A system usage focusing on large heavy industries like airplane manufacturing.

10. A system usage towards digital twin and data ponds.

Table 3.3: Organized Ideas with description

Idea name Description

Speed zones Divide the factory into speed zones that restrict the speeds for material carriers.
It could also establish temporary zones because of maintenance and similar activities.

Production equipment tracker Shows how much mobile production equipment has moved on,
for example, a shift, a day, or around a certain product.

Table 3.4: Ideas linked to personas

Idea name Description Production planner Production engineer Plant manager
Speed zones ... X X
Production equipment tracker ... X X
... ... ... ... ...

3.4.3 Evaluate
When moving from the explore phase in the Design Thinking process normally a ma-
terializing phase starts where concepts are tested and implemented. In this researched
this step was swapped with an evaluation phase where the system usages only were
tested. To evaluate problem areas that was found in the understand phase new meet-
ings with previous interviewees were set-up. This made it possible to not only evaluate
the findings but also ask additional questions regarding the production. A focus group
was also conducted. The result from the evaluation allowed for prioritization of the sys-
tems usages. This was then used to build the use-cases discussed in 1.2. The use-cases
can be found in Appendix B Examples of what was shown at these sessions is presented
in 3.8.

Evaluation with production engineer
Many of the concepts from the ideation-phase was linked to the production engineering

29



persona. Because of this additional interviews with individuals working in these roles
was considered necessary to evaluate the best system. Since interviews already had
been conducted with these individuals the understanding of the subject was good and
less explanations was needed. The system usage drafts was at this session presented as
they were intended to be shown to a potential customer. A semi-structured interview
format was used to ask questions when showing the content. Example questions were:

• Explaining a use-case

• Ask for a first opinion

• Ask for feedback of specific parts

• Ask for overall usefulness of use-case

Focus group
A focus group was conducted to test both the problem findings from the understand-
phase and the system usages. Unlike the other evaluation sessions the participants in
this focus group had not been part of the study earlier. This required a briefing that
explained the subject and the research aims. The group consisted of people experienced
in production and production planing. Discussions could because of this start at a
relatively advanced level. As a first topic the group discussed how positioning was used
today and what benefits came from using that data. This discussions was bridged to
productions problems and uncertainties originating from lack of positional knowledge.
By listening to the groups argumentation’s the defined problems from the exploration-
phase could be evaluated. The group was reminded that no answers were wrong and
encourage to speak freely.

To evaluate the early versions of the usage drafts the group was asked to describe how
they would use a positing system if it was given to them for free. The reason for stating
this scenario was that it removed economical aspects which would make it easier to
discuss usage more freely. It also allowed for the individuals to imagine a system that
was tailored to their own needs instead of the entire organization. By asking what
they would like to be able to measure, monitor or visualize data regarding industrial
needs of different roles were gathered. Finally the focus group was shown the system
usage drafts and was asked to asses their usefulness. The group was encouraged to
be critical of the content and reminded that the data quality depended upon their
honesty. Answers aligning with previous data was used to reinforce earlier observations
while critique acted as a tuner to create better usages.

30



Figure 3.8: Visualization of different velocities and a path at the test site

31



4
Results

This chapter will present identified problems that indoor positioning can solve, identified
users of an indoor positioning system and how the system can generate value for the
identified users.

4.1 Identified problems that indoor positioning can
solve

The interview data suggested that one of the biggest problems on the production sites
was the lack of high-quality data, easily accessible data regarding the production. Fur-
ther, the analysis allowed for this problem to be divided into two themes of data-related
problems. These themes were: missing data and misused data. Missing data refers to
unavailable data that users thought would benefit decision-making and improvement
suggestions. Misused data refer to low-qualitative data that still is used for calculations
of KPI-values.

4.1.1 Missing data
A majority of the interviewees reported that they lacked data to support the use of data-
driven decisions. Typical things they wished they had more information about was the
localization of equipment and vehicles, variation in the production, and throughput
times of products. The uncertainty of these parameters caused decisions to rely on
employees’ production experience, rather than actual production data. Not being able
to localize equipment was also described as a source of frustration when it caused
disruption and delays in the production.

The tested system shows promise when it comes to tracking bigger objects in an indus-
trial setting. Especially vehicles since they can supply a positioning tag with electricity.
By analyzing the historical data, graphs could be drawn to show up-time and route vari-
ations of the material carriers and other objects. This can increase feedback on daily
activities and makes comparison between different production cycles possible. The in-
terviews also revealed that information regarding the specific tool operations like which

32



tool that performed a certain task, was very valuable. By using positioning data to
track the location of the tools and linking it to activity this could be achieved.

4.1.2 Misused data
When asking interviewees to describe the process of data usage in their work, problems
regarding misused data emerged. Misused data in this instance refers to data of low-
quality that still is used. Low-quality data was stated by our interviewees to frequently
yield odd-looking KPI’s, such as OEE-values exceeding 100%, which is impossible.
This was often caused by human errors during data collection that resulted in faulty
measurements. The problem is so common that the production staff has learned to
simply ignore KPI outliers.

4.2 Identified users of an indoor positioning system
Data from the interviews and observations was also used to identify three personas
that would use the system output. However, from the ideation process it became clear
that most ideas were in some way connected to the production engineer, see figure 4.1.
Because of this, this persona is highlighted in the result to emphasize the importance
of having this role in mind when developing indoor positioning. Apart from being the
main user of such a system, the production engineer also was perceived as the role with
best knowledge regarding system requirements.

In one of the interviews the production engineer stated that The biggest problem is not
that we don’t have data, it is that data that is gathered finds no later usage and We have
a lot of routing problems that stems from the low quality of the factory floor. As stand
alone comments they would still have been interesting but it is the combination that
shows the width of the production engineers scope. Both physical problems and data
gathering projects concerns this role. This means that the production engineer needs
systems that can give increased visibility of the production e.g. position data.

Another example of how data pointed towards the production engineer as the main user
came from the evaluation sessions of the ideas. A quote from this sessions was I want
all of this, but first I must know that it is economically defensible. The interest of the
possibilities of indoor positioning were unmatched by other roles that was encountered.
Production engineers also had a better understanding of the relation between system
integration from an economical point of view than other factory workers.

33



Figure 4.1: The production engineer persona

An additional way to visualize this is to show how the different personas related to
the organizations structure described by Mintzberg (1979). The production engineer
mainly work with optimizing the manufacturing processes unlike the plant manager
and production planner that works much closer to the core structure, see figure 4.2.

Figure 4.2: Mintzbergs organizations structure related to personas

34



4.3 Identified value adding system usages for main
users

The qualitative data suggests that indoor positioning have potential to improve em-
ployee safety, increase productivity and gain insight in production flow. However, the
interview data showed ambiguous results regarding how the potential best is realized.
Individuals working in the manufacturing industry tended to highlight that indoor posi-
tioning solutions must be able to communicate with other systems. When analysing the
possible solutions for positioning systems it was clear that some solutions required more
data than just positioning data, especially when the data would be used for production
controlling.

Industrial researchers also showed a positive attitude towards indoor positioning but
said that many of the terms regarding the subject needed clarification. The lack of
definitions regarding technical terms, such as digital twin, was described as a friction
for communicating between stakeholders in the manufacturing industry.

This study suggests that positioning systems in manufacturing industries should ini-
tially be used for:

• Increasing safety and productivity at the manufacturing site

• Supplying data for manufacturing KPI calculation

• Manufacturing process analysis

4.3.1 Increase safety and productivity at the manufacturing
site

Using data from the test system at SKF enables calculation of near misses of the
material handling equipment. A near miss happens when two vehicles, or one human
and a vehicle, are closer than a certain "near miss"-threshold. By counting the number
of near misses in different areas heat-maps can be drawn. This information can be used
to visualize areas where accidents are more likely to occur.

Integration is required to enable control over material handling equipment. However, if
this limit is removed geo-fencing in the manufacturing is made possible. Geo-fencing is a
digital geographic zone where connected vehicles can be controlled (Trafikverket, 2020).
Controlling actions in these zones would enable equipment to dynamically change the
speed to fit the surrounding. If no humans are close by higher velocities can be allowed.
This will provide solutions that can increase flow, which generate a higher ratio of value
added activities according to Modig and Åhlström (2012).

35



4.3.2 Supply data for manufacturing KPI calculations
Positioning data can be used to calculate several common manufacturing KPIs. Exam-
ples of such KPIs are Idle time, Availability Rate, Down time and On Time Delivery.
Data from the SKF test site can be used to visualize Idle vs Operating time for three
different forklifts, see figure 4.3.

KPIs are used to monitor the production process (Amrina et al., 2018) and manu-
facturing responsible acts upon these indicators. Alongside the Industry 4.0 principal
decentralization decisions should move closer to the devices (Herman, Pentek, 2016).
KPI calculations from positioning data can be used for controlling the process directly
without human interference. An example of this inherits from the On Time Delivery
KPI (See 2.1). By letting the production process control systems prioritize orders the
process could automatically prioritize urgent orders.

36



Figure 4.3: Utilization of three different forklifts at the test site

37



4.3.3 Manufacturing process analysis
Lean principals, as bottleneck tracking (Liker & Meier, 2006) are common approaches
to production problems. Analyzed positioning data can support production engineers to
detect bottlenecks. Figure 4.4 shows two different rounds on the same route performed
by the same forklift, but at different times. Analyzing historical data will also enable
better understanding of AGV behaviours. Today this data is only available during a
limited time after error raises. Using an indoor positional system would remove this
limit and allow for comparison of problem scenarios.

Figure 4.4: Comparison between two forklifts at the SKF test site

38



5
Discussion

This thesis has investigated what problems that can be solved with radio-based posi-
tioning in the manufacturing industry. Identified problem areas were then linked to
the potential users of such a positioning system. By applying design thinking method-
ologies, solutions were developed to generate the most value for identified users. This
chapter will discuss the thesis results, but also the learning process that took place
during it. The aim is to increase the understanding of the result by explaining the
learning in an iterative process.

5.1 Result interpretation and implications
The result support the claims of Samama (2019) that addresses the need for clear
constraints of an indoor positioning system. By specifying the technical requirement
downsizing of the number of potential solutions is made possible, allowing for compar-
ison and priorition. This will likely benefit both user and developer because it forces
communication in a more narrow field. However, the results also transcend the need for
technical constraints and shows that better understanding of the end users first must
be satisfied to develop meaningful solutions. This is done by highlighting the produc-
tion engineer persona to emphasize the need for system functionalities that deals with
production optimization.

Cooper (1984) describes that strategies that finds a balance between technological
prowess and a strong market orientations are those who sees the best results (s.156).
A strong market orientations should in this statement be interpreted as product de-
velopment deriving from market demands (ibid.). Indoor positioning does not see any
commercial use as of now (Samama, 2019), because of this identifications of system
users and definitions of their needs must first be established to locate the market de-
mand. This makes the result of identified system users particularly interesting because
they are rarely in charge of system purchases, which mean that they often have little
influence of what system will be accuired.

Case studies limits generalization and analyze through statistical inference (Dubois &
Gadde, 2002). The strength of the case approach instead comes from its ability to in-

39



depth insight in empirical phenomenons. This is especially true if the framework and
theory is allowed to evolve through iteration during the research (ibid.). The design
thinking method relies on the same back-and-forth mechanism to continuously add to
problem and user understanding. Because of this the result should not be used to
make broad statements about the manufacturing industry, but instead used as insight
into what problems that indoor positioning can solve. The validity of this result will,
therefore, rest on the iterative process rather then its ability to give general answers to
the research questions.

Combining positioning data from the test-site with interview findings from industry
stakeholders yielded results that showed that there is an interest in the usage of data
generated by a positioning system. Waste elimination is crucial to approach a lean
manufacturing process (Liker & Meier, 2006) and there is a need for tools that can sup-
port data-driven decision making (Amrina et al., 2018). Position data collected over
time will give the engineers working in manufacturing environment high-resolution data
to base their analyses and KPIs on. Compared to manually collected data, data from a
continuously operating positioning system will generate more cost effective data. Man-
ufacturing engineers are a limited resource. Systems enabling more time for analyzing
data and less time spent on data collection will make it possible to utilize the capacity
of the manufacturing engineers more economically.

This study also shows the upsides of a Gemba approach where interesting findings come
from close contact with the manufacturing environment (Richardson & Richardson,
2017). If all data collection for KPI measurement in a future manufacturing scenario will
be atomized the manufacturing engineers might miss potential interesting findings in
the manufacturing process during data collection. Close observation in manufacturing
environments was an important part of understanding how indoor positioning could be
used. Because of this, it is hard to recommend the production engineer to solely rely
on a positioning system to analyze processes. Our findings instead suggests that using
data to complement observations during Gemba is a more suitable system usage.

One main problem found in this study is the misuse of data by the manufacturing
industry. Accepting that positioning data likely will be misused as well one must find
ways for proper data usage. However, interpreting the analyses from positioning data
with real know-how of the process setup decreases the risks of data misuse. Independent
of the data quality, the interpretation of the analyzes will be easier if the engineers have
solid process know-how, gathered from Gemba.

The result also showed that many of the ideas required further investigation of sys-
tem boundaries to enable viable solutions. This study shows that material handling
control systems can not be accessed by a superior system. This limits Ericssons test
system to only monitoring functions, since controlling actions such as giving a material
carrier allowance to a certain area requires permission from the carrier manufacturer.
Future studies is therefore encourage to analyze possible joint system development with
other manufacturing equipment companies. In collaboration; system boundaries could

40



become more clear, which would enable a more focused market strategy.

Learning in iterative processes
As mentioned in the method; answering the research question was not a linear process.
Instead iterations were used to continuously update the framework, theory and overall
understanding of the subject. This ‘back and forth’ motion over different research
activities is in this study portrait as a unique trait of the design thinking process.
In reality, however, this movement is the underlying mechanism of learning that allows
new knowledge to influence the end result. Dubois and Gadde (2002) describes learning
as the essence of all research projects. Despite this, the readers seldom are allowed to
follow the actual learning process of the writers, even though it is the most important
part of the research process(ibid.). Just like in any subject the transformation from
tacit to explicit (Werr, 1998) knowledge could aid other researchers by giving insight
into how ideas take form during a project.

In the early stages of the research time was spent reading up on the subject to under-
stand both technical and organizational aspects of indoor positioning. Understanding
the vocabulary used in a certain field is of-course a must when trying to understand
a subject. It was especially important since lack of subject knowledge would make it
difficult to construct relevant interview questions.When interacting with users in the
early stages of design projects, it is important to try different approaches and constantly
iterate a set of questions and topics to find the most suitable way for accessing the users
knowledge.

The research project was performed with Ericsson as a project sponsor. When the
project started, the stakeholders had interpretations of how the system output might be
used. Although the research team was free to choose the research method. Going to the
users for answers as done in design thinking (the chosen method) includes defining the
problem. But what if the defined problem can not be solved with the project sponsors
offered product or service? The research team understood that the stakeholders and
the users had some different thoughts on how to use positioning data.

Questioning "why" when performing a design project enables double-loop learning (Argyris,
2002) but the research team can easily step outside the project scope when questioning
the purpose of the project. The user needs might not be compatible with how the
system is designed. Using iterative design methods like design thinking enables many
learning opportunities for the research team, but it can also create conflicts with the
sponsor’s goals. Therefore, it is important to set up and maintain concrete limitations
for the project in order to get clear results. In this project, we have tried to identify
user needs and define the ones that fit within the project scope. But one could also
have investigated the user needs without the limitations, and questioning "why" even
further. Such double-loop approaches in an iterative manner open up for answers far
away from clear answers and thus are of little use for project sponsors.

41



Some knowledge gained during this project is the result of such double-loop learning
processes. When describing the research methods and results one tries to transfer the
gained knowledge to the reader. We as researchers want to give the reader the best
possible chance to learn from our study. But adding all findings from the research
project to the report will blur the message. Therefore only the most important findings
and the process steps taken to reach them are added to the report for the highest
possible readability.

5.2 Ethical aspects
Information about where objects are located can be sensitive, both for the employees
and the company where they work. Manufacturing companies suffer from competition
from other actors on the market. Manufacturing setups and value chains can be the
difference between a successful company and an unsuccessful company. Therefore keep-
ing positioning data secret is crucial for the company. A positioning system must not
share information to unauthorized users. Even though the positioning data might be
classified, data about dead objects don’t interfere with GDPR laws.

But positioning data, and visualized positioning data where humans are involved re-
quire consideration about integrity. The GDPR laws must not be violated when doing
research. Our research methods in this project involve interviewing people and observ-
ing work procedures. To ensure that no integrity violations against GDPR laws occur
we must keep the data about people anonymous. One must not be able to identify the
interviewees by reading this report.

Keeping data anonymous is also important when visualizing positioning data via a
positioning system. Among colleagues, visualized positioning data might be used in a
way that is not intended by us who developed it. For example, workplace bullying is
experienced by many employees every day 1, and positioning services might be used
by staff for bullying. By visualizing poor performance together with user data the
positioning system could be used for matters that the unions would reject.

Protest against positioning systems has previously been held by the unions 2. The
surveillance possibilities for corporate organizations increase with positioning systems.
For Ericsson company, as a provider of positioning services, integrity must be considered
before the system is launched.

Lastly, positioning data provides information that can be used to control the con-
nected equipment. Inefficient use of the equipment can be spotted, and the fleet can
be managed and used in better ways with positioning services. Higher throughput of
the installed equipment will in an extent put less environmental footprint since the
consequence of investing in un-utilized due to lack of information can be prevented.

1https://www.bbc.com/news/business-51182651
2https://www.svt.se/nyheter/inrikes/kritik-mot-gps-overvakning-av-personal

42



6
Conclusion

This thesis aimed to investigate if there is a potential to solve problems in the manu-
facturing industry with radio-based indoor positioning. To do this, problems regarding
data usage were identified and how a positioning system can be used to increase pro-
ductivity, gain insight into the production flow, and improve employee safety. Based
on qualitative analysis it could be concluded that indoor positioning is perceived as
a promising technology with the potential to improve production processes. However,
this requires the solutions to have a strong connection to the system users and the
manufacturing industry KPI’s.

The data also identified three personas that would work with a positioning system,
and the production engineer as the main user. Further more, analysis using the AIM
method reveled that problems in the manufacturing industry could be grouped into -
Data is missing and Data is misused. From this, value adding usages for the production
engineer were identified by using design methodologies described in section 2.5. The
best system usages were found to be:

• Increase safety and productivity at the manufacturing site

• Supply data for manufacturing KPI calculation

• Production process analysis

This study has confirmed the potential of radio based indoor positioning both in theory
and with system user evaluations. The literature emphasized the focus on technical
constraints. By applying design thinking methodologies this research have aimed to
expand this by also involve the user perspective and Lean principles regarding waste
eliminations. Denning (2013) describes design as the creation of artifacts and innova-
tion as the process of making people adopt the artifact into the organization. We are
convinced that understanding of the manufacturing system users and the environment
where they operate is key to develop ideas that can go from designed applications to
used innovations.

43



References

Alänge, S. (2009). The affinity-interrelationship method aim. A Problem Solving Tool
for Analysing Qualitative Data Inspired by the Shiba “Step by Step” Approach.

Amrina, U., Firdaus, A., et al. (2018). The selection of productivity key performance
indicators for car manufacturing companies using integrated performance mea-
surement system. Sinergi, 22 (2), 101–106.

Argyris, C. (2002). Double-loop learning, teaching, and research. Academy of Manage-
ment Learning & Education, 1 (2), 206 - 218.

Basalla, G. (1988). The evolution of technology. Cambridge University Press.

Benyon, D. (2014). Designing interactive systems : a comprehensive guide to hci, ux
and interaction design. Edinburgh: Pearson Education.

Börnfelt, P.-O. (2009). Arbetsorganisation i praktiken: en kritisk introduktion till
arbetsorganisationsteori. SNS förlag.

Clardy, A. (1997). Studying your workforce. Sage.

Cooper, R. G. (1984). New product strategies: what distinguishes the top performers?
Journal of Product Innovation Management, 1 (3), 151–164.

Curedale, R. (2019). Design thinking : process & methods. Los Angeles, CA: Design
Community College Inc.

Denning, P. J. (2013). Design thinking. Communications of the ACM , 56 (12), 29–31.

Dooley, L. M. (2002, aug). Case Study Research and Theory Building. Advaces in
Developing Human Resources, 4 (3), 335–354. doi: 10.1177/1523422302043007

Dubois, A., & Gadde, L.-E. (2002). Systematic combining: an abductive approach to
case research. Journal of Business Research, 55 (7), 553–560.

Dudovskiy, J. (2016). The ultimate guide to writing a dissertation in business studies:
A step-by-step assistance. Pittsburgh, USA.

Eisenhardt, K. M. (1989). Building theories from case study research. Academy of
management review, 14 (4), 532–550.

44



Esaiasson, P., Gilljam, M., Oscarsson, H., Towns, A. E., & Wängnerud, L. (2017).
Metodpraktikan : konsten att studera samhälle, individ och marknad. Wolters
Kluwer.

Gibbons, S. (2016). Design thinking 101. Retrieved July 31, 2016, from https://
www.nngroup.com/articles/design-thinking/

Given, L. (2008). The SAGE Encyclopedia of Qualitative Research Methods. 2455 Teller
Road, Thousand Oaks California 91320 United States: SAGE Publications, Inc.
doi: 10.4135/9781412963909

Gu, Y., Lo, A., & Niemegeers, I. (2009). A survey of indoor positioning systems for
wireless personal networks. IEEE Communications surveys & tutorials, 11 (1),
13–32.

Hartson, R., & Pyla, P. S. (2012). The UX Book: Process and Guidelines for Ensuring
a Quality User Experience. doi: 10.1016/C2010-0-66326-7

Herman, Pentek, O. (2016). Design Principles for industrie 4.0 Scenarios.
Retrieved 2020-01-28, from https://ieeexplore.ieee.org/document/
7427673?arnumber=7427673{&}newsearch=true{&}queryText=industrie4
.0designprinciples

Kirkeby, O.-F. (1990). Abduction in: Vetenskapsteori och metodlära. en introduktion,
ed. h. Anderson. Studentlitteratur .

Liamputtong, P. (2011). The use of focus group methodology in the health and social
sciences. Focus Group Methodology: Principles and Practice, 87–106.

Lichtman, M. (2014). Qualitative Research for the Social Sciences. 1 Oliver’s Yard,
55 City Road London EC1Y 1SP: SAGE Publications, Inc. doi: 10.4135/
9781544307756

Liker, J. K., & Choi, T. Y. (2004). Building deep supplier relationships. Harvard
business review, 82 (12), 104–113.

Liker, J. K., & Meier, D. (2006). The toyota way fieldbook: A practical guide for
implementing toyota’s 4ps. New York, NY: McGraw-Hill.

Loranger, H. (2014, Aug). Ux without users is not ux. Retrieved from https://
www.nngroup.com/articles/ux-without-user-research/

Mautz, R. (2012). Indoor positioning technologies.

Mintzberg, H. (1979). The structuring of organizations. Prentice-Hall, Inc., Englewood
Cliffs.

Modig, N., & Åhlström, P. (2012). Detta är lean: lösningen på effektivitetsparadoxen.
Stockholm School of Economics (SSE) Institute for Research [distributör].

45

https://www.nngroup.com/articles/design-thinking/
https://www.nngroup.com/articles/design-thinking/
https://ieeexplore.ieee.org/document/7427673?arnumber=7427673{&}newsearch=true{&}queryText=industrie4.0designprinciples
https://ieeexplore.ieee.org/document/7427673?arnumber=7427673{&}newsearch=true{&}queryText=industrie4.0designprinciples
https://ieeexplore.ieee.org/document/7427673?arnumber=7427673{&}newsearch=true{&}queryText=industrie4.0designprinciples
https://www.nngroup.com/articles/ux-without-user-research/
https://www.nngroup.com/articles/ux-without-user-research/


Phillips, P. P., & Stawarski, C. A. (2008). Data collection: Planning for and collecting
all types of data (Vol. 175). John Wiley & Sons.

Reinhard, G. (2016). A strategist’s guide to industry 4.0. Retrieved 2020-
01-27, from https://www.strategy-business.com/article/A-Strategists
-Guide-to-Industry-4.0?gko=a2260

Richardson, T. M., & Richardson, E. (2017). The toyota engagement equation : How to
understand and implement continuous improvement thinking in any organization.
McGraw-Hill Education.

Samama, N. (2019). Indoor positioning: technologies and performance. New Jersey :
Wiley-IEEE Press.

Trafikverket. (2020). Geofencing. Retrieved from https://www.trafikverket.se/
resa-och-trafik/forskning-och-innovation/aktuell-forskning/
transport-pa-vag/geofencing/

Trost, J. (2007). Kvalitative intervjuer. Pozkal: Studentlitteratur.

Werr, A. (1998). Managing knowledge in management consulting. In Academy of
management proceedings & membership directory (p. A1 - A7).

Wikberg Nilsson, A. (2015). DESIGN process och metod. Lund: Studentlitteratur AB.

Womack, J. P., Jones, D. T., & Roos, D. (1990). The machine that changed the world,
rawson associates. New York: Free Press.

Yin, R. K. (2014). Case study research: Design and methods (applied social research
methods). Sage publications Thousand Oaks, CA.

46

https://www.strategy-business.com/article/A-Strategists-Guide-to-Industry-4.0?gko=a2260
https://www.strategy-business.com/article/A-Strategists-Guide-to-Industry-4.0?gko=a2260
https://www.trafikverket.se/resa-och-trafik/forskning-och-innovation/aktuell-forskning/transport-pa-vag/geofencing/
https://www.trafikverket.se/resa-och-trafik/forskning-och-innovation/aktuell-forskning/transport-pa-vag/geofencing/
https://www.trafikverket.se/resa-och-trafik/forskning-och-innovation/aktuell-forskning/transport-pa-vag/geofencing/


A
Personas

i



ii



B
Use-case

iii



Positioning
use-cases for 
manufacturing industries



Digital Zones are a defined area where one can 
measure different parameters and declare local rules. 

Digital zones are based on a positioning technology, 
that enables real time location updates from 
connected objects. 

Whatis Digital Zones? 



Around objects: 

Zones can be defined around objects. All objects 
connected to the positioning system can be 
surrounded by a zone.  Peopling carrying connected 
gadgets can also be positioned. 

In the factory:

At the shop flor, areas can be divided in zones. These 
zones are not connected to a mobile object, but to 
virtual geometrical boundaries.  

Different zone-types



The manufacturing industry constantly working 
towards lower costs, increased productivity, and safer 
work environments. Digital zones can: 

• Ensure safer work environments
• Increase productivity
• Lower the costs

Benifits  for  the 
manufacturing  industry



Rules in digital zones can be declared in order to 
control objects connected to the positioning system.  
Suitable rules for a manufacturing industry are: 

• Speed limitation
• Allowed operating areas
• Allowed operations (see quality assurance 

use-case)

With connected people the advantages are even 
bigger. For example, when the system recognizes a 
person and a vehicle in the same area vehicle slows 
down until there is no risk for collision. 
Operators and machines can share the same space 
with out risk of injured staff. The need of physical 
fences and obstacles will be unnecessary

Use-case: Geofencing



Quality assurance according to ISO standards (e.g. 
ISO 9001) requires supervision of production 
parameters. Connected equipment delivers data to 
databases. It’s important to know which product 
which values refer to. 
At assembly lines, this can be done by knowing the 
product’s position and the tool position. Common 
in the automotive industry is to use wired tools, 
with a certain cable length.  The tool thereby is 
only  allowed operate on one product at the time. 
A radio positioning system will allow you to use 
wireless production equipment instead of wired.

Use-case: ISO Quality 
assurance



• 3 classes

• Class 3: 

• No supervision

• Class 2: 

• Torque supervision

• No data storage

• Class 1:  (CC / SC)

• Torque supervision  

• Rotation angel supervision

• Data storage

• Class 1,2 is Positioning critical (You must know which product 
you're working on)

Use case example –Screwdriver



Use-case: Heatmaps

Based on position data heatmaps shows the most 
intensive zones. All tracked objects can be 
monetarized this way.

One can set up zones around objects and in the 
facility. Then it is possible to track if, when, and 
which objects that might conflict. The number of 
near misses, or collisions can be calculated based on 
positioning data.  



Technicalrequirements

Monitoring:

• For monitoring activities the system requires 
coverage where you want to monitor.

• All objects that being monitored must be 
connected to the positioning system. 

• For monitoring activities at least 2 anchors 
must be within reach.



Technicalrequirements

Controlling:

Besides the requirements for monitoring 
controlling activities has further technical 
requirements:

• Connected objects (eg. forklifts). The controlled 
object must be connected to a network (LAN). 

• The controlling system must connect to the 
vehicles CAN* bus.

• A defined decision hierarchy between the 
system and the vehicle. 



Description:

A Controller Area Network (CAN bus) is a robust vehicle bus 
standard designed to allow microcontrollers and devices to 
communicate with each other's applications without a host 
computer. It is a message-based protocol, designed originally 
for multiplex electrical wiring within automobiles to save on 
copper, but can also be used in many other contexts. Definition 
from Wikipedia.

There is no standard for communication between superior 
systems and material handling equipment in the manufacturing 
industry according to our interviewees. 

It is complicated to get sensor data from the vehicle CAN bus. 
Therefore the vehicle might ask the positioning system for 
permission rather than being order by the positioning system. 

CAN bus Communication in vehicles



Vehicle data possible to successfully 
combine with positioning data
The manufacturing engineering analyzes can be enhanced by combining 
positioning data with other data formats.

• Data from a connected vehicles (e.g. forklift) CAN bus system:
• Speed
• Steering angle
• Telescope position 
• Actuator information
• Break information
• Accelerometer information



Costexamplegeofencing
The use-case includes geofencing where one can set-up and manage speed 
zones. This will enhance safety but in larger facilities it can also increase 
productivity.  Here is an Example: Increase of vehicle speed in safe zones. 
Investing in a system like this, at a facility with 100 manual non automated 
vehicles an increased maximum vehicle speed  at safe areas from 6 to 12 
Km/h might improve the productivity with up to 17,5 %.  Since 1 driver + 
forklift costs 680 000 per year, a decrease from 100 to just 98 drivers will ret.

INVESTMENT COST: 1 000 000 

RUNNINGCOST:  250 000 / year

SAVINGS: up to 11Msek

PAY OFF:  1 year



Costexamplewirelessscrewdriver 
At a large automotive manufacturing sites visited during the research project 
changes in the production setup occurs on daily basis. Screw bandage is one 
of the most used fastenings at automotive sites.  For CC classes 2 and 1 
positioning of the tool is required. Average  time to move a stationary 
screwdriver (mounted via cable) is 16h. This operation can not be done during 
running production, it needs to be scheduled to a production free period (e.g. 
weekend). 

Moving a CC class 2 or 1 screwdriver will cost approximately: 

This is excluding the soft costs regarding changes in the system setup. The un-
flexible stationary screwdrivers also makes the production engineers limited to 
areas where these are installed for certain operations. Therefore some minor 
improvements are not possible and the production process is not as flexible as 
it could be. 



• Large Manufacturing sites 
• Intensive Logistic Operation Companies
• Workforce intensive manufacturing
• Companies with a mix of automatic and 

manual activities

Potential 
Customers





Manufacturing 
process analysis

Positioning
use-cases for 
manufacturing
industries



The manufacturing industry constantly improves its 
production processes. The improvement process vary 
from site to site, but common approaches divers from 
the LEAN philosophy and  or the SIX SIGMA 
methodology.

The improvement process often starts with an 
identified production problem, for example 
production capacity issues. Manufacturing process 
analysis refers to the analyses performed to solve the 
issue. 

What is Manufacturing 
process analysis? 



Manufacturing engineering analyses and 
work studies:

When performing a manufacturing analyze you can use 
position data to compare different work setups. 
Traditionally these analyzes are done manually on paper, 
but with a positioning system, you can perform multiple 
analyses simultaneously.  Continues data collection can also 
assure good data quality over time, and for different 
shifts/workforces. 

Work studies can be performed on equipment and 
workforce. Compared to manual work studies you’ll get a 
better resolution on the collected data.



Types of manufacturing engineering 
analyses and tools where positioning data 
can be used:
Examples of work studie activites where 
positiondata can be improving the quality:

• WBB – Work Balance Board. 

• VSM – Value Stream Mapping. 

• Frequency studies.

• Bottle neck tracking

Examples of what position data can show: 

• Process Deviations

• Wait time

• Utilization

• Best practice / Path



Manufacturing engineering software (potentilal
analyzing software) : 
• MVV (Casat)
• AXXOS
• NOVOTEK



KPIs compatible with position data:

The manufacturing industry constantly measures different 
Key Performance Indicators to monitor the production 
processes.

These indicators rely on data provided by the supervision 
systems (SCADA) and aims to describe the process current 
and historical state. 

KPIs possible to calculate with positioning data: 

• OEE – Overall Equipment Effectiveness
• AR –Availability Rate
• PR – Performance Rate
• FTY – First Time Yield
• FPY – First Pass Yield
• VA – Value Adding time
• DT – Down Time
• Idle
• ELR – Employee Late Ratio
• On Time Delivery



Advantages with positioning data for 
manufacturing analyzes  
Processes with large operation sites or with high equipment 
density has complex flows of booth material and equipment. 
All production system has bottlenecks, processes where the 
production capacity is limited. At complex operation sites 
with many tools and equipment carriers, these bottlenecks
can be hard to find, since they are caused by dependencies 
between the product and the equipment. By tracking the 
equipment you provide the engineers with reliable 
information. 

Positioning data can be used for visualizing:

• Reduce variation between shifts
• Reduce the distance equipment is moved
• Reduce the number of equipment movements
• Plan when to move equipment
• Calculate the right number of equipment
• Fast reliable data when abnormalities occurs.




	Introduction
	Background
	Purpose
	Limitations
	Research Questions

	Theory
	The indoor positioning problem
	The Manufacturing Industry and how it is measured
	Industrial evolution
	Lean Prodcution
	Gemba
	KPI - Key Performance Indicators in the manufacturing industry

	Organizational structures according to Mintzberg
	Industry 4.0
	Design methods
	Design Thinking
	Affinity Interrelationship Method -AIM 
	Braindrawing

	Persona
	Knowledge gaining and transfer

	Method
	Research design
	Case Study
	Systematic combination

	Litterature study
	Data collection
	Quantitative data
	Qualitative data

	Identification of valuable system usage
	Understand
	Explore
	Evaluate


	Results
	Identified problems that indoor positioning can solve
	Missing data
	Misused data

	Identified users of an indoor positioning system
	Identified value adding system usages for main users
	Increase safety and productivity at the manufacturing site
	Supply data for manufacturing KPI calculations
	Manufacturing process analysis


	Discussion
	Result interpretation and implications
	Ethical aspects

	Conclusion
	References
	Personas
	Use-case