Content uploaded by Peter Funk
Author content
All content in this area was uploaded by Peter Funk
Content may be subject to copyright.
FACTORY-IN-A-BOX – SOLUTIONS FOR AVAILABILITY AND
MOBILITY OF FLEXIBLE PRODUCTION CAPACITY
Mikael Hedelind1, Mats Jackson1 and Peter Funk2
Mälardalen University,
1Department of Innovation, Design and Product Development
2Department of Computer Science and Engineering
1P.O Box 325, SE-631 05, Eskilstuna, Sweden
mikael.hedelind@mdh.se
Johan Stahre and Rikard Söderberg
Chalmers University of Technology,
Department of Product and Production Development
Johan Carlsson
Fraunhofer Chalmers Research Centre
Department of Industrial Mathematics
Mats Björkman
Linköping University,
Department of Management and Engineering
Mats Winroth
Jönköping University,
Department of Industrial Engineering and Management
Abstract: The objective of this paper is to present examples of how to realize a flexible and reconfigurable
production system. An ongoing research project in Sweden called Factory-in-a-Box will be presented which is
one research initiative within this area. The purpose of the Factory-in-a-Box project is to develop solutions for
mobile production capacity on demand. Three key features have been identified as enablers for these kinds of
production capabilities: mobility, flexibility, and speed. The concept consists of standardized modules that can
be installed in e.g. containers and easily transported by trucks, rail vehicles, and boats. The modules can easily
be combined into complete production systems and reconfigured for new products and/or scaled to handle new
volumes. The goal of the Factory-in-a-Box project is to build fully operative production modules that are
developed in close cooperation between different academic and industrial partners. This paper will present the
reults from these demonstrators giving examples of the usability of the Factory-in-a-Box concept in industry.
Keywords: Flexibility, Reconfigurability, Mobility, Speed, Production System,
Demonstrator
1. INTRODUCTION
Globalization makes competition within
manufacturing more difficult to sustain, and it is
recognized that low cost and high quality alone are
not enough to guarantee a firm’s competitive position
in the market place. The uncertainty in markets and
rapid introduction of new products has created a
growing need for flexible, reconfigurable, and
responsive manufacturing systems. Meeting customer
demands requires a high degree of internal flexibility,
as well as the ability to reconfigure operations to suit
new demands.
The national Swedish ambition to increase the
number and size of Swedish manufacturing industries
is counteracted by lack of production capacity
available on demand and at any location. Companies
without in-house production capacity, have few
options if they need resources for pilot production.
Instead, manufacturing orders are often placed in
low-wage countries, e.g. China or Eastern Europe.
Thus, to develop the next generation of products and
services, there is a need to find and implement new
innovative methods and concepts that will support
industry in generating new ideas to quickly realize
these into successful products and competitive
production systems. There is a need for flexible and
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
1
reconfigurable production systems, enabling quick
product realization as well as flexibility and
reconfigurability (Jackson, 2000). The question is
how to develop and implement such a production
system?
The objective of this paper is to present and give an
example of how to realize a flexible and
reconfigurable production system. An ongoing
research project in Sweden called Factory-in-a-Box
will be presented which is one initiative within this
area (Jackson and Zaman, 2006). The objective of the
this project is to build fully operative production
modules that are developed in close cooperation
between different academic and industrial partners.
This paper will present the reults from these
demonstrators as examples of the usability of the
Factory-in-a-Box concept in industry.
2. METHODOLOGY
This paper is based on a literature review, a series of
interviews, as well as case studies that have been
used to collect data and experiences from the
development of the demonstrators in the Factory-in-
a-Box project. In collaboration with industrial
partners a number of problem areas was selected and
case-studies was conducted to explore the different
areas and the applicability of the Factory-in-a-Box
concept.
In general, case study method is the preferred
strategy when 'how' or 'why' questions are being
posed, when the investigator has little control over
events, or when the focus is on a contemporary
phenomenon within some real-life context (Yin,
1997), which is relevant for this work.
Case studies may be criticized for lack of statistical
reliability and validity. Furthermore, it is argued that
it is not possible to test hypotheses. To overcome this
dilemma, it is increasingly important to select a
representative case and to validate the result
continuously, and not simply at the end of the study.
It is also important to describe the actual case
carefully and only to draw conclusions that are valid
exclusively for similar systems.
3. THE FACTORY-IN-A-BOX CONCEPT
In a competitive environment, where the products
have the same performance, quality and
functionality, the process of developing products
within shorter intervals compared to the competitors
becomes increasingly important (Tidd et.al., 2001).
The product itself is a smaller part of the complete
offering to the customer, with branding, design,
financing, services, smart products and other aspects
becoming increasingly important. Fragmented
markets stress the need for abilities to continuously
adapt to new demands and to integrate new
technologies. In a business environment dominated
by change and uncertainty it is becoming more and
more important to define and sustain the competitive
advantage of the corporation.
Several different philosophies regarding future
production systems exist which share similar views
about changing and turbulent business environment.
One similarity in modern production system
philosophies is the idea of autonomy and modularity.
For example, Mass Customization, Agile production
systems and Holonic Manufacturing Systems are the
modern production philosophies, which are based on
the concepts of modularity and autonomy, see e.g.
Piller (2002), Yusuf et.al. (1999), Jackson (2000),
and Molinari et.al. (2004).
Production system philosophies as e.g. Mass
customization, Agile, or Holonic systems are not
very specific on how to reach the desired goals.
There is a need for applications and implementations
of these production system philosophies; the Factory-
in-a-Box concept is one initiative within this area
(Ask, 2006).
The Factory-in-a-Box concept consists of
standardized production modules that can be installed
in a container and transported by e.g. a truck or by
train. The modules may then rapidly be combined
into production systems that can be reconfigured for
new product and/or scaled to handle new volumes.
Production capacity may be provided as a mobile and
flexible resource that rapidly can be tailored to fit the
needs of a company, at a specific point of time. The
emphasis on mobility in the Factory-in-a-Box
concept is important in Sweden, where geographic
limitations are a reality.
The Factory-in-a-Box concept presents a future
possibility for a production-on-demand market.
Mobility, flexibility, and speed are order-winners on
that market. Three examples of customer segments
for the Factory-in-a-Box are:
• Company A has a new product design, but
without production capacity. The options for
the company are to invest in new production
capacity or to outsource production. If the
company chooses to outsource, chances are
that production will be placed abroad and that
the company may lose control of their
product. Company A may lease a Factory-in-
a-Box production system for the time needed
and then return the system to the lessor
providing the production modules.
• Company B has a peak in their production,
which exceeds their capacity to produce. The
company may need to outsource production
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
2
or make large investments. Company B may
lease production capacity in the form of a
Factory-in-a-Box as an alternative to
outsourcing.
• Company C is a large enterprise, which
wants to ensure quality and availability of
the manufacturing of a sub-system,
provided by a supplier. Company C can
help a supplier by providing a Factory-in-a-
Box module to improve quality or handle
volume variation.
It should be possible to place a Factory-in-a-Box
close to product development or customers within the
distribution chain. A likely scenario is that the
Factory-in-a-Box can be rented or leased from
production specialists, i.e. a type of functional sales
of production capacity. This would improve the
preconditions for a larger degree of production, and
hence product development, to remain locally.
There are a number of requirements for the Factory-
in-a-Box concept in order to realize the key features
of Mobility, Flexibility, and Speed. These
requirements are:
• Modules that are easy to transport to the
production site as well as to move them
within site, e.g. external and internal
mobility.
• Ability to dynamically adapt the degree of
automation and flexibility.
• Reconfigurability in order to meet changing
demand and automatic/semi-automatic
configuration of modules and system is
prerequisites for scalability for changing
production volumes and for fast ramp up of
the production.
The use of standardized production modules provides
autonomy and reusability. A Factory-in-a-Box
installed at a company should be integrated with the
company’s existing technical production capacity and
its present workforce. The intention is to balance
automation and manual labour in the Factory-in-a-
Box modules. Thus, each configuration will include
e.g. automation requirements, operator staff
requirements, configuration simulation modules, and
case-based experience/knowledge databases.
4. REALIZING A FACTORY-IN-A-BOX
The Factory-in-a-Box demonstrators have been
accomplished by teaming the resources and
knowledge of four running ProViking projects:
ExAct, DYNAMO, Flexible and Accurate
Automation, and Robust Design & Variation
Simulation. The Factory-in-a-Box project has thus
formed a unique combination of flexible and accurate
manufacturing technology, concepts for strategic
development of automation, and tools for intelligent
reuse of experience, knowledge, and information.
4.1 CONTRIBUTION FROM THE ROMUS
RESEARCH GROUP AND THE “DYNAMO”
PROJECT
The DYNAMO project is part of the ROMUS
research group and the research is conducted at
Jönköping University and Chalmers University. The
contribution of the DYNAMO project has provided
the Factory-in-a-Box concept with tools for design,
measurement, visualization, and management of
dynamic levels of automation (LoA). Automation
strategies have been investigated connected to the
demonstrator projects, proposing how to achieve the
right level of automation for each specific life-cycle
situation of the production module by utilizing LoA-
variation. The goal has been to increase robustness
during multiple stages of the module (such as start-
up, ramp-up, and operation) by varying automation
level of tasks related to operations, transfer, testing,
materials handling, and supervision when necessary.
LoA describes the relationship between human and
machine in terms of function/task allocation, and
should be set within a pre-defined span between 1
(manual) and 10 (full automation) for each task. For
each demonstrator, requirements have been specified
for how much the automation level should be
possible to vary. It means that requirements on
technology, IT-systems, and human skills are
necessary. Creating robustness is made during design
and reconfiguration where most parameters are set,
and LoA could, therefore, be utilized as a design
variable.
By implementing the research results from the
ongoing Dynamo project in the Factory-in-a-Box
demonstrators, it has been possible to illustrate the
potential of controlled dynamic automation levels for
both product and production practitioners. It has also
been a necessary precondition to make mobile
production capacity on demand possible.
4.2 CONTRIBUTION FROM THE DIVISION OF
ASSEMBLY TECHNOLOGY/PRODUCTION
SYSTEMS, LINKÖPING UNIVERSITY, OF THE
“FLEXIBLE AND ACCURATE AUTOMATION”
PROJECT
The “Flexible and Accurate Automation” (FlexAA)
contribution has been concentrating on the technical
realization of the Factory-in-a-Box concept and
demonstrators. The main research question for the
FlexAA project has been how to make the production
module/cell accurate and flexible enough to both
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
3
fulfil product quality requirements in an assembly- or
light-machining cell, and to do this rapidly and
effortlessly. The FlexAA project has contributed to
both the operation planning level as well as to the
hardware level.
On the operation planning level, process-oriented
operation planning and 3D-CAD, simulation
software, has been investigated to shorten and
simplify path planning on a robot-oriented
abstraction level, and processes on a task-oriented
abstraction level. By using generic robot
programming in the execution of the process at hand,
in combination with embedded system controller
over standard TCP/IP networks, the sequence of
operation on the factory floor with robots and
operators, can be quickly implemented. This shortens
lead-time in ramp-up, and simplifies changes and
changeovers throughout the manufacturing processes.
On the hardware level, production modules have
been integrated into mobile platforms. Equipment
that will be easily mounted and reconfigured on
standardized rigid base plates with flexible fixation
points have been investigated. This constitutes the
Factory-in-a-Box containers that can easily be moved
by a truck and/or forklift.
Existing experience from partners in the FlexAA
project has been used, using 6DOF flexible tooling
modules that do not require conventional calibration.
These tools have been investigated and tested in the
Factory-in-a-Box project. This approach is based on
building assembly fixtures that are configured and
reconfigured by the robot that also perform the
assembly- and lightweight machining operation.
4.3 CONTRIBUTION FROM THE MÄLARDALEN
UNIVERSITY RESEARCH GROUP OF THE
“EXACT” PROJECT
Experience is one of the most valuable assets in the
manufacturing industry, both during design and
development of products and manufacturing systems.
During continuous development and improvement of
production methods, technology and techniques to
achieve quality and production capacity together with
cost reductions experience is important. A key issue
for the Factory-in-a-Box project is to achieve
flexible, efficient and quick configuration and
production ramp up, which is easy to maintain and
adjust if requirements change on product or if input
parts specification change. Experience is essential,
and if all experience related to a box is stored and can
be reused for similar production tasks. The value of
the box increases by offering greater flexibility,
faster and more reliable configuration and
production. Valuable experiences to reuse are:
designs and configurations, fault reports, diagnoses
and solutions, quality improvements, cost reductions,
production increase actions, and software
components used in producing a specific product.
Also subparts of the production process are seen as
experience components. If similar products are
produced, the “programming” of the operations may
become the task of composing previous successfully
sub-parts into a complete program. Experience and
knowledge is spatial and only relevant in specific
contexts. A designer designing a specific part of the
production process will have access to relevant
experience and previous solution to the same or
similar problems that may be reused after some
manual or automatic adaptation. Parts of the software
components also have “learning” capability, which
will improve their performance. Finally, experience
collected in other Factory-in-a-Box’es can be reused
by any Factory-in-a-Box (standards and remote
facilities will be a part of the Factory-in-a-Box
concept).
All properties are parts and subprojects of the ExAct
project, developed in more generic form for
production industry. This is a complex reality, where
many compromises have to be made. Some of the
projects have to deal extensively with how to
integrate with, and change, a large and often
inflexible production process in order to preserve the
valuable features of the subprojects. In the Factory-
in-a-Box, the context and environment is controllable
and limited. The resources will be used to produce a
number of fully functional Factory-in-a-Box systems.
This will convince industry of the advantage and
motivate them to invest in modifications, enabling
the vision of the ExAct project (flexible, learning and
experience sharing semiautomatic and automatic
tools and systems continuously improving
production, quality, and cost reductions).
4.4 CONTRIBUTION FROM WINGQUIST
LABORATORY/FRAUNHOFER CHALMERS
CENTRE AND THE “ROBUST DESIGN &
VARIATION SIMULATION” PROJECT
Robust design and variation simulation are important
quality assurance activities in the product realization
process. Geometrical variation, originating from
individual manufacturing and assembly processes,
often propagates and accumulates during production,
resulting in non-nominal products and production
equipment. Geometrical quality problems are often
discovered during pre-production or when the
product is getting ready for market introduction. A
change in the product or production concept at this
stage often results in huge costs for product and/or
production changes, market delays, and bad
publicity. Therefore, more and more efforts are made
in early concept phases to virtually verify product
and production concepts with respect to geometrical
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
4
variation. These efforts can be used as a natural base
for non-nominal path planning to meet future
demands on accuracy and throughput.
A long-term vision within the field of production
technology is the virtual factory, with high level of
accuracy regarding realism and functionality. Early
programming, simulation, verification, and
visualization of virtual production equipment make it
possible to reduce the ramp up time in the real
factory. Despite that modern industries use virtual
prototypes to replace physical prototypes, visualize
assembly processes and program industrial robots
off-line, the full potential of the virtual factory is still
not reached. A major limitation is programming time.
Most programming of motions and paths for robots
and equipment is still generated manually, since the
existing support for automatic path planning and load
balancing is very limited. Another limitation is the
geometrical accuracy between the virtual model and
the physical reality. Today, all virtual models used
for robot and assembly simulation and verification
are nominal. However, in the real world, all
equipment, parts and subassemblies are inflicted by
geometrical variation, often resulting in conflicts and
on-line adjustments of off-line generated paths.
Increased demands on ramp-up time, flexibility and
plant throughput require robot paths to be generated
and verified with respect to both robot precision and
assembly tolerances. Therefore, the goal is to create
knowledge, simulation tools and working procedures
that will enable non-nominal path planning for rigid
bodies and industrial robots. The project integrates
two basic disciplines that have never been combined
before, variation simulation and automatic path
planning. Non-nominal path simulation and
verification in concept phases reduces cycle time as
well as need for physical verification meets demand
on flexibility and speed by avoiding unnecessary
tight tolerance, design changes and on-line
adjustment of robot programs. Automatic non-
nominal path-planning supports the Factory-in-a-Box
vision of mobility and reconfigurability by allowing
for fast robot program updates on site.
5. FACTORY-IN-A-BOX DEMONSTRATORS
The key features to realize a Factory-in-a-Box are
flexibility, mobility, and speed. The concept consists
of standardized modules that can be installed in
containers and easily transported by e.g. trucks, rail
vehicles, boats etc. The modules shall be easily
combined into complete production systems and
reconfigured for new products and/or scaled to
handle new volumes. The goal of this project is to
build five fully operative demonstrators – Factory-in-
a-Box production cells– that are developed in close
cooperation between different academia’s and
industrial partners. The goal with the demonstrators
is to exemplify and realize the Factory-in-a-Box
concept; they will be practical examples of the
usability of the concept in industry. All
demonstrators are a practical solution for a particular
function(s) and provide a real business case for the
concept. The five different demonstrators are
described in the paper.
5.1 DEMONSTRATOR 1 – AUTOMATIC
ASSEMBLY WITH FOCUS ON FLEXIBILITY
A first example of a Factory-in-a-Box module has
been developed and demonstrated within ABB
Robotics’ production system – an automatic
production module to assemble robot components.
The overall goal of this pilot demonstrator was to
develop an automatic production module, which
assembles robot controller cabinets, meeting the
overall Factory-in-a-Box requirements of flexibility,
speed, and mobility. The demonstrator has been
developed in parallel with an ongoing product
development project of a new robot controller at
ABB Robotics: “IRC5”.
• Flexibility: In order to assemble different
variants of cabinets with short set-up time, it
is necessary to have flexible equipment and
fixtures. There will be a need for
reconfiguring the module and its components
while still having robust and efficient
manufacturing.
• Speed: Short set-up time is vital for the
success of this module. The Factory-in-a-Box
module will enable a structured production
requirement process and a support for design
of future cabinet variants. A “standard”
Factory-in-a-Box module will also enable
virtual system configuration and module
modelling and simulation.
• Mobility: The Factory-in-a-Box module will
have to be designed as a “Mobile Platform”
to be moved anywhere within ABB Robotics
production system. Possibly also moved to a
supplier or another production site in Bryne,
Norway.
The vision of this demonstrator has been a
production system that can be assembled and
configured according to the company needs and that
can be delivered to any location. Demonstrator 1 has
explored this vision and tried to make this a reality.
The focus of this demonstrator has been to
investigate the following requirements that were
specified in the original project plan;
• Modules that are easy to transport to the
production site as well as to move them at the
site, e.g. external and internal mobility. In
order to attain internal mobility i.e. air
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
5
cushions can be used for fast and smooth
transportation of modules or for the entire
Demonstrator.
• Reconfigurability in order to meet changing
demands and automatic/semi-automatic
configuration of modules and system are
prerequisites for scalability for changing
production volumes and for fast ramp up of
production.
• Reusability of system components and
modules together with simple and fast
simulation and programming makes
conditions for faster and cheaper system
solutions and system robustness towards
disturbances, especially during ramp-up of
production. The reusability also makes it
possible to achieve a profitable reduction of
production capacity. This is as important as
the ability to increase the capacity. The
reusability makes it possible to reuse the
equipment in other applications in the same
or in other companies.
• Standardized hard- and software interfaces
and integrated highly flexible production
equipment, integrated metrology, and
sensor-based calibration, combined with
sensor-integrated robot/equipment control
are prerequisites for flexibility/agility.
The Factory-in-a-Box module at ABB Robotics
consists of two robots, a number of fixtures, a gluing
station, a folding station, and robot handled tools.
Material and components are presented to the robots
via conveyors and carriers placed in a lock-system.
The cell is designed for mobility using mobile base
plates that either are heavy enough to be placed
directly on the floor or, as in the case of the robot, a
platform which can be secured to the floor by
vacuum or air cushions for fast and smooth
transportation. Figure 1 shows an illustration made to
show that each piece of equipment is thought of as
being placed on a piece of a puzzle; the different
parts are mobile, with well defined interfaces
between them. Figure 2 shows one of the robots and
one fixture mounted on platforms.
Fig. 1. Illustration of the concept of
Demontstrator 1; mobile prodution equipment
with well defined interfaces
Fig. 2. Picture of one of the mobile fixtures and one
robot on a platform
During the Demonstrator 1 project, software for cell
programming and control was developed (Hedelind
and Hellström, 2007). This software was developed
in order to enable the hardware reconfiguration that
was needed in the resulting robotic working cell.
The cell uses flexible equipment, and is designed for
reconfigurability. Four different types of controllers
are today assembled in the cell. Through technical
solutions, supporting mobility and flexibility, the
requirements of speed are achieved by e.g. quickly
reconfiguring the cell at another location and /or
introducing a new product. The I/O communication
in the cell is established through wireless
communication pads to enable easy reconfiguration.
The two robots are equipped with swivels and tool-
changers so that they may handle all product variants,
and also to make it easy to introduce future product
variants that require other tools for handling.
Factory-in-a-Box module at ABB Robotics has been
commercially developed and put into operation since
December 2006.
5.2 FACTORY-IN-A-BOX 2 – WELDING WITH
FOCUS ON MOBILITY
Demonstrator 2 is developed in collaboration with
Pharmadule Emtunga (PHEM). PHEM is a supplier
of modular facilities to the offshore, telecom, and
pharmaceutical industries. At present the company is
striving to implement the same concept in their
manufacturing system as they have in their products,
i.e. modularisation.
Demonstrator 2 is a semi-automated manufacturing
cell, which is used for cutting, bevelling, and welding
of carbon steel pipes. All machinery will be fitted
into a standard container, which also will contain,
fume hood exhaust, lighting, computer terminal etc.
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
6
Factory-in-a-Box 2 has approached the key concepts
flexibility, mobility, and speed as follows:
• Flexibility - In order to weld different
variants of pipes as well as conducting
different joint preparations with short set-up
time, it is necessary to have flexible
equipment and fixtures. There will be a need
for reconfiguring the module and resources,
e.g. configuring the demonstrator online and
still having a robust and efficient
manufacturing. The optimal level of
automation will be investigated in the
project. Different machining operations in
the module are drilling, joint preparation,
and cutting. Pipes will have different
dimensions in terms of diameter and
thickness.
• Mobility - The Factory-in-a-Box module
will have to be designed on a “Mobile
Platform” to be moved anywhere within the
production system, to a supplier, or to site.
The equipment should easily be mounted
and reconfigured on standardized rigid base
plates with flexible fixation points. The
Factory-in-a-Box module should be
moveable by a truck – specifying the need
for a standardized container.
• Speed - There is a need to quickly
reconfigure the module without a long set-
up time. Operations should be started as
soon as possible after movement within the
manufacturing system or transport to site.
The Factory-in-a-Box module will reduce
the welding time compared with today’s
manual process. With increased automation,
quality will be improved and more
consistent, and the disturbance time will
decrease.
Demonstrator 2 can perform the following
operations;
• Cutting and seam preparation of pipes
• Manual welding when fitting the pipes
together
• Orbital welding
• Flexible fixture to handle the different pipes,
as shown in figure 3
One of the major challenges with Demonstrator 2
is the usage of known technologies in a new
application and context. Orbital welding of
carbon steel pipes with straight-angle chamfers
has never been done before within the company.
Fig. 3. Illustration of the flexible fixtures developed
in demonstrator 2
PHEM has had a business concept in mind when
developing the Demonstrator 2. This kind of mobile
orbital welding equipment could make a great asset
for a lot of different construction and installation
sites where carbon steel pipes are being welded. The
Factory-in-a-Box module can be leased out to PHEM
by suppliers or be owned by PHEM themselves. The
Demonstrator 2 will be implemented within PHEM
operations during the fall of 2007.
5.3 FACTORY-IN-A-BOX 3 – FOUNDRY WITH
FOCUS ON MOBILITY
To further demonstrate the applicability of the
Factory-in-a-Box project to industry, a third example
of a Factory-in-a-Box module will be demonstrated
within the Swedish foundry industry in cooperation
with the Swedish Foundry Association. To maintain
its competitiveness a foundry must meet increasing
demands for efficient production and a good working
environment.
Varnäsföretagen AB is a Swedish foundry that
produces sand-casted aluminium goods. The
company has modern and highly operational facilities
for both casting and subsequent machining.
However, the process step between the two,
deburring and grinding has been neglected, as it is in
many other foundries. This middle step between the
casting and machining is important, but often
performed manually using handheld tools. This type
of work has a tradition of being ergonomically
unsuitable and generally inefficient, thus motivating
Varnäsföretagen to take part in the Factory-in-a-Box
project. The aim of the third demonstrator is to create
an automatic solution for deburring of casted
products.
Demonstrator 3 has approached the key concepts
flexibility, mobility, and speed as follows:
• Flexibility - In order to handle different
components with short set-up time, it is
necessary to have flexible equipment and
fixtures. There will be a need for
reconfiguring the module and resources, e.g.
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
7
configuring the robot online and still having
a robust and efficient manufacturing.
• Mobility - Demonstrator 3 will have to be
designed on a “Mobile Platform” to be
moved anywhere within the production
system, or to another foundry company. The
equipment could be encapsulated into a
container to enable easy transport, and
containment during production.
Demonstrator 3 should be moveable by a
truck.
• Speed - There is a need to quickly
reconfigure the module without long set-up
time. The Factory-in-a-Box module will
reduce the handling and material removal
time compared with today’s manual process.
With the right level of automation, quality
will be improved and the disturbance time
will decrease.
During 2006 and 2007 a feasibility-study has been
performed at the company, which included a
mapping of the production process at the company
and a study of the products that are being produced.
After that the results of the study were analyzed it
was concluded that not all the operations could be
performed within one automatic solution. The ideal
solution would be to use several small and cheap
standardized automation cells that could perform a
couple of operations each. This project will focus on
developing one cell that can perform a set of
operations that can be used for some of the deburring
work. This first cell could for example be used to saw
off larger pieces from the casting followed by some
milling and grinding operations. These operations are
today a burden for the persons doing the manual
work.
The final concept that has been developed includes a
robot mounted inside a container. The robot is placed
on a flexible steal-beam system that can be moved in
and out of the container. A flexible fixture has been
proposed that should be able to handle all the
products that are supposed to be machined in the cell.
The concept has been partly realized in a lab
environment and simulated fully in a virtual
environment, and whether it will be converted into an
operational cell is dependent on the company and
circumstances outside the research projects control.
5.4 FACTORY-IN-A-BOX 4 – FUNCTIONAL
SALES WITH FOCUS ON FLEXIBILITY
Demonstrator 4 has been developed in association
with FlexLink Systems. FlexLink’s focus is
automation of production flow within the following
processes: Assembly - Filling - Machining -
Packaging. FlexLink will, in this project, use their
Dynamic Assembly System (DAS) concept in order
to demonstrate the principles in the Factory-in-Box-
project - flexibility, mobility, and speed - in a real
customer case.
Demonstrator 4 has approached the key concepts
flexibility, mobility, and speed as follows:
• Flexibility - Different variants of products
with short set-up time require flexible
equipment and fixtures. There will be a need
for reconfiguring the module and resources
online to a customer specific product while
still having a robust and efficient
manufacturing. The optimal level of
automation will be investigated in the
project.
• Mobility - The Factory-in-a-Box module
will have to be designed as a reconfigurable
“Mobile Platform” to be moved anywhere
and reused for a new customer in a case of
leasing.
• Speed - Short set-up time is vital for the
success. The “programming” should be fast,
which may demand reuse of experience or
knowledge sharing.
During 2005 and 2006, a pre-study at the company
has been performed which has included an
investigation of “functional sales” and a mapping of
interested companies for this commercial solution. A
number of companies have been contacted (over 40
different companies) and possible candidates for a
FlexLink and Factory-in-a-Box commercial
application where identified.
After identifying one company that was interested,
the project has participated in a quoting phase. No
order has been placed so far. The Demonstrator 4
project has investigated the concept of functional
sales of manufacturing capacity. The project has
shown that industry is interested in this concept, but
the circumstances and details in the offering may
have to be changed. This type of functional sales are
already being used in other types of industries, but is
still not a common solution within manufacturing
industries. No “real” example has been generated so
far, which was the initial objective.
5.5 FACTORY-IN-A-BOX 5 – MANUAL
ASSEMBLY WITH FOCUS ON MOBILITY
This part of the project is conducted in cooperation
with Bombardier Transportation. Bombardier
Transportation in Västerås is facing a market where
the customers are becoming more and more powerful
due to the strong global competition. Many
customers have strong wishes that part of the
production should be carried out locally in order to
create new jobs. Instead of building factories, which
will be abandoned as soon as the order is processed,
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
8
Bombardier and the Factory-in-a-Box project aim to
develop mobile production facilities, which can be
re-located as soon as the production of an order is
finished. This can lead to winning orders on markets
otherwise closed, which will create job opportunities
not only for foreign countries, but also for
Bombardier in Sweden since the main part of the
order is produced here.
Factory-in-a-Box 5 has approached the key concepts
flexibility, mobility, and speed as follows:
• Flexibility - Different variants of products
with short set-up time require flexible
equipment and fixtures. There will be a need
of reconfiguring the module and resources to
a customer specific product while still
having a robust and efficient manufacturing.
• Mobility - The Factory-in-a-Box module
will have to be designed as a “Mobile
Platform” to be moved anywhere and reused
for a new customer and a new project.
• Speed - Short set-up time is vital for the
success as well as readiness for
transportation
An analysis has shown that a Factory-in-a-Box,
mobile production capacity can be used to reallocate
working opportunities in a cost-effective way when
the customers demand this as a part of the business
deal. The concept contains four parts: a technical
solution, a logistics solution, a training solution for
the local labour, and a methodology for how to move,
install, and put the Factory-in-a-Box into production.
A fully developed Factory-in-a-Box enables a
substantial reduction of the resources needed for
sharing experience and knowledge compared with a
conventional outsourcing strategy. The feasibility-
study has resulted in two technical solutions for
moving and housing production capacity abroad: a
special container and a modular building solution. An
illustration of a special expandable container solution
is shown in figure 4. The choice of final solution will
be decided upon including a prognosis on usage
frequency, production capacity needed, and the
strategically value this solution gives Bombardier.
Fig. 4. Illustration of the mobile workstation concept
of Demonstrator 5.
In order to exemplify a valid solution with a cost-
analysis, the project has focused on a technical
solution for the assembly of High-Voltage boxes
demanding a capacity of 1-2 boxes a week. Even
using this delimitation, the feasibility-study indicates
that the Factory-in-a-Box easily can be modified to
handle a much higher production volume. It also
indicates that the same kind of solution can be useful
for other Bombardier products. The feasibility-study
has also resulted in a checklist that the person
responsible for the installation of the Factory-in-a-
Box can use in order to shorten the start-up time at
the customer. The feasibility study ended in a
demonstration of a container solution that was
temporarily installed at Bombardier in Västerås, and
the concept is used in different offered solutions to
customers.
Besides being used as an enabler to win orders, the
feasibility-study has shown that the Factory-in-a-Box
concept provides a substantial potential for cost-cut
in the ongoing production at Bombardier. Thus, the
feasibility-study recommends that the development
of a prototype should be started at the company.
6. SUMMARY AND CONCLUSION
This paper has presented the Factory-in-a-Box
project, the four different underlying research
projects that contribute to the project, and the five
demonstrators that have been developed in order to
demonstrate the Factory-in-a-Box concept. The four
underlying research projects, as presented in section
4, do all contribute with their own specific research
in order to realise the Factory-in-a-Box concept. The
five demonstrators that have been developed together
with the industrial partners are all examples of how
this concept may create “mobile production capacity
in demand” – as the project set out to do.
Demonstrator 1 was developed during two years and
is now integrated into production within a real
manufacturing system. Demonstrator 2 has been
developed in a virtual setting and evaluated, and the
industrial partner will develop the physical
demonstrator during the fall of 2007. Demonstrator 3
was fully developed in a virtual environment, and
partly developed in a laboratory setting. The future of
Demonstrator 3 lies in the hands of the industrial
partner, which may decide to develop and use the
concept at a later time. Demonstrator 4 was an
attempt at leasing or selling production capacity. The
concept has been offered to several customers, and
time will tell if a real investment will be made.
Demonstrator 5 was realised in a temporarily setup,
and a lot of theory around how to realise mobile
production facilities was developed within that
demonstrator project. The concept of Demonstrator 5
has been offered to customers of the industrial
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
9
partner and whether any real implementation will be
made is left to see.
The scientific contribution of the Factory-in-a-Box
concept is primarily the attempt to exemplify and
implement a flexible and reconfigurable production
system in reality. In the Factory-in-a-Box project we
have indicated that it is possible to develop
conceptual as well as operational manufacturing cells
that meet the requirements mobility, flexibility, and
fast set-up and ramp-up of production. Implementing
a Factory-in-a-Box is proposed as an innovative
solution and a means to exploit new competitive
factors enabling future competitiveness within
manufacturing industry.
The concept of mobile production capacity has been
evaluated together with the concept of functional
sales. The fourth demonstrator project had functional
sales as the main priority, and an extensive feasibility
study and a large number of offerings to customers
have been conducted. However, the results from the
study are inconclusive, and to fully evaluate the
concept more cases will be needed.
Among the different competitive factors that are of
importance in the Factory-in-a-Box project, mobility
is perhaps the key-enabler. Mobility can be used as a
competitive advantage in several different ways:
reconfiguration of a manufacturing system, to
provide production capacity where needed – when
needed, and enabling companies to make an
investment in one piece of equipment and reuse it at
several production sites or within several projects.
ACKNOWLEDGEMENTS
Factory-in-a-Box is a joint project between 4
universities and more than 6 multinational and
national companies, (www.factoryinabox.se, in
Swedish), the project is funded by participating
industrial partners and the Swedish Foundation for
Strategic Research (http://www.stratresearch.se)
through the ProViking programme
(http://www.proviking.se).
REFERENCES
Ask, A. (2006) Factory-in-a-Box – An enabler for
Flexibility in Manufacturing Systems.
Mälardalens högskola, Licentiate thesis,
Mälardalen University Press
Hedelind, M., Hellström, E. (2007). Programming of
reconfigurable robotic working cells. To be
published in the proceedings of The 4th
International Federation of Automatic Control
Conference on Management and Control of
Production and Logistics, Sibiu, Romania, 27-30
September
Jackson, M. (2000). An Analysis of Flexible and
Reconfigurable Production Systems – An
Approach to a Holistic Method for the
Development of Flexibility and
Reconfigurability. Linköping, Studies in Science
and Technology. Dissertation No. 640,
Linköping, Sweden
Jackson M, Zaman A (2006) Factory-in-a-Box –
Mobile Production Capacity on Demand.
Proceedings of the 2006 IJME – INTERTECH
Conference October 12-21 New York City USA
Molinari, T. L., F. Pierpaoli, A. Urbani, and F.
Jovane, (2004). New Frontiers for
Manufacturing in Mass Customization. CIRP
Journal of Manufacturing Systems, Vol. 33, No.
3
Piller, F.T. (2002). Mass customization: four
approaches to deliver customized products and
services with mass production efficiency.
Engineering Management Conference, IEEE
International, Vol.2
Tidd, J., J. Bessant, K. Pavitt (2001). Managing
Innovation. Second Edition, John Wiley & Sons,
Ltd, England
Yin, R.K. (1997). Case Study Research, SAGE
Publications Inc., London
Yusuf, Y.Y., M. Sarhadi, A. Gunasekaran (1999).
Agile manufacturing: The drivers, concepts and
attributes. International Journal of Production
Economies Vol. 62
Copyrights belong to the authors of this paper
Swedish Production Symposium 2007
Hedelind...
10