Intelligent and Flexible Worker Assistance Systems

Assembly Assistance Platform for Planning Assisted Assembly and Rework

as Well as Execution of a Worker-Centered Assistance

Rainer Mueller, Matthias Vette-Steinkamp, Leenhard Hoerauf, Christoph Speicher

and Attique Bashir

Center of Mechatronics and Automation Technology, Eschberger Weg 46, 66121 Saarbruecken, Germany

Keywords: Worker Assistance, Manual Assembly, Rework Station, Industry 4.0.

Abstract: In assembly, reworking stations are often barely automated work places, where no technical equipment

supports the worker, validates the process execution or documents process results. However, worker

assistance systems are used to guide and support the worker through manual assembly. Nevertheless, the

concept of providing information to the worker is limited to screen based output media and assistance systems

are just compatible to supplier selected devices. The objective of the presented research and development in

this paper is an interactive worker assistance system which combines qualification-based worker support and

intelligent process validation especially for rework stations. Beside the worker assistance system, a planning

environment is developed for an efficient generation of work plans and content for the worker assistance

system.

1 INTRODUCTION

Nowadays, manufacturers are confronted with many

challenges such as market demands for high quality

products and decreasing prices. At the same time, the

demand for individualized products raises which

results in an increasing product variety and in smaller

batch sizes at the manufacturer’s side. To satisfy these

customer demands, manufacturer have to find the

right way to combine the flexibility of manual

processes and the reliability of automated systems.

(Mueller, 2016a)

In most industrialized countries, the described

challenges paired with the demographic development

intensifies the situation and the need for supporting

systems in manual assembly. According to (Weidner

et al., 2015), supporting systems can be distinguished

between technical systems that substitute humans and

systems that support humans. Furthermore, they state

that a technical system can be considered as

supporting system if it fulfills the following points:

1. The technical system supports the worker with his

tasks, without replacing him

2. The technical system can always be overruled by

the worker

3. The worker operates the technical system

4. The system does not pose any danger for the

worker

Beyond that supporting systems consider physical,

psychological, communicative or organizational

improvement. An assistance system is a supporting

system that can be distinguished in systems for

physical or informational support (Hinrichsen et al.,

2016).

In final assembly lines, these supporting systems

are frequently implemented to guide the worker

through the assembly process. Whereas in rework

areas, experienced workers do the rework without the

help of an assistance system because standard

assistance systems do not possess the ability to handle

dynamic failure conditions. Furthermore, these

systems obstruct the worker during his work because

they do not consider the experience level of the

worker. Due to the minor application of assistance

systems at rework area, failures occurring during

manual rework process aren’t detected in time and

cause expensive reworks again.

This paper presents the development of an

assembly assistance platform. The platform’s purpose

is planning assisted processes through use of the

worker assistance system, guiding the worker through

the job and rework process validation in industrial

production. The platform consists of a planning

Mueller, R., Vette-Steinkamp, M., Hoerauf, L., Speicher, C. and Bashir, A.

Intelligent and Flexible Worker Assistance Systems - Assembly Assistance Platform for Planning Assisted Assembly and Rework as Well as Execution of a Worker-Centered Assistance.

DOI: 10.5220/0006613900770085

In Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2018) - Volume 2: HUCAPP, pages

77-85

ISBN: 978-989-758-288-2

environment to design dynamic process flows and a

control system. The control system connects planning

environment, worker assistance system and

intelligent resources.

The ongoing research and development of the

assistance platform is presented in following

chapters: Chapter 2 gives an overview of the state of

the art. Chapter 3 de1scribes the current situation and

requirements for worker assistance in the rework

station of an automotive supplier. In Chapter 4 the

developed solution is presented, beginning with a

short analysis of current worker assistance systems

which are found in manufacturing companies

nowadays. Chapter 5 summarizes the results of the

research work and gives an outlook towards future

work.

2 STATE OF THE ART

To support the worker cognitively during manual

processes, conventional media such as paper, tables

and other drawings are widely used (Wiesbeck,

2014). However, these traditional concepts have the

disadvantage that employees have to look at different

objects and search for the correct information by

flicking through the instructions. This is time

consuming (Lusic and Fischer, 2016) and hinders the

parallel handling of the product. The information

supply is not efficient by using this unsuitable display

techniques and can lead to assembly errors e.g. due to

misinterpretation. These manuals are time-invariant,

and therefore cannot represent information in

temporal sequences directly. Hence, they burdens

more cognitive load to the worker (Lusic and Fischer,

2015). Furthermore, manuals have to be updated

contently and physically if they are available as

printouts because they do not have any direct link to

a digital planning environment. Process instructions

written down on paper are more likely to be ignored,

due to the extra handling effort.

Therefore, analogue and conventional

information carriers have to be substituted by digital

media and to provide the information in the most

ergonomic way to the worker. The information

representation is just one part of the assistance. It

avoids failures by providing the worker with

information for the right execution of the process.

However, the second part is to check if a supposed

completed process is executed as specified. Thus,

assistance systems are equipped with sensors to

evaluate the process. For example, camera

technology is widely used to compare target state and

actual state of a process.

Currently, there are several worker assistance

systems available on the market. Looking at them

reveals different stages of assistance. The main

component of assistance systems is their information-

displaying character. Moreover, they often have an

evaluating component like cameras to detect

differences between actual status and target status of

the process or rather of the product.

A closer look at the available cognitive assembly

assistance systems reveals that these systems consist

mostly of two parts. The first part is the planning

environment, whereas the second part is used to

support the worker during the processes. In the

planning environment, the user or a planner defines

the processes and parameters which are used during

the process later. The following characteristics are

found in most worker assistance systems:

▪ Listing of processes to be executed

▪ Pointing to the current process (to be finished)

▪ Illustration of the process

▪ Manual quitting of the process

These systems are often restricted to the provider’s

specific resources. Even though they integrate several

systems like pick-by-light, the amount of available

resources integrated in the assistance system is

limited. Moreover, most common worker assistance

systems are limited to a worker guidance which

appears in the form of a monitor with a job list and

further human readable information like figures.

Sometimes, there is only one sensor like a push button

for quitting the process. Planning environments are

used to configure the assistance systems, but are

limited to integrate further resources. Because of this,

processes with other resource requirements cannot be

listed on the job list. E.g. camera based systems

cannot validate the torque applied in a tightening

process. Especially in the manual assembly or rework

station where a lot of complicated and different kinds

of processes have to be validated, the limitation of

hedging technologies can be seen as insufficient.

Most environments generate process lists by defining

them step by step. These static job lists are unsuitable

for a huge amount of variants as well as error code

dependent rework processes. An error code

corresponds to a defective behavior of the product

and allows to determine the reason of the failure. It is

provided by a test bench. However, changing the

rework job is quite common because another failure

reasons can be detected by the worker during the

process execution.

Therefore, the system must pose the functionality

to change the rework job dynamically on an error

code, which is not provided by common systems.

HUCAPP 2018 - International Conference on Human Computer Interaction Theory and Applications

3 ANALYSIS OF THE INITIAL

SITUATION

As the assembly assistance platform is developed for

the application in real production environment, the

conditions and initial situation are analyzed

beforehand. The case is the assisted and ensured

rework on faulty automatic transmissions after the

final assembly and commissioning in production. As

an automotive supplier, the transmission

manufacturer faces high quality demands and is

obligated to secure and document its processes. Due

to the operational structure and economic reasons,

this is achieved in the final assembly but not in the

rework area.

An automatic transmission is made of a large

quantity of parts which are assembled through

complex processes. In the final assembly, (semi-

)automated assembly lines are installed which

provide a high degree of process validation as well as

an extensive process documentation. The process

validation results in high quality manufactured

products. To validate the processes, the assembly

lines are equipped with sensors and electronic

resources that ensure the correct assembly process

execution. In a testing bench, additional tests are

performed to validate the quality and functionality.

Faulty transmissions are sent together with an error

code to the rework stations in order to get repaired

and brought in a salable state.

In contrast to the final assembly lines, rework

stations are structured as single workplaces and are

dominated by a high degree of manual assembly

processes. This allows a required degree of flexibility

to repair all kinds of product variants and perform

different rework jobs and processes at one workplace.

But compared to the final assembly lines, the process

validation and documentation in the rework station

are insufficient, due to the lack of automated and

technical equipment. Hence, highly skilled workers

are needed which additionally undergo a special

training program to be able to handle the rework job.

Especially the lack of process validation can lead to

mistakes during the process. In the final assembly

line, the process validation ensures the correct and

scheduled performance of a process as well as process

documentation, which is necessary for traceability

purposes to customers.

Since rework jobs require highly qualified

workers with many years of experience, in case of

sudden absence personal replacement cannot be

found easily. Furthermore, experienced worker

undergo a capacity intensive training to be qualified

to work at the rework area. Thus, one objective of the

proposed assembly assistance platform is to reach a

similar degree of process validation as in final

assembly lines, but maintain the workers’ adaptivity

and ability to make situation based decisions. Another

objective is to cover worker qualification and training

on the job. Therefore, the assembly assistance

platform consists of four important modules:

▪ Assistance system for guiding and qualifying

workers

▪ Intelligent resources to validate and document the

processes

▪ A control system that distributes planning data to

resources and assistance system and connects all

systems with each other

▪ A planning environment for creating dynamic and

parametrized processes lists

The worker assistance system has to guide the worker

through the rework job and support him to avoid any

mistakes.

Following, the user profile is described. The

developed system addresses women and men alike

and there is no difference in the information/ content

provided by the system or how the content is

displayed. The user’s age varies from early 20s to 60.

Therefore, IT-affine and non-affine people alike are

confronted with the system in their daily work. To

address both types of users, users will be qualified

and trained for working with the system.

Furthermore, the user is involved in the design phase

of the system and has the opportunity to give

feedback concerning the design of the assistance

system. As mentioned before, the average user is a

skilled worker with at least three years of work

experience in a similar production department, where

similar but often “easier” tasks have to be performed.

Hence, the user is familiar with the basic

functionalities of the product (automatic

transmission). Because a lot of product variants and

different process have to be mastered, three skill

levels (beginner, advanced and expert) have to be

considered. Beginner starting in the rework area are

supported by experts and trained for about half a year.

He needs a higher grade of assistance e.g. of difficult

processes and resource because he is still in learning

phase and has to be trained. Users exceeding the

training phase are of course more experienced. On

this level, they don’t need small stepped assistance

and further explanation of processes, but additional

hints or information e.g. during a bolt tightening

process considering the tightening sequence or during

the rework on a variant, which they don’t handle /

repair very often. Experts on the other hand don’t

need any hints. On this level, assistance is seldomly

needed and detailed information is considered

Intelligent and Flexible Worker Assistance Systems - Assembly Assistance Platform for Planning Assisted Assembly and Rework as Well as

Execution of a Worker-Centered Assistance

disturbing rather than helpful. Therefore, the

assistance system will provide only information about

major repair steps and most importantly, give

feedback to successful or faulty processes during

repair. This last point was addressed by beginners and

experts alike. To close the loop between human action

and system behavior or rather the performed task a

feedback from the assistance system is given. The

feedback informs the user if a process is quit as

planned or if something needs to be considered there.

It needs to be considered that workers considered

experts don’t have more decision rights. The main

difference is the degree of assistance, which is needed

to complete the task.

The rework job is a set of processes which

consists of three stages: the disassembling, the

repairing process and the assembling of the product.

Because of many different error codes in combination

with a high amount of product variants, the worker

assistance system has to generate the process and job

list dynamically. Since the relationship between error

code and job list is not unique, the worker should have

the opportunity to take another process and job list, if

he detects another reason for the error aside from the

given job list code. Therefore, the assembly

assistance platform has to allow a dynamic and

resource independent generation of process

sequences and an easy integration of resources.

Since the assembly assistance platform is used at

the rework area, there are four requirements

formulated which are similar to (Aehnelt and Urban,

2015):

1. Information

2. Parametrization of resource and process execution

3. Monitoring and checking

4. Documentation

Information considers the ability of a system to

provide the user with human readable data. A

visualization complements the pure textual

information as well as other forms of visualization by

showing how and where to handle parts.

Parametrization and process execution concerns

the data for smart resources. E.g. electronic resources

can be configured through parametrization.

Electronic bolt tightening tools often have the

capability to drive a predefined bolt tightening curve

and tighten bolts with certain parameters.

Continuous monitoring and checking of the

process can be validated. As long as the current

process is not quitted the worker knows that the

process is not completed as planned. In case of a

system failure, most assistance systems allow the

worker to quit a process manually if the worker is sure

that a process is finished correctly.

Documentation is mostly the last action of a process.

If the process is finished certain process parameters

have to be stored due to traceability purposes.

Every requirement listed above is fulfilled by a

resource of the assembly assistance platform. E.g. the

first requirement of the assembly assistance system is

to provide the worker with information. The resource

which can be used for that is a monitor. The second

requirement is to execute the process. For this, a smart

resource can be used. So for every requirement listed

above a resource can be used to fulfill it.

4 DEVELOPMENT OF AN

ASSEMBLY ASSISTANCE

PLATFORM FOR THE

MANUAL ASSEMBLY

To use the assembly assistance platform in a modern

factory environment, the following requirements and

boundary conditions are considered:

1. The planning environment can use already

existing data e.g. from the product design

department

2. The planning environment is user friendly and can

be used easily without any programming

experience

3. The planning environment can handle multiple

product variants

4. The planning environment can store the plan data

in a suitable way so that a control system can use

these data

5. The control system can read error codes and

variant codes to generate a rework job list from an

existing overall process list dynamically

6. The worker assistance system allows to

dynamically expand the rework job if another

failure reason is assumed

7. Additional resources can be easily integrated into

the worker assistance system

8. The worker guidance offers reporting and

feedback possibilities

Beside the analysis of final assembly as well as

current state of rework area, the requirements

considering the product and processes are also

analyzed. Based on the analysis of the initial situation

and a performed Failure Mode and Effects Analysis

(FMEA), the critical product parts and processes

which have to be validated during rework by the

resources and functions of the assembly assistance

platform are documented. As a result, the following

requirements have to be considered and provided by

the assembly assistance platform:

HUCAPP 2018 - International Conference on Human Computer Interaction Theory and Applications

1. Checking if the right part is removed or assembled

as well as positioned (e.g. tolerances) correctly

2. Checking if a disassembled part is placed in a

defined component tray

3. Checking if a disassembled part is disposed as

anticipated

4. Checking if the correct part is picked and

assembled in the product

5. Checking if a process is performed as planned e.g.

if a bolt tightening operation is executed correctly

regarding tightening process, torque, angle and

sequence

With the understanding of product, process and

resources as well as the boundary conditions and

requirements, the concept for the assistance platform

is developed. The assembly assistance platform

consists of four different modules. In Figure 1, the

concept and framework of the platform with different

modules is shown and unites the planning

environment as well as the worker assistance system.

Besides, the human role can be seen. The role is

mainly to perform processes on the product. The

worker guidance is the main interface between human

and worker assistance system. It provides the user

with main information about the product as well as

the process but also allows to give feedback e.g. when

a process needs to be quit. Another way to provide

information to the user is by using the smart

resources. Whenever the resources are not supposed

to be used, they are disabled, which is a hint to the

user that he eventually missed a step before or didn’t

commit it correctly.

Figure 1: Worker assistance platform consisting of a

planning environment and worker assistance system.

In Figure 2, the current state of the worker

assistance system with implemented resources is

shown. As for now, it consists of a 2D camera, an IR-

Camera, a laser projector and a worker guidance

system. Further resources and functionalities to fit the

described requirements are planned.

Figure 2: Current state of the work station with laser

projector (1), 2D Camera (2), IR camera (3) and a worker

guidance system (4).

4.1 Planning Environment to

Configure Assisted Job Lists

To setup the worker assistance system, a planning

environment is needed. To reduce programming

effort, the planning environment has a GUI. Since the

execution of the processes leads to the completion of

the rework, the aim of the planning system is

primarily to generate a resource afflicted generic

process list (generic rework job list). The list allows

to repair any component of any product variant and

even helps handling the worst case scenario in which

all components need to be replaced (or repaired).

Therefore, the list contains all assemblies,

subassemblies and parts (hereinafter referred to as

product components) of the product’s variant group.

The approach to create that list is to begin with

analyzing the product, derive the processes and assign

the appropriate resources.

Initial point for the planning are the product data.

Since there are always product variants which are

similar to one another, a generic Bill of Material

(BOM) can be used, representing all product

assemblies, subassemblies and parts of the variant

group. The first plan data to be specified is the error

code. It has to be attached to the products assemblies,

Intelligent and Flexible Worker Assistance Systems - Assembly Assistance Platform for Planning Assisted Assembly and Rework as Well as

Execution of a Worker-Centered Assistance

subassemblies or parts. The error code determines

which product part needs to be handled, repaired or

replaced and allows the control system to calculate

the disassembly path. The product data and

requirement which determine how the product is

assembled can be received from other department,

like the development department.

Knowing the product requirements, in the next

step all disassembly as well as assembly processes are

derived and linked to product components. Thus, for

each component a corresponding process exists and a

generic process structure (similar to the BOM

structure) is created. Since the planning data are used

later to control the worker assistance system, process

data needs to be enriched with resource configuring

data, which are human readable as well as machine

readable. Machine readable data are meant to be used

as input data for resources or contain information to

receive data from the resource to validate the process.

For example, process parameters such as bolt-

tightening-torque, bolt-tightening angle, article

number, guidance information, etc. are some of these

parameters.

The next step after creating the generic process

structure is to create variant specific process

structures by loading variant specific BOMs. In case

of matching, the according process steps are kept

whereas not matching process steps are removed.

Process parameters have to be received from other

departments or derived from the product structure. By

known process parameters, the last step is to define

variant specific resources to execute the processes.

The job list in the planning environment contains

the disassembly as well as the assembly. In Figure 3,

an overall rework job list is shown which represents

the worst case scenario, meaning that every product

component needs to be handled and the whole

product needs to be disassembled and assembled. The

dark squares represent the processes. Since the

assembly and disassembly can vary considering the

used resource, the assembly process is marked with a

dash to symbolize the difference. Up shifted (and

down shifted) process (e.g P2.1, P2.2, P2.3) are

attached to subassemblies. The upper (bright) layer

contains the disassembly path while the bottom one

(dark) contains the assembly path. The repairing

process is not regarded yet. They are defined

separately and attached to the error code. A rework

process can contain:

▪ Replacing a presumably defect part

▪ Assembling a missing part skipped (forgotten) in

the final assembly

▪ Checking for craze or damage during assembling

▪ Exfoliation and cleaning the parts afterwards

Figure 3: Overall process list with disassembly (bright lane)

and assembly (dark lane) processes, repair processes are not

considered yet.

However, the overall process list does not pose

any rework job list. The work job list depends on the

error code and lists process steps that lead the worker

to the affected product component and consists of a

disassembly stage as well as a repairing and an

assembly stage. The rework list is generated by the

control system.

4.2 Control System to Generate Error

Code Dependent Rework Jobs and

to Distribute Data to the Right

Resources

Since the planning environment and worker

assistance system are separated, there needs to be a

control system which executes the rework job as

generated based on product variant and error code.

The control system needs two initial information to

generate the rework job. The first information is the

product variant and the other the error code. Based on

the error code, the faulty product part is determined

and a rework job list is generated. The rework job list

contains all processes which are necessary to get to

the product repaired. In Figure 4 and Figure 5, two

different rework job lists for different rework cases

and different degrees of disassembly are shown.

Using the plan data set in the planning environment,

the job list is dynamically created by the control

system and depends on the given error code. The

repairing process is settled in the turning point

between disassembly (upper lane) and assembly

(bottom lane).

Figure 4: Case 1: Rework job shown as graph. The repair

process is the turning point between disassembly and

assembly.

HUCAPP 2018 - International Conference on Human Computer Interaction Theory and Applications

Figure 5: Case 2: Upstream assemblies are not

disassembled, but taken out all in one.

However, during the rework another reason of

failure can always be discovered than suggested.

Therefore, the worker assistance system has to allow

to switch the path to another part, if another reason is

assumed. When the rework job list is set, the control

system reads the first process and distributes the

machine data to the corresponding resource.

When the process is executed, it awaits a feedback

signal from the monitoring resource. If the feedback

is positive, the process data are stored for

documentation purposes and the next process can be

executed. However, the worker has the possibility to

choose another reason of failure. The control system

then calculates the path from the current position to

the new targeted position.

The combination and connection of planning and

control system allows to generate new rework jobs

based on defined reference products and processes,

also new paths during rework. This and the

description of process parameters beforehand make a

dynamical system possible, without modelling every

single rework job and product variant with their

processes.

4.3 Worker Assistance System to

Support Worker during the

Rework Job

The worker guidance in form of a user interface (UI)

is the access point for the worker to interact with the

assistance system. The UI presents data similar to

(Aehnelt and Bader, 2015) considering the process,

the product as well as the resources. The data are

human readable as shown in Figure 6. The processes

are listed on the left side (1) and are ticked green by

the system whenever they are finished according to

plan. In the same list, the next process is already

highlighted and hints to the task to be carried out next.

The current process execution is presented by a

picture or an animation (2) based on the product data.

This visualization helps the worker to find the right

assembly place at the product. Especially for

beginners or less experienced workers, an animation

is helpful to find the right mounting place on the

product. Another advantage is that a picture can be

easily created from given a CAD (Computer-aided

design) model. On the right side of the UI,

information considering the resource to be used (3) is

provided. Beneath the resource information there is

an extra field for additional background information

considering the process or certain product (4). Its

purpose is to explain why certain processes need to be

done to develop awareness and avoid attempts

passing processes, due to the lack of knowledge.

Beneath the product part, there is a detailed

explanation of the process in case that the worker

does not know the process at all (5). At the bottom,

there are three buttons placed. The first button allows

the worker to demand more information considering

the current process (6). The second one quits the

process manually (7). Manual process quitting is

shown in the process list by an orange tick beside the

process. The third button allows the user to toggle

between expert and beginner mode (8). The purpose

of these modes is to provide beginners and

inexperienced worker with sufficient information but

avoiding annoying experienced worker with an

information overflow. This mode is designed for

beginners as well as trainings and qualification for

certain variants. Moreover, if an experienced worker

needs more information for just one process, he

always has the option to do so. The UI is implemented

as a web application and can be used platform

independent. Being an important part, the UI is

considered a resource and a part of the worker

assistance system. A further visualization system is

the laser projector (1 in Figure 2) which can be used

to show the exact order of bolt tightening processes

or highlighting a component which has to be handled

or dis/assembled.

In the current state of the work station setup, the

process checking/ validation is done by two different

systems. First of all, there is a 2D camera (2 in Figure

2) which will only check if the live captured image

matches the reference image. In case it does, the

process is triggered as correctly carried out. This kind

of checking is only applicable if the monitored

product property is limited to a completeness control.

The 2D camera can be replaced by 3D cameras to

ensure that an element is placed correctly e.g.

regarding the depth. Since the image evaluation

technology is processing intensive, it is mostly

reduced to a small area and is not suitable for every

assembly process. Since reference images mean that

appropriate processes need to be recorded once, an

alternative can be to generate the reference data from

CAD to avoid media disruption. Other technologies

involve camera technology (C) too, but reduce the

range of processed wavelength. In this case e.g. only

Intelligent and Flexible Worker Assistance Systems - Assembly Assistance Platform for Planning Assisted Assembly and Rework as Well as

Execution of a Worker-Centered Assistance

infrared LEDs can be detected and used to determine

the position and orientation of an IR-LED afflicted

object.

Figure 6: Worker guidance to lead the worker through the

assembly and allow him to interact with the worker

assembly system.

5 CONCLUSIONS AND

OUTLOOK

This paper describes the development of a human

centred assembly assistance platform which allows to

design and configure the worker assistance system.

To configure the worker assistance system and setup

the job list with less programming knowledge, a

planning environment for assembly planning

engineers is developed. In this environment,

information about parts, processes and resources are

linked with each other to create the maximum job list

for rework. Depending on an error code the job list is

dynamically modified by the control system. For the

modification and reduction of the maximum job list,

a specific rework process is added which describes

the repair work to be executed. The information of the

job list is processed by a worker guidance system to

provide human readable information to the worker.

The worker assistance system is equipped with

resources to document process results like cameras

and EC-driver. Moreover, the worker assistance

system allows short termed change of rework jobs as

well as different stages of disassembly, if the worker

detects unexpected failures.

For the planning environment, a concept exists

which will be implemented as a software solution in

the near future. On the worker assistance system side,

development considering the process evaluation are

still ongoing and concentrated on the development of

a 3D camera based completeness checking. This shall

improve the reliability of recognition and allow easy

creation of reference data by using CAD models,

which can configure any rework station.

ACKNOWLEDGEMENTS

This paper was written in the framework of the

research project NeWiP which is funded by the

German Federal Ministry of Education and Research

(BMBF) and supervised by the lead partner PTKA-

Karlsruher Institut für Technologie under the funding

code 02P14B203.

REFERENCES

Aehnelt, M., Bader, S., 2015. Information assistance for

smart assembly stations, Scitepress.

Aehnelt, M., Urban, B., 2015. In The Knowledge Gap:

Providing Situation-Aware Information Assistance on

the Shop Floor, Second International Conference,

HCIB 2015, Springer.

HUCAPP 2018 - International Conference on Human Computer Interaction Theory and Applications

Botthoff, A., 2015, Hartmann, E. A., Zukunft der Arbeit in

Industrie4.0, Springer Vieweg, Berlin.

Friedmann, M., Trapp, T. U., Stoldt, J., Langer, T., Putz,

M., 2016. A Framework for Information-driven

Manufacturing, 49th CIRP Conference on

Manufacturing Systems.

Hinrichsen, S., Riediger, D., Unrau, A., Assistance Systems

in Manual Assembly In Direct Digital Manufacturing in

the Context of Industry 4.0, Volume 01/2016, p. 3 - 14.

Lusic, M., Fischer, C., 2015. Worker information systems:

state of the art and guideline for selection under

consideration of company specific boundary

conditions, 48th CIRP Conference on manufacturing

systems - CIRP CMS 2015.

Lusic, M., Fischer, C., 2016. Static versus dynamic

provision of worker information in manual assembly: a

comparative study using eye tracking to investigate the

impact on productivity and added value based on

industrial case examples In 49

CIRP Conference on

Manufacturing Systems- CIRP CMS 2016.

Mueller, R., Hoerauf, L., Vette, M., Speicher, C., 2016a,

Planning and developing cyber-physical assembly

systems by connecting virtual and real worlds, Cirp

2016.

Mueller, R., Vette, M., Scholer, M., and Ball, J., 2016b,

Assembly Assistance and Position Data Feedback by

Means of Projection Lasers, SAE Technical Paper.

Weidner, R., Redlich, T., Wulfsberg, J.P., 2015, Technische

Unterstützungssysteme, Springer Vieweg, Heidelberg.

Wiesbeck, M., 2014, Struktur zur Repräsentation von

Montagesequenzen für die situationsorientierte

Werkerführung, Herbert Utz Verlag, München.

Intelligent and Flexible Worker Assistance Systems - Assembly Assistance Platform for Planning Assisted Assembly and Rework as Well as

Execution of a Worker-Centered Assistance