Available online at www.sciencedirect.

com

Available
ScienceDirect
Available online at www.sciencedirect.com
online atonline
www.sciencedirect.com
Available at www.sciencedirect.com
ScienceDirect
Procedia CIRP 00 (2022) 000–000
ScienceDirect
ScienceDirect
Procedia CIRP 00 (2022) 000–000
www.elsevier.com/locate/procedia

Procedia Procedia
CIRP 00 CIRP
(2017)106
000–000 www.elsevier.com/locate/procedia
(2022) 132–137
www.elsevier.com/locate/procedia
9th CIRP Conference on Assembly Technology and Systems
Intelligent overhead9thcrane improves
CIRP Conference operator
on Assembly ergonomics
Technology and Systems and productivity
Intelligent overhead crane improves operator ergonomics and productivity
28tha,CIRP
Abdellatif Bey-Temsamani *, Gert Design Conference,
Schouterden b May
, Jan Van den 2018,
Bergh c Nantes,
, Joram France
Meskens a
, Taner Incircid ,Karel Kellensb
a
DecisionS,
b Flanders Make, Belgium
A new methodology
Abdellatif Bey-Temsamani
b
to
ACRO research group,*, analyze
Gert Schouterden
Department
EDM
a,
the, Jan
of Mechanical
–
functional
Van denKU
Engineering,
UHasselt, Flanders
c
Bergh
Leuven,
Make,
and
, Joram physical
Meskens
ROB core
Belgium
, Taner
lab Flanders
c
architecture
Incirci
Make, of
Belgium,Karel Kellens
a d b

existing products forDepartment

an assembly oriented product family identification
DecisionS, Flanders Make,a
Belgium
ProductionS, Flanders Make, Belgium
d
b
ACRO research group, of Mechanical Engineering, KU Leuven, ROB core lab Flanders Make, Belgium
EDM – UHasselt, Flanders Make, Belgium
c

ProductionS, Flanders Make, Belgium

d

* Corresponding author. Tel.: +32Paul

16 910Stief *, Jean-Yves
800; E-mail Dantan, Alain Etienne, Ali Siadat
address: abdellatif.bey-temsamani@flandersmake.be

Écoleauthor.
* Corresponding Nationale
Tel.:Supérieure
+32 16 910d’Arts et Métiers,
800; E-mail Arts et
address: Métiers ParisTech, LCFC EA 4495, 4 Rue Augustin Fresnel, Metz 57078, France
abdellatif.bey-temsamani@flandersmake.be

Abstract
* Corresponding author. Tel.: +33 3 87 37 54 30; E-mail address: paul.stief@ensam.eu

Operator ergonomics is a crucial factor for well-being at work. In this paper, an intelligent overhead crane (hoist system) is
Abstract
presented to assist operators in handling large loads. Existing overhead cranes have several downsides when it comes to operator
Operator ergonomics
ergonomics
Abstract is a crucial
and productivity. factor
During for well-being
operation, at work.
the equivalent In this
inertia paper,
of the an intelligent
overhead overhead
crane is heavily crane (hoist
affecting system) of
the flexibility is
presented to to
an operator assist
move operators
a load. in handling
This large
can lead loads. Existing
to potential overhead
muscles cranes
issues and to have several downsides
low productivity. when itthis
Throughout comes to aoperator
work, design
Inergonomics
today’s of
upgrade and
business
an productivity.
environment,
existing overhead During
the trendoperation,
crane towards
that allows the
more equivalent
toproduct
improve inertia
variety
both and of the overhead
the customization
flexibility of is crane
systemisand
theunbroken. heavily
Due affecting
to this
ergonomic the of
development,
issues flexibility
the need of
operators,
agile
an and reconfigurable
operator to move aproduction
load. This systems
can emerged
lead to to cope muscles
potential with various products
issues and and
to lowproduct families. To
productivity. design andthis
Throughout optimize
work, production
a design
is described. With human operators being central in the load's manipulation, it is not only important to consider the ergonomics
systems
upgrade asofwell
an as to choose
existing the optimal
overhead crane product
that matches,
allows to product both
improve analysis the methods
flexibility areof needed.
the Indeed,
system most
and of the known
ergonomic issues methods
of aim to
operators,
but also
analyze a
the usability
product or one
of thesefamily
product
machines
on the
during
physical
the design
level.
stage.
Different
This families,
product
includeshowever,
assessment may
of where
differ
controls
largely in terms
canof
be positioned,
the number and
is described.
which With humanfeedback
operatorsand being central in are the provided
load's manipulation, allitof
is these
not only important to consider the ergonomics
nature offorms of control,
components. This fact impedes anfeed-forward
efficient comparison and how
and choice of appropriate affect
product efficiency
family and effectiveness
combinations of the
for the production
but also the
solution usability
in the specific of context.
these machines
After during the
capturing thedesign stage. This
requirements of ofincludes assessment
operators in different of where controls
contexts, a prototypecan be positioned,
is built which
system. A new methodology is proposed to analyze existing products in view their functional and physical architecture. The aim is to cluster
which
includes
these forms
products of
several control,
upgrades
in new assembly feedback
compared and feed-forward
to existing
oriented product systems.
families are provided
for theFirst, and how
a semi-automatic
optimization all of these
movement
of existing assembly affect efficiency
system
lines and the and
(“Follow-Me”) effectiveness
creation of future of the
is implemented.
reconfigurable
solution
This system
assembly in the
systems. specific
reduces
Based context.
operator
on Datum After
forces
Flowand capturing
Chain,increases the requirements
handling
the physical speed
structure of for of operators
the movement in different
in the X-Y
products is analyzed. contexts, a
plane. Secondly,
Functional prototype
subassembliesthe usageis built which
of wireless
are identified, and
aincludes
functional
controls several
is upgrades
analysis
explored is as wellcompared
performed. to existing
Moreover,
as (color-coded) a hybridsystems.
labeling First,
functional
and a semi-automatic
and on themovement
physical architecture
led-indicators graph
crane. systeman(“Follow-Me”)
(HyFPAG)
Finally, is the output
intelligent is implemented.
which
gripping depicts the
system for
This
largesystem
similarity panels reduces
between
is designed operator
product families
and forces and increases
by providing
constructed. Thisdesign handling
grippingsupport
systemspeed
to both, forproduction
is capablemovement in theits
system
of adjusting X-Y
plannersplane.
gripping Secondly,
and points the usage
productautomatically
designers. ofillustrative
Anto wireless
the most
example
controls ofisa explored
nail-clipperasiswell
usedas to (color-coded)
explain the proposed methodology.
labeling An industrial
and led-indicators casecrane.
on the study on two product
Finally, families ofgripping
an intelligent steering system
columnsfor of
suited positions.
thyssenkrupp Presta France is then carried out to give a first industrial evaluation of the proposed approach.
large panels is designed and constructed. This gripping system is capable of adjusting its gripping points automatically to the most
©suited
2017 The Authors. Published by Elsevier B.V.
© 2022positions.
The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 28th CIRP Design Conference 2018.
This
© 2022 Theopen
is an access
Authors. article by
Published under the CC
Elsevier Ltd.BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)
Selection
© 2022 TheandAuthors.
peer-review under
Published responsibility
by Elsevier Ltd.of the scientific committee of the 9th CIRP Conference on Assembly Technology
Keywords: Assembly; Design method; Family identification license (https://creativecommons.org/licenses/by-nc-nd/4.0)
This is an open access article under the CC BY-NC-ND
and
This Systems
is an open
Peer-review underaccess articleofunder
responsibility the CCcommittee
the scientific BY-NC-ND of thelicense
9 th CIRP(https://creativecommons.org/licenses/by-nc-nd/4.0
Conference on Assembly Technology and Systems )
Selection and peer-review under responsibility of the scientific committee of the 9th CIRP Conference on Assembly Technology
and Systems
Keywords: Intelligent overhead crane; Operator ergonomics; Productivity; Load handling

1.Keywords:
Introduction
Intelligent overhead crane; Operator ergonomics; Productivity; Load handling of the product range and characteristics manufactured and/or
1. Introduction customization
assembled in thismakes this In
system. need
thiseven morethe
context, profound. Solutions
main challenge in
Due to the fast development in the domain of employed today
modelling and as state is
analysis of now
practice
not to handle
only to materials
cope with can be
single
The ability to handle loads greater than that allowed by customization makes this need even more profound. Solutions
1.
communication Introduction
and an ongoing trendobjective
of digitization and divided into
products, two
a limited groups:
product (i)
rangethe orconventional
existing (hoist-based)
product families,
human anatomy has remained a central of humanity employed today as state of practice to handle materials can be
digitalization,
The ability manufacturing
to handle enterprises are facing important devices
but also tothat
be include
able to jacks,
analyze and cranes,
to rope
compare hoists.
products etc.
to (ii)
define
across civilizations sinceloads
evengreater
before than that allowed
we began by
recording divided into two groups: (i) the conventional (hoist-based)
challenges
human anatomy in objective
today’s market
has remained a environments:
central objectiveaasofcontinuing
humanity collaborative robotic systems or “cobots”. For
new product families. It can be observed that classical existingthe former group
history. This has inspired inventions, old as the devices that include jacks,
tendency
across towards
civilizations reduction
since of
even product
before
wheelbarrow five thousand years ago, to as new as development
we began times and
recording (hoist-based),
product families currently available
are regrouped incranes,
function rope hoist
industrial hoists.
of clients etc. (ii)
devices are
or features.
collaborative to
cumbersome robotic
use systems
and exert or “cobots”.
dynamic For the
stresses on former
the group
worker’s
shortened
history. This
exoskeletons product
and lifecycles.
objective
smarthas In addition,
inspired
assistive there
inventions,
devices. is an
as increasing
In today’s oldmodern
as the However, assembly oriented product families are hardly to find.
(hoist-based),
body currently available industrial[2].hoist devices are
wheelbarrow
demand
factory the fiveof workers
of customization,
ability thousand
being years
at the moderately
to handle ago,
same timeto as in anew
heavy as
global
loads On duethe to low
product orfamily
no inertia compensation
level, products differ In some
mainly cases,
in two
cumbersome
operators even torefrain
use andfromexertusing
dynamic stresses
installed on theleading
systems, worker’s to
exoskeletons
competition and
with smart
is andassistive
waycompetitors devices.
all over the In today’s
world. This modern
trend, main characteristics:
in an efficient remains extremely relevant [1]. This body low or no(i)inertia
due to ergonomic
detrimental
the number
issues
of components
compensation
and long-term [2]. and (ii)
In some
health
the
cases,
problems
factory
which
improves istheinducing
ability ofthe
efficiency, workers
reduces tocycle
development handle moderately
from
time andmacro heavy
prevents loads
to injury.
micro type of components (e.g. mechanical, electrical, electronical).
operators
[3]. even refrain
This effect is mostfrom using installed
pronounced systems,
for moderate leading
loads, to
as for
in an efficient
markets,
Rapidly results
reducingway
in isdiminished
and remains
batch sizes lot extremely
andsizes duerelevant
requirements [1].mass
to augmenting
of This Classical methodologies considering mainly single products
detrimental ergonomic issues and long-term health problems
improves
product efficiency,
varieties reduces cycle
(high-volume time and production)
to low-volume prevents injury.[1]. or
[3].solitary,
This effect already
is most existing
pronounced product families analyze
for moderate loads, as the
for
Rapidly
To cope with
2212-8271 ©reducing
2022this batch Published
Theaugmenting
Authors. sizes and
variety requirements
as as to beofable
wellLtd.
by Elsevier mass
to product structure on a physical level (components level) which
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)
identify possible
Selection and peer-review optimization potentials
under responsibility in the
of the scientific existing
committee of the 9thcauses difficulties
CIRP Conference regarding
on Assembly an and
Technology efficient
Systems definition and
2212-8271 ©system,
production 2022 The it Authors.
is Publishedtobyhave
important Elsevier
a Ltd.
precise knowledge comparison of different product families. Addressing this
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)
Selection and peer-review under responsibility of the scientific committee of the 9th CIRP Conference on Assembly Technology and Systems
2212-8271 © 2022 The Authors. Published by Elsevier Ltd.
This is an©open
2212-8271 2017access article Published
The Authors. under theby CC BY-NC-ND
Elsevier B.V. license (https://creativecommons.org/licenses/by-nc-nd/4.0)
Peer-reviewunder
Peer-review under responsibility
responsibility of the
of the scientific
scientific committee
committee of theof28th
the CIRP
9 th CIRP
DesignConference
Conferenceon2018.
Assembly Technology and Systems
10.1016/j.procir.2022.02.167
 Abdellatif Bey-Temsamani et al. / Procedia CIRP 106 (2022) 132–137
2 / Procedia CIRP 00 (2019) 000–000
heavy loads (e.g., > 100 kg) the use of supportive systems is
hard to avoid by operators. While for the latter group (cobot),
although commercial cobots are already safe and include smart
control principles, they can’t handle enough payloads required
by industrial needs. Considering the competitiveness of cobots
in comparison with available industrial robots while focusing
on safety parameters, cobots with above 10 kg payload capacity
are expected to hold significant market attractiveness in the
near future.
In this paper, an innovative concept to upgrade industrial
hoist (overhead crane) systems with intelligent features to
make them ergonomic and smart for moderate loads handling
will be introduced.
The paper is organized as follows: in Section 2, an overview
of moderate loads handling requirements is presented with a
human-centered approach. In Section 3, the innovative
architecture of the smart hoist system is presented. This
includes, the FollowMe principle, ergonomic HMI and
ergonomic gripping system. In Section 4, experimental results
are described and discussed. In Section 5, a summary of the
results and next research challenges are given.
2. Human-centered requirements capturing
To capture the industrial requirements for intelligent overhead
cranes a contextual inquiry has been performed at three
manufacturing sites of three multinational companies. A social
and societal ethical procedure [4] approved the method for this
contextual inquiry. The following method has been used: first
companies selected a specific spot at their production floor of
which they thought the introduction of an intelligent overhead
crane would make most sense. They then selected workers that
would be willing to participate in the study based on a limited
set of criteria provided by the researcher conducting the study.
The researcher then visited the site, discussed the approach of
the study with the selected participant(s), ensured the signing
of an informed consent form and then started the contextual
inquiry. During this inquiry, the researcher observed the
participant in his natural working environment, captured movie
clips, asked questions and made notes. The researcher
anonymized all data where necessary, and only kept
anonymized data for further processing.
In total, 7 operators participated in the study. Within each
company, these operators worked in a team. The teams were
split in different ways: at one site, operators mostly worked
alone in their work cell, another team worked within a shared
space along a conveyor belt, and the final team worked together
in a single work cell. All teams had at least one overhead crane
at their disposal that they should use in order to limit the
physical load to an accepted level. The actual use however
differed. Some operators hardly used the overhead crane; others
used them always when required.
Several factors influenced the use or disuse across all teams.
All operators know the importance of ergonomic behavior and
they tried to comply as far as they thought it did not hamper
meeting the team’s targets. This implicit social pressure
affected the usage decisions.
Some factors that negatively influenced the usage are directly
related to the support system itself: physical stress related to the
inertia of the system when rotating or starting and stopping
horizontal movement; the ease of attaching and detaching
loads; and the degree to which (parts of) the overhead crane
hindered execution of other tasks. The ease of attaching and
detaching is not limited to the gripper, but also involves
133
additional actions that have to be performed, with or without
technological support. These can include fetching and aligning
the hoist’s gripper(s) and/or remote or putting the object in a
suitable position.
To effectively address these issues, and increase usability, a
hoist system has been gradually enhanced on three aspects
affecting the effectiveness, efficiency and satisfaction of
operators, or in one term, usability [5].
To demonstrate potential gains, both a smart hoist architecture
and a research prototype involving conscious decisions about
the control over and information provided by the system for
executing certain actions were developed. Using on the results
of contextual inquiry, a design space [6] that supports this
thinking along several dimensions has been defined:
1. Physical load. The physical load on the operator while
executing an action.
2. Speed control. The flexibility of speed control; to what
degree can an operator influence the speed of
execution.
3. Place of control. The relevant parts of the control
interface should be easy to reach during each task,
while limiting the effect on other actions.
4. Function specification. The degree to which the
system provides information about the involved
components of the human-machine interface (HMI).
In its simplest form, this can be a label. More details
can be provided with feedforward [7].
5. Effect specification. The degree to which the system
provides information about the effect of using a
control component of the HMI. I.e. how the system
provides information about the execution of an action
beyond what is directly visible in the environment, in
human-computer interaction this is referred to as
feedback.
6. Instructions. Some tasks may be infrequently used
and/or may require instructions.
The smart hoist architecture focuses on a specific set of actions
and related requirements that arose from the requirements at the
company sites. It targets a scenario in which operators should
move a large variety of irregularly shaped flat objects with
various dimensions across a relatively large distance using an
overhead crane. This scenario combines several properties of
the observed cases in which we identified research challenges.
The first action is flexible horizontal movement in any direction
(variable speed for horizontal movement, with limited load,
controlled on handler or product). This function should be
largely invisible to the user, except for turning it on or off.
A second major component is flexible gripper configuration,
which should have limited physical load, speed can be fixed
(based on observations, mostly a single speed level was used).
As the proposed gripper has 5 degrees of freedom during
configuration, automated gripper configuration should be
default to limit the amount of necessary actions. A manual
backup option is required in case of problems with automated
gripper placement detection. The place of control is either on
the operator or on the handler. Function and effect specification
is necessary for both automated and the manual backup. A
limited amount of instructions may be necessary for the manual
backup, as it should not be necessary to use it much.
Besides those actions, the HMI should provide access to the
default actions: up/down (fixed speed) and gripper activation.
134 Abdellatif Bey-Temsamani et al. / Procedia CIRP 106 (2022) 132–137
/ Procedia CIRP 00 (2019) 000–000
Figure 1 provides a visualization of the hoist characteristics in
the initial situation (current practice) and the envisioned
situation for these tasks: gripper placement, gripper activation,
horizontal and vertical movement. Improvements both in load’s
gripping and hoist motion features are clearly highlighted in
comparison to current practices.
Figure 1: Visual depiction of the characteristics of a default overhead
crane and its properties for gripper (placement configuration and
activation) and movement (horizontal and vertical). Starting condition
(current practice) and envisioned overhead crane.
3. Innovative overhead crane architecture
3.1. Follow-Me principle
Traditionally, when an operator wants to move a load using
a manual overhead crane, he / she exerts a sideways force on
the load. Through tension in the hoist chain/cable, the mobile
parts of the overhead crane are moved. This requires relatively
much force from the operator, not only because of the inertia
of the load, but also because of the inertia of the mobile parts
of the overhead crane itself.
The Follow-Me principle focuses on compensating the forces
that are required to move the inertia of the mobile parts of the
overhead crane. Figure 2 shows the Follow-Me principle in a
schematic way. If an operator moves the load, the angle
between the hoist chain/cable (𝛼𝛼) will be non-zero. By using
this angle as an input to a proportional-derivative feedback
control loop (PD-controller), the moving parts of the hoist can
be actuated to compensate for their inertia. The amount of
actuation can be changed by tuning the PD-controller gains.
Figure 2: Follow-Me principle: the inertia of the hoist is compensated
by actuating the hoist based on the angle between the chain/cable and
centerline of the hoist (α). This results in lower operator forces.
3
A multibody simulation model of an overhead crane, used to
compare Follow-Me principle with a traditional/passive system
was performed. This simulation shows that the forces required
to move the load from one point to another can be reduced
drastically using the Follow-Me principle, as can be seen by
comparing Figure 3 and Figure 4. For this operator movement-
profile, the settle time and peak operator force are reduced by
a factor ≥ 2 using the Follow-Me principle.
Figure 3: Results of an operator movement using a passive overhead
crane. The settle time is ~20s and the peak operator force is ~170N.
Figure 4: Results of an operator movement using an overhead crane
actuated with the Follow-Me principle. The settle time is ~10s, with a
peak force of ~60N.
3.2. Ergonomic and adjustable gripping system
In order to attach the presented Follow-Me hoist system to the
products to be handled, an ergonomic gripper system has been
developed. Starting from the industrial needs of the involved
companies, the design requirements of this gripping system
were defined as:
• Modular - Traditional grippers for moderate and
heavy loads (>25kg) are typically customized for
single products or product series leading to both
significant investments and changeover times. Target
here was to design a modular gripper frame able to
handle a wide variety of products. Therefore the
gripper frame should be compatible with both multiple
grasping principles [7] and a variable number of
gripping points.
• Adaptable – the position of the specific gripping
(contact) points can be adjusted in the X-Y plane.
• Scalable – by using standard components (e.g. linear
guides, motors, gears…) the gripper frame can be
easily scaled towards other product ranges.
 Abdellatif Bey-Temsamani et al. / Procedia CIRP 106 (2022) 132–137
4 / Procedia CIRP 00 (2019) 000–000
• Automatable – next to the manual (backup scenario)
repositioning of the grasping points, the gripper should
be able to adjust itself in an automated scenario.
• Safe and user-friendly – of course the gripper system
should work in a safe and ergonomic way and easy
controllable by the human operator (e.g. HMI in
Section 3.3)
Figure 5 illustrates the final gripper frame design with both the
symmetric as well as asymmetric variants in the Y-direction.
Figure 5: 3D-model of the modular and adjustable gripper system with
4 gripping points and both symmetric (top-right) and asymmetric
(down-left) variants in the Y-direction.
The presented gripper frame has a maximum payload of 300 kg
and is equipped with 4 Goudsmit Magvacu combi grippers
which can be positioned in a rectangular frame ranging between
275x580mm and 1100x1050mm by 5 SEW Compact Extra-
Low Voltage Drives. These drives can be controlled in three
different ways:
• Manual mode (backup scenario) via forward and
backward buttons on HMI.
• Semi-automated via predefined coordinates in look-up
tables (e.g. product variants indicated by ERP system)
• Fully-automated via coordinates determined by an
integrated vision system mounted on the gripper.
The related vison algorithm as well as steering of the motors
and grasping points run on an integrated RevPi Core 3+.
As illustration, Figure 6 shows the application of the gripper
frame on a flat panel with asymmetric ventilation grille.
Figure 6: Application of the gripper system on a flat panel with an
asymmetric ventilation grille.
3.3. Ergonomic HMI
To make the system relatable for companies, a state-of-the-art
commercial control offering has been selected as starting point
for the HMI: the SIATECH BabHAND [8]. It is a wireless
control that can be attached magnetically to an armband and
that can be wired to a ring with two buttons that are mapped to
one row of the colored buttons on the wireless control (see
Figure 7). The figure shows that the buttons are both colored
135
and have arrows that indicate the corresponding movement
directions.
As discussed in section 2, for default operations we only need
vertical movement as horizontal movement is done using the
FollowMe principle (section 3.1). For the default operations we
decided to add labeling (Figure 8), while making the mapping
as intuitive as possible: start grippers (green), stop grippers
(red). For the control of gripper placement we use left-to-right
chronological order; first take a picture (yellow), then move the
grippers (blue).
The manual backup for gripper movement is expected to be a
less used process and involves too many potential actions to fit
on a controller. We split this process up in two parts: deep push
for movement axis selection with button placement
corresponding to axis placement (Figure 7, middle) and we use
the original button labeling to move the grippers along the
activated axis. Colored LEDs provide feedback on the currently
active axis and active gripper movement. Short instructions are
provided on the central part of the handle to help operators to
successfully use these functions. This part of the handle can
also function as a holding place for the controller when the
operator is holding the gripper handle. The middle part is
recessed to ensure reachability of the buttons while holding on
the handle.
Figure 7: Wireless controller with default button assignment, axis
selection for manual movement and actions when a horizontal axis is
selected
When the operator is not holding the handle; e.g. to get a better
view on alignment of the gripper with the product or the
product with its destination, the controller can be put on the
operator’s arm. The operator can use the buttons on the ring
controller to carry out the actions considered most important at
that specific moment. Selecting which row of colored buttons
is controlled by the ring controller is done using the buttons on
the black half circle.
Figure 8: Rough sketch of the (envisioned) led notifications for
showing system status and the gripper handle with room for the
magnetic controller in the middle
4. Experimental results and discussion
4.1. Follow-Me principle
The Follow-Me principle has been tested on a dual axis
overhead crane (Figure 9). The focus here is on movement
136 Abdellatif Bey-Temsamani et al. / Procedia CIRP 106 (2022) 132–137
/ Procedia CIRP 00 (2019) 000–000
along the Y-axis, as the inertia of the moving parts is the
biggest in that direction (due to the mass of the moving beam).
Figure 9: a dual axis overhead crane to test the Follow-Me principle.
Figure 10 shows the measured operator force over time
during a single movement cycle for 2 tested loads (12kg and
97kg) with both a disabled (passive) and enabled Follow-Me
principle (Follow-Me). The cycle is defined as:
• a straight movement along the Y-axis for 2.5m
• a backwards movement along the Y-axis (back to
the start position)
The cycle is performed at a pace that feels natural for the
tested load and system (passive / Follow-Me).
From this test it can be seen that the maximum operator force
is lowered by a factor ~2 by using the Follow-Me principle. The
time to perform the cycle in a natural way is also reduced
drastically (factor is depending on the load).
In the passive case, the force profile shows multiple peaks
during the cycle. The moving parts of the overhead crane are
overshooting the desired position of the load. To stabilize the
load, the inertia of the moving parts needs to be stopped, for
which extra operator force is needed causing the extra peaks.
With an actuated overhead crane, the overshoot is drastically
reduced. This means that it is much easier to stabilize the load.
Because of these faster settling times, the cycle time can be
reduced drastically.
Figure 10: Measured operator forces over time during a single
movement cycle for 2 tested loads (12kg and 97kg) with both a
disabled (passive) and enabled Follow-Me principle (Follow-Me)
When moving a heavy load, the resulting angle between the
chain/cable and the centerline will be relatively small
5
compared to when moving a lighter load. This means the
operator needs to exert more force to move the load.
In consequence, the system can be improved by
estimating/measuring the load that is being lifted. If an estimate
of the load is known, the PD-controller gains can be adapted to
account for the change in load and thus making the Follow-Me
principle more reactive to the change in angle.
Figure 11 shows the maximum operator force for increasing
loads for an overhead crane with Follow-Me principle disabled
(passive), enabled (Follow-Me) and with a load adaptive
Follow-Me principle (Adaptive Follow-Me). The maximum
operator forces can thus be further reduced using a load
adaptive feedback controller.
Figure 11:More operator force is needed for increasing loads. Using
an overhead crane equipped with (adaptive) Follow-Me principle, the
operator forces can be reduced drastically
4.2. Fully integrated system
The presented gripper (Section 3.2) has been attached to the
hook of the lift of the overhead crane system. Wireless HMI
receiver unit has been integrated with the control system of
overhead crane. Furthermore, both an electrical interface which
carries control signals, supply power and safety connections as
well as the required pneumatic connections have been made
between the overhead crane control system and the gripper. In
this way all sub systems have been integrated with each other
as shown in Figure 12.
Figure 12: Integration of gripper (left). HMI assembly close up (right)
With the button to function mapping of , by using only one
hand, it is possible to control the gripper manually while
moving in horizontal XY plane assisted by Follow-Me
function. This has the advantage that other hand is freed up to
perform any other operation or function in parallel. Balancing
and positioning the gripper, controlling lift up and down
motion, balancing and supporting the operator himself to avoid
an ergonomically risky posture are just few examples that can
 Abdellatif Bey-Temsamani et al. / Procedia CIRP 106 (2022) 132–137
6 / Procedia CIRP 00 (2019) 000–000
be achieved by freeing up a hand of the operator during manual
handling operations.
The integrated system has been tested with a safety fence frame
and wooden plate as the load. These loads were successfully
picked up using gripper’s magnetic and vacuum suction cups
and by adjusting gripper arms at right size that is best suited for
the load. In Figure 13, on the left, the metal frame was gripped
at center point while keeping the arms of the gripper in
contracted state. On the right, gripper arms were extended
which provided a clear advantage with regards to ease of
handling and improvement in lower back posture.
Figure 13: Bad lower back posture (left photo) when gripper arms are
not adjusted for the load. Better gripping posture is realized by the
extension (right photo).
Finally, several pick and place type test runs have been
performed with the integrated system; picking up various loads
moving them with Follow-Me function and placing them on
another location. More tests are foreseen with operators who
have experience working with hoists. The idea is to collect their
feedback via an operator acceptability survey after
experimenting with the integrated system. The survey results
can provide additional insight for further improvements with
respect to ergonomics, smoothness and speed of operations for
future implementations.
5. Conclusions and next steps
In this paper, we presented an innovative design of a hoist-
based overhead crane system to handle moderate loads
(<100kg). The innovative architecture covers different
technological components, namely an intelligent system (based
on visual sensing and PID controller), that reduces operator’s
forces by reducing the inertia of the overhead crane system, a
flexible gripping system that reduces posture problems of
operators while gripping a load, and an ergonomic HMI, that
allows to free up operator’s hands for multi tasks in a
production floor. These technological components were
137
implemented in an existing traditional overhead crane system
and upgrade it to a next generation ergonomic loads handling
system.
Experimental results clearly show benefits on operators
forces reduction (a factor x2 reduction), as well as ergonomic
loads handling through the flexible gripping and HMI systems.
Different prototypes have been built by the research team
with incremental upgrade and additional options. The initial
prototype based on one axis motion has bee recorded with a
practical scenario [9].
The research team is further researching techniques to refine
the HMI. The goal is to have it always available and usable,
even when faced with a comprehensive set of infrequent control
actions.
Acknowledgements
This research is supported by Flanders Make, the strategic
research Centre for the Manufacturing Industry and the
Flemish Innovation and Entrepreneurship Agency through the
research project 'SMARTHANDLER_ICON’ (project number:
HBC.2018.0251).
The authors would like to thank all project’s partners for their
inputs and support to make this publication.
References
[1] Riyaz et al., A review of various material handling systems, International
Journal of Innovations in Engineering and Science, vol. 2, nr. 10,2017.
[2] Davood et al., Effect of neck posture on cervicothoracic loads in overhead
crane operators, International Journal of Occupational Safety and
Ergonomics, ISSN 1080-3548, 2018.
[3] Muthukumar et al., Study on discomfort in mobile crane operation,
International Research Journal of Engineering and Technology, Vol. 4,
Issue 12, 2017.
[4] The social and societal ethical committee of Hasselt University in Belgium.
https://www.uhasselt.be/Social-and-Societal-Ethics-Committee. Last
visited: September 29, 2021.
[5] EC International Electrotechnical Commission: Din en iso 9241{110:2020
ergonomics of human-system interaction - part 110: Interaction principles.
https://www.iso.org/obp/ui/#iso:std:iso:9241:-110:ed-2:v1:en
[6] Van den Bergh and Heller. 2021. MyLYL: Towards Flexible Interaction
Design for Operator Assistance Systems, presented at International
Workshop on Human-Centered Software Engineering for Changing
Contexts of Use, https://sites.google.com/view/hcse-workshop-
interact2021/agenda.
[6] Vermeulen et al. 2013. Crossing the bridge over Norman's Gulf of
Execution: revealing feedforward's true identity. In Proceedings of the
SIGCHI Conference on Human Factors in Computing Systems (CHI '13).
Association for Computing Machinery, New York, NY, USA, 1931–1940.
DOI:https://doi.org/10.1145/2470654.2466255.
[7] Fantoni, et al., Grasping devices and methods in automated production
processes. CIRP Annals – Manufacturing Technology. Vol. 63, 679–701,
2014.
[8] SIATECH BABHAND specifications. Last visited: September 29, 2021.
https://www.siatech.fr/en/products/#products-babhand
[9]
https://www.youtube.com/watch?v=GTKRwtfGzCo&list=PLP6P3b3pzc
HmFeyYyZSyezwY6nfjTrBS9&index=5