Advances in Production Engineering & Management

ISSN 1854-6250

Volume 16 | Number 4 | December 2021 | pp 431–442

Journal home: apem-journal.org

https://doi.org/10.14743/apem2021.4.411

Original scientific paper

The impact of the collaborative workplace on the production
system capacity: Simulation modelling vs. real-world
application approach
Ojstersek, R.a,*, Javernik, A.a, Buchmeister, B.a
aUniversity

of Maribor, Faculty of Mechanical Engineering, Maribor, Slovenia

ABSTRACT

ARTICLE INFO

In recent years, there have been more and more collaborative workplaces in
different types of manufacturing systems. Although the introduction of collaborative workplaces can be cost-effective, there is still much uncertainty about
how such workplaces affect the capacity of the rest of production system. The
article presents the importance of introducing collaborative workplaces in
manual assembly operations where the production capacities are already limited. With the simulation modelling method, the evaluation of the introduction
impact of collaborative workplaces on manual assembly operations that represent bottlenecks in the production process is presented. The research presents
two approaches to workplace performance evaluation, both simulation modelling and a real-world collaborative workplace example, as a basis of a detailed
time study. The main findings are comparisons of simulation modelling results
and a study of a real-world collaborative workplace, with graphically and numerically presented parameters describing the utilization of production capacities, their efficiency and financial justification. The research confirms the expediency of the collaborative workplaces use and emphasise the importance of
further research in the field of their technological and sociological impacts.

Keywords:
Simulation modelling;
Production system capacity;
Industry 5.0;
Assembly line;
Human-robot collaboration;
Collaborative workplace
*Corresponding author:
robert.ojstersek@um.si
(Ojstersek, R.)

Article history:
Received 7 September 2021
Revised 11 December 2021
Accepted 13 December 2021
Content from this work may be used under the terms of
the Creative Commons Attribution 4.0 International
Licence (CC BY 4.0). Any further distribution of this work
must maintain attribution to the author(s) and the title of
the work, journal citation and DOI.

1. Introduction
Optimization of production systems has been an attractive research field for many decades. Researchers are constantly wondering how to improve production system capacity or use it as efficiently as possible. In recent years, there have been an increasing number of collaborative machines that, together with workers, form high flexible, economically justified collaborative workplaces. Collaborative machines are to some extent already well studied, but their impact on the
collaborative workplace, on workers and more broadly on the manufactured system is often unknown. Where is the turning point when a collaborative workplace is economically, socially and
from a capacity standpoint justified? We want to answer this complex research question.
Researchers have been asking for years who is working with whom (human with robot, or vice
versa) [1]. This issue, given the complexity of the social dimensions of the collaborative workplace
and the security parameters, raises a lot of unanswered questions from the worker's point of view
[2]. The findings show that we are talking about a hybrid research area, associated with a concept
of Industry 5.0 [3, 4], where safety meets ergonomics, technological efficiency, and the unanswered
431

Ojstersek, Javernik, Buchmeister

question of the integrated impact of collaborative workplaces on the production system [5]. As we
know, proper ergonomic analysis of workplaces significantly improves their productivity [6], but
this is only proven in manual assembly workplaces; how different production parameters can affect the collaborative workplace is not known. Only general guidelines for the preparation and
arrangement of the collaborative workplaces are given [7], where the authors still draw parallels
with the manual assembly workplaces [8]. The shortcomings of such research are highlighted
when we want to analyse in detail the impact of collaborative workplaces on the sustainable justification of the production system [9]. The authors cite limitations in terms of different time, cost
and technological suitability of collaborative workplace parameters. Determining the appropriate
"collaborative" parameters [10] is crucial in the use of efficient and safe (for worker and robot)
workplaces [11]. Research work presents that states of safe parameters of speed and acceleration
of collaborator robots in a common workspace [12], but how the change of parameters affects the
efficiency of the collaborative workplace and other production capacities is hard to define [13].
Due to these limitations, researchers want to provide a general methodology for the introduction
of collaborative workplaces [14], but when one of the general advantages is high flexibility of collaborative machines and associated production systems [15] in which we include them, the implementation is very demanding, most often made individually [16]. The answers to the questions
about the feasibility of introducing collaborative workplaces must thus respond to an appropriate
investment strategy [17] and the sustainable justification of such workplaces and their wider impact [9]. Correlations between these parameters [18] can be well represented by simulation modelling methods [19], where an integrated approach to planning and deployment of collaborative
workplaces can be evaluated and the collaborative workplace constructed accordingly [20]. Recent research shows that it will be necessary to know the technological behaviour of the collaborating machine and, more importantly, its sociological impact on the co-worker [21]. The response
that a co-worker may have to a collaborative workplace is complex and individual according to
the employee's condition. More broadly, the impact of a collaborative workplace can significantly
change production capacities and their efficiency. It should be emphasized that the collaborative
workplace can be placed in different types of production systems, in different configurations,
which represent an additional complexity of its optimization [22].
In our research work we want to answer the question of determining the collaborative workplace parameters (time study and financial norms) when introducing it into an existing production system. In doing so, we focus on the use of simulation modelling methods and the evaluation
of a real-world collaborative workplace. Data from detailed time and costs analysis will enable the
implementation and comparison of the broader impact of the collaborative workplace on the entire production system, where a comparison will be made between manual assembly and collaborative workplaces. The research is based on the study of the production system of assembly line
and attempts to improve its limited production capacity.

2. Problem description

Optimizing an assembly production line system is a major challenge if the system is already at the
minimum possible takt time and is no longer able to optimize assembly processing time for individual workplaces. Such an assembly line system, when orders increase, faces the inability to
achieve the desired quantity of products with limited assembly capacities. In recent years, manual
assembly workplaces have been automated and robotized, and such workplaces have some limitations as production capacity increases, investment costs increase, new equipment is introduced,
and the size of such fully automated cells increases assembly line footprint. Given that the assembly line production system presented in Fig. 1 and the corresponding processing time data in Table 1 indicate assembly line constraints at manual assembly workplaces Mas8 and Mas9, where the
assembly processing time is equal to the line takt time. The question of the feasibility of introducing collaborative workplaces where the manual assembly is upgraded with the capacities of a collaborative robot, whose initial investment and introduction to an existing job is less demanding,
is questionable.
432

Advances in Production Engineering & Management 16(4) 2021

The impact of the collaborative workplace on the production system capacity: Simulation modelling vs. real-world …

2.1 Production system description
The research problem deals with the products assembly line with ten manual assembly workstations (Mas) and associated workstations processing times (work-element times) presented in
Fig. 1 and Table 1 respectively. The assembly line production system has a certain line takt time
of 54 s, a constant speed of the conveyor belt of 1.2 m/min, an additional mark-up coefficient of
the conveyor belt length of 0.05. The assembly line is carried out in three shifts in five working
days a week. Workers in the manual jobs of the assembly have a certain useful number of working
hours in a shift, lasting 7.5 h. Transport to the initial station of the assembly line and shipment of
finished products is carried out with the use of forklifts. Input semi-finished products and component assembly accessories are always available to the assembly workers.
Workplace
Processing time (s)

Mas1
52

Table 1 Assembly line manual workplaces processing times
Mas2
Mas3
Mas4
Mas5
Mas6
Mas7
Mas8
51
49
48
52
51
52
54

Mas9
54

Mas10
48

Fig. 1 Production system – manual assembly line with ten workplaces (3D model)

Eqs. 1 to 7 represent a numerical calculation of the assembly line characteristics. Numerically
determined parameters are consistent with real-world production systems and serve as a basis
for building a simulation model. For further calculations, next variables are defined:
takt
Uc
nc
ns
ηc
ηl
Qe
Md
lc
kl
vc
dw
dp
tf
t1
E

Takt time
Useful capacity
Useful number of working hours in one shift
Number of shifts
Worktime efficiency coefficient
Line efficiency coefficient
Quantitative efficiency
Number of workplaces
Conveyor length
Mark-up coefficient for the conveyor length
Conveyor speed
Distance between workplaces
Distance between products on the conveyor
Product’s flow time
Operation processing time
Additional number of products on the conveyor

Eq. 1 defines useful capacity of the assembly line per working day, including three shifts working schedule, 7.5 h of useful working hours and worktime efficiency coefficient of 0.92. High worktime efficiency coefficient is used in relation to assembly line characteristics.
𝑈𝑈𝑐𝑐 = 𝑛𝑛𝑠𝑠 · 𝑛𝑛𝑐𝑐 · 𝜂𝜂𝑐𝑐 = 3 · 7.5 · 0.92 = 20.7

Advances in Production Engineering & Management 16(4) 2021

h

day

or 1242

min
day

(1)

433

Ojstersek, Javernik, Buchmeister

In corelation to defined takt time of 0.9 min, which is minimum possible takt time for presented
operations in Table 1, and defined line’s useful capacity, quantitative efficiency is defined as presented in Eq. 2.
𝑄𝑄𝑒𝑒 =

𝑈𝑈𝑐𝑐

𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡

=

1242
0.9

= 1380

pcs

(2)

day

With the know number of workplaces (Md = 10), takt time and total processing time the final
assembly line theoretical efficiency is determinated by Eq. 3.
𝜂𝜂𝑙𝑙 =

∑ 𝑡𝑡1

𝑀𝑀𝑑𝑑 · 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡

=

509

10·54

= 0.943 or 94.3 %

(3)

Defined number of workplaces, known distance between workplaces (dw = 2.16 m) and proposed mark-up coefficient for the conveyor length (kl = 0.05) the optimum conveyor length is defined by Eq. 4:
(4)
𝑙𝑙𝑐𝑐 = 𝑀𝑀𝑑𝑑 · 𝑑𝑑𝑤𝑤 · (1 + 𝑘𝑘𝑙𝑙 ) = 10 · 2.16 · (1 + 0.05) = 22.68 m
Distance between products on the conveyor (dp) is known when the speed of conveyor is defined (vc = 1.2 m/min) and multiplied with the takt time of 0.9 min. Shown by the Eq. 5:
𝑑𝑑𝑝𝑝 = 𝑣𝑣𝑐𝑐 · 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 1.2 · 0.9 = 1.08 m

An additional number of products on the conveyor is defined by the Eq. 6.
𝐸𝐸 = (𝑀𝑀𝑑𝑑 − 1) ·

𝑑𝑑𝑤𝑤
𝑑𝑑𝑝𝑝

= (10 − 1) ·

2.16
1.08

= 18

(5)

(6)

Knowing the number of workplaces, additional number of products on the conveyor, distance
between workplaces and distance between products on the conveyor products’ flow time can be
defined by Eq. 7.
(7)
𝑡𝑡𝑓𝑓 = (𝑀𝑀𝑑𝑑 + 𝐸𝐸 ) · 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = (10 + 18) · 0.9 = 25.2 min
2.2 Collaborative workplace (CWas) description

Our own designed flexible collaborative workplace, in Fig. 2, consist of a worktable ⑦, a collaborative robot UR3e ⑧, a collaborative gripper Robotiq 2F-85 ⑨, a pallet of semi-finished products
④, pallet of finished products ⑤, and three types of semi-finished products need to be assembled
(type one yellow brick ①, type two green brick ② and type three 4×2 brick ③).
To run simulation models and study the capacity of the assembly line production system, it was
necessary to determine the processing time of collaborative assembly operation between worker
and collaborative robot. To determine the most accurate processing times, we carried out the time
study evaluation with different speeds and accelerations of the collaborative robot (Table 2), evaluating different workers, in sitting and standing positions.

434

Fig. 2 Layout of a collaborative workplace

Advances in Production Engineering & Management 16(4) 2021

The impact of the collaborative workplace on the production system capacity: Simulation modelling vs. real-world …

Speed (%)
100
80
60

Table 2 Collaborative robot speeds and acceleration data
Linear movement
Joint movement
(mm/s) Acceleration (%) (mm/s2)
add
Speed (%) (°/s) Acceleration (%)
750
100
2000
100
180
100
600
80
1600
80
144
80
450
60
1200
60
108
60

(°/s2)
360
288
216

By performing evaluation of the seventy-two iterations of the assembly operation, we were able
to accurately determine the collaborative workplace’s processing time, human operational time,
optimal speed of the collaborative robot, and gain a better understanding of the collaborative robot influence on the worker.
The assembly operation consisted of simple assembling of three semi-finished products ①, ②
and ③ into one finished product ⑤. In the initial stage, the collaborative robot picks up a semifinished product ③ and move it to the assembly location (Fig. 2). At the assembly location, collaborative robot stops and waits for the worker to attach two semi-finished products ① and ②.
At this stage of the assembly operation, the worker attaches two semi-finished products ① and
② to the semi-finished product ③, which is held in the collaborative gripper. The attachment
position of the two semi-finished products ① and ② was determined, the green brick ① is always at the top, the yellow brick ② is always at the bottom, while the order of the composition
is: first the yellow brick is attached followed by the green brick. After the three semi-finished
products ①, ② and ③ were assembled into a finished product ⑤, the collaborative robot move
and place the finished product on the pallet with the finished products. The working process is
finished when the pallet of finished products ⑥ is filled. It should be noted that the work process
was carried out in a laboratory environment, so the position of the semi-finished products pallet
④ was fixed, while in a real-world assembly line operation the pallet would be transported by a
conveyor.

3. Simulation modelling

Given the presented line assembly production system and the problem of improving limited production system capacity by introducing collaborative workplaces, we used simulation modelling
to build a simulation model of the assembly line production system and to analyse the collaborative workplace in detail. Initially, the input parameters of the assembly line production system
presented in Section 2 were upgraded by numerical modelling of the manual and collaborative
workplaces costs, further used to study individual workplaces financial justification.
For the simulation model, the workplaces cost calculation (Mas and worker-robot collaborative
workplace CWas) was performed. Table 3 presents the data and cost calculation of the workplaces
provided for the results implementation into a discrete event simulation environment Simio. Obtained data provides the basis for the validation of the obtained results in Section 4.
Table 3 Workplaces cost calculation data
Cost calculation parameter
Mas
Purchase value of the machine (€)
11,666
Machine power (kW)
0.1
Workplace area (m2)
6
Depreciation period (year)
7
Useful capacity of the machine (h/year)
5216
Machine write-off value (€/h)
0.32
Interest (€/h)
0.01
Maintenance costs (€/h)
0.02
Production system area costs (€/h)
0.12
Electrical energy consumption costs (€/h)
0.02
Machine operational costs (€/h)
0.49
Workplace total costs (€/h)
11.54
Workplace cost per item (€/piece)
0.173

Advances in Production Engineering & Management 16(4) 2021

CWas
35,000
0.1
6
7
5670
0.88
0.03
0.06
0.11
0.02
1.1
12.76
0.185

435

Ojstersek, Javernik, Buchmeister

3.1 Production system modelling
The assembly line production system was modelled in the Simio software environment. The simulation model shown in Fig. 3 represents the assembly line, where all ten workplaces are devoted
to manual assembly stations. The input parameters of the assembly line are the same as presented
in Section 2, in addition, the parameters of workplaces costs evaluation according to mathematical
modelling in Section 3 are added. The simulation model operates in three shifts, five working days
a week. The model assumes that input materials and semi-finished products are always available,
the system operates at 94.3 % efficiency rate. There are no unknown failures during the assembly
operation. The main purpose of the simulation model is to evaluate the possibility of introducing
collaborative workplaces to existing manual assembly workplaces with limited capacity.
In the intermediate graphic presentation (Fig. 3) we can observe that in the manual assembly
jobs M_As8 and M_As9 bottlenecks of the production system appear, potentially these two manual
assembly workplaces represent the final capacity of the evaluated assembly line. Since these two
workplaces are about equalizing the time of the assembly cycle and the time of the assembly line
takt time, it is advisable to optimize these two workplaces to raise production system capacity.

Fig. 3 Simulation model of the manual workplace’s assembly line (2D model)

As a proposal to increase the production system capacity, the assembly line in Fig. 4 represents
the introduction of one collaborative workplace, where one worker serves two collaborative robots. Fig. 4 shows this workplace with one AsCw1 worker workplace and two collaborative robots
in AsCr1 and AsCr2 workplace. As shown, instead of ten workers in production, we now have only
nine workers. In consideration, we have eight manual assembly workplaces and one collaborative
workplace including worker and two robots. With the input parameters of the production system,
all parameters of manual assembly workplaces remain unchanged. We added the input data for
the collaborative workplace. The preliminary phase of graphic presentation of the simulation
model shows the elimination of previous (Fig. 3) bottlenecks and the potential increase in the
production system characteristics.

Fig. 4 Simulation model of the proposed collaborative workplaces assembly line

3.2 Collaborative workplace modelling

Collaborative workplace design and collaborative assembly operation were modelled in Siemens
Process Simulate environment. The Process Simulate software environment allows us to model
different systems or scenarios, simulate operations (machine or human), analyse human movements, optimize the production system, create robot programs, etc.
We have started by modelling the collaborative workplace and added all the necessary components for the collaborative work process, as shown in Fig. 5a. To perform the actual simulation,
we first had to define the correct kinematics of the collaborative robot and the collaborative gripper. Properly defined kinematics is crucial to the functionality of the simulation model, as the same
program of collaborative robot is running inside the simulation model as in the real-world
436

Advances in Production Engineering & Management 16(4) 2021

The impact of the collaborative workplace on the production system capacity: Simulation modelling vs. real-world …

application. After defining the exact locations of components, we have started to create the program for collaborative robot. In the program, we adjusted the movement of collaborative robot
according to the range, kinematics, speed, type of movement and human safety. After completing
the program in Process Simulate environment, we have transferred the program from the virtual
to the real-world collaborative robot, through an integrated interface, where we only checked
proper functioning and safety of the program.
After ensuring the relevance of the collaborative robot and the human-robot collaboration, the
collaborative operation was simulated (Fig. 5b). The goal of simulating human work was to compare the simulation processing time against a real-world study human processing time.
Table 4 shows the results of the real-world collaborative workplace time study evaluation, in
which we conducted a time study of four workers with different ages (between 25 and 45 years).
Workers were instructed for the correct assembly operation order and needed collaborative
workplace knowledge. When performing time study, we have unknowingly changed the speed of
the robot for the workers and automatically measured and recorded the assembly process processing time. We performed seventy-two iterations to study the time of the collaborative assembly
operation. The results in Table 4 represent the average results of these iterations for an individual
worker and total average assembly processing time with respect to the robot speed and acceleration. The total average processing time of collaborative assembly was used in both the simulation
model of the production system, in Simio, and the collaborative workplace, in Process Simulate.
a)

b)

Fig. 5 Simulation model of human-robot collaboration

Table 4 Real-world evaluated workers (Wi) collaborative workplaces processing times
Workers processing time (s)
Robot speed/acceleration (%)
W1
W2
W3
W4
60
90.146
89.796
86.950
86.566
80
81.038
75.234
68.444
67.434
100
64.544
62.108
59.286
63.318

4. Results and discussion

Average
88.36
73.04
62.31

The results in Table 5 show the simulation modelling results of the costs and the utilization rate
of an individual manual assembly workplaces. The workplaces cost depends on the number of
processed products in the simulation time of five working days, working in three shifts. According
to the determination of the assembly line production cost per individual piece and the type of
workplace (data presented in Table 3), we can see how the costs affect the number of production
pieces on the assembly line. More important is the parameter of workplace utilization, for which
the utilization of the first workplace Mas1 is not relevant, since the simulation model assumes a
constant supply of semi-finished products to the first assembly workplace. However, we can see
that the highest utilization rate is in the workplaces Mas5, Mas7, Mas8 and Mas9. Based on a detailed
Advances in Production Engineering & Management 16(4) 2021

437

Ojstersek, Javernik, Buchmeister

analysis (throughput time, average time in station and number of entered/exited products) of the
results, we find that the bottleneck of the production system is represented by the workplaces
Mas8 and Mas9, where the assembly processing time is equal to the assembly line takt time. Workplaces Mas8 and Mas9 are at the maximum of their capacity and prevent smooth flow of products
through other workplaces. As we can see, the numerical results confirm the preliminary graphical
representations of the simulation model and suggest the importance of optimizing these two
workplaces.
When introducing a collaborative workplace (replacement of the Mas8 and Mas9), which contains
one CWas1 robot collaborative workplace and two CR1 and CR2 collaborative robots, the simulation
results in Table 6 prove the feasibility of introducing such workplaces at evaluated assembly line.
Table 6 shows the simulation results according to three different speed levels of collaborative
robots. The results, as in Table 5, show the values of job costs according to the number of assembled products and associated to workplace utilization rate. As we can see, at 60 % of the robot's
speed, the bottleneck in the assembly line workplace is already eliminated, in which case the collaborative worker and the robot are equally utilized (CWas1: 92.2 %, CR1: 84.14 % and CR2: 85.45 %).
The results of utilization rate prove a consistency of other manual assembly workplaces, which,
however, approach the maximum capacity according to the results. Given the value of the cost per
piece, we see a huge reduction in the cost of collaborative compared to manual assembly workplaces. Reducing costs is essential, as two collaborative robots represent significantly lower costs
than one additional worker.
As the speed of the robots increases, their occupancy decreases (robots have more capacity to
be used), but this does not significantly affect the rest of the assembly line workplaces, as the operator needs his/her time to properly assemble the parts on the collaborative robot. The cost of a
collaborative workplace does not change, at all different speeds, the collaborative workplace enables the production of all available semi-finished products to be assembled. Based on the results,
we can conclude that the production capacities are increased but the other workplaces’ capacity
is limited, potentially appearing new production line bottlenecks.
WP type
Cost (€)
Utilization (%)

Mas1
939.39
100

Table 5 Simulation model manual assembly line results
Mas2
Mas3
Mas4
Mas5
Mas6
Mas7
939.23 939.06 939.06 938.74 938.57 938.41
98.06
94.2
92.26
99.93
97.99
99.9

Mas8
903.48
99.88

Table 6 Simulation model collaborative workplace assembly line results

CR1 and CR2 speed and acceleration 60%
WP type
Mas1
Mas2
Mas3
Mas4
Cost (€)
939.39
939.23
939.06
939.06
Utilization (%)
100
98.06
94.2
92.26
CR1 and CR2 speed and acceleration 80%
WP type
Mas1
Mas2
Mas3
Mas4
Cost (€)
939.39
939.23
939.06
939.06
Utilization (%)
100
98.06
94.2
92.26
CR1 and CR2 speed and acceleration 100%
WP type
Mas1
Mas2
Mas3
Mas4
Cost (€)
939.39
939.23
939.06
938.9
Utilization (%)
100
98.06
94.2
92.26

Mas5
938.9
99.93

Mas5
938.74
99.93
Mas5
938.74
99.93

Mas9
903.31
99.86

Average CWas1 processing time 88.36 s
Mas7
CWas1
CR1
CR2
938.41
938.244
29.78 30.25
99.9
92.2
84.14 85.45
Average CWas1 processing time 73.04 s
Mas6
Mas7
CWas1
CR1
CR2
938.57
938.41
938.24
29.78 30.27
97.99
99.9
92.2
69.55
70.7
Average CWas1 processing time 62.31 s
Mas6
Mas7
CWas1
CR1
CR2
938.57
938.41
938.24
29.78 30.28
97.99
99.9
92.2
59.33 60.33
Mas6
938.57
97.99

Mas10
903.15
88.75
Mas10
937.26
92.1
Mas10
937.75
92.14
Mas10
937.91
92.16

Table 7 and Fig. 6 show the comparative average simulation results on which we find that the
assembly line total cost in comparison with manual assembly workplace costs and the introduction of one collaborative workplace are reduced by 8.34 %. The reduction of the average workplaces utilization is minor, as collaborative robots are at any speed fully occupied. We can see that
just one collaborative robot would be too few. In this case, the worker in the collaborative workplace would have to wait a long time for the next assembly operation to be performed, that time
is significantly less justified in terms of cost and capacity than serving a pair of collaborative robots. Given the number of finished products, we can assume that the average number of finished
products increases by 3.83 %, when introducing a collaborative workplace, at this state the number of finished products approaches the theoretical capacity of the production system. The theoretical assembly line capacity is limited by the longest processing time of the individual workplace.
438

Advances in Production Engineering & Management 16(4) 2021

The impact of the collaborative workplace on the production system capacity: Simulation modelling vs. real-world …

In current state it is represented by the workplaces Mas5 and Mas7, with a processing time of 52 s.
An interesting fact is the number of unfinished products in the production system (remaining
products in system RPis), which represents the size of intermediate stocks. Considering that there
were bottlenecks in the manual assembly line, we can see that this number is reduced by as much
as 93.67 % in the introduction of the collaborative workplaces. This result demonstrates how the
elimination of bottlenecks has a positive impact on production capacity and its justification.
Fig. 7 shows a simulation model of the product assembly process and time study of needed
worker time to assembly one product. With the help of a simulation model, we can accurately
determine the phases of assembly, the needs of the worker movement and the robot operations.
The simulation model itself assumes the optimal speeds of such a collaborative workplace in corelation to the input parameters. Created simulation model assumes the assembly of one product,
which includes three semi-finished products. The assembly phase is divided into five sub-phases
(phase a – starting position, phase b – preparation of yellow and green semi-finished product,
phase c – placement of yellow semi-finished product on a semi-finished product in robot gripper,
phase d – placement of green semi-finished on a semi-finished product in robot gripper and phase
e – final worker position). The initial assembly time is represented by the variable ts = 0 s and the
final time of the worker assembly phase by the variable tf. The results prove that the simulation
model predicted the working time of the worker assembly per product it would be 2.04 s, which
is on average equivalent to 80 % of the robot speed criteria compared to the results in Table 8
where the four-worker real-world time study was performed.
Four evaluated workers have assembled nine consecutive products during the study. In Table
8, the results of the individual assembly processing times are captured between t1 and t9. Workers
assembled the product in a sitting position at three different robot speeds, unaware of the real
speed of the robot. Presented results prove that different workers, and their working abilities can
affect the assembly operation processing time, as shown in a simulation model, an average assembly processing time can be used for variety of workers. The results prove the expediency of evaluating the collaborative workplace in both real and simulation environments.
Table 7 Manual vs. collaborative workplace comparison results
AL Type
Mas
CR speed 60 %
CR speed 80 %
Total AL costs (€)
9282.4
8508.2
8508.5
Average utilization (%)
97.1
94.2
91.5
Throughputs (pcs)
5507
5714
5717
RPis (pcs)
221
14
11

CR speed 100 %
8508.5
89.7
5718
10

Fig. 6 Workplace comparison results

Advances in Production Engineering & Management 16(4) 2021

439

Ojstersek, Javernik, Buchmeister

Fig. 7 Collaborative workplace simulation model – product assembly phases

Table 8 Real-world collaborative workplace worker product assembly processing time study
Worker 1, age of 46 years
CR speed (%)
Workers assembly processing time by product (s)
t1
t2
t3
t4
t5
t6
t7
t8
60
1.956
2.056
2.008
1.900
2.536
2.264
3.122
2.082
80
1.898
2.184
2.124
1.982
1.084
2.076
1.830
1.922
100
2.044
2.192
2.012
1.958
2.460
1.940
2.338
1.754
Worker 2, age of 27 years
CR speed (%)
Workers assembly processing time by product (s)
t1
t2
t3
t4
t5
t6
t7
t8
60
2.262
1.496
1.700
1.684
3.506
1.594
1.624
1.628
80
1.984
2.672
1.602
1.660
1.558
1.860
1.862
2.066
100
2.056
4.380
1.960
1.722
1.914
1.884
1.449
2.178
Worker 3, age of 29 years
CR speed (%)
Workers assembly processing time by product (s)
t1
t2
t3
t4
t5
t6
t7
t8
60
1.674
1.474
1.358
1.254
1.230
1.454
1.118
1.370
80
1.416
1.460
1.366
1.330
1.350
1.758
1.306
1.454
100
1.400
1.314
1.414
1.406
1.468
1.226
1.422
1.448
Worker 4, age of 25 years
CR speed (%)
Workers assembly processing time by product (s)
t1
t2
t3
t4
t5
t6
t7
t8
60
1.406
1.922
1.242
1.358
1.366
1.240
1.594
1.380
80
1.126
1.552
1.196
1.016
1.698
1.152
1.362
1.336
100
1.770
1.566
1.644
1.106
1.682
1.658
1.478
1.222

t9
2.616
1.944
2.356
t9
1.966
1.824
2.048
t9
1.184
1.288
1.118
t9
1.700
1.164
1.222

The obtained results prove the expediency of introducing collaborative workplaces in the positions of manual workplaces with limited capacities. The positive impact of collaborative workplaces is reflected in the entire production system capacity increase.
Presented simulation results of manual workplaces prove that they can identify bottlenecks in
the production system, which need to be eliminated to achieve higher production capacities. Described graphical and numerical results accurately describe the place where the introduction of a
collaborative workplaces is appropriate. In the present case, this is the Mas8 and Mas9 workplaces,
where the workplaces processing time is equal to the line takt time.
With the help of simulation modelling, we have introduced a collaborative workplace to this
assembly line station, where one worker serves two collaborative robots. The collaborative robot
operates in three different modes of speed and acceleration. Based on the results, we find that the
correct setting of the speed of the collaborative robot is key to achieving full utilization of capacities of the collaborative workplace. It should be noted that exceeding the optimal speed of a collaborating robot may have a negative impact on the worker, as excessive speed and acceleration
cause discomfort to the worker and longer waiting times for the robot to proceed with the next
440

Advances in Production Engineering & Management 16(4) 2021

The impact of the collaborative workplace on the production system capacity: Simulation modelling vs. real-world …

operation. At too high robot speeds and inability to achieve shorter assembly times on the side of
the worker, congestion can occur due to poorly performed work of the worker. The correct choice
of robot speed and the corresponding optimal process time of robot service is crucial, as evidenced by the simulation results of the collaborative workplace impact on the production system,
where we see that the increasing robot speed beyond the robot service limit has no positive effect
on the collaborative workplace production system. In general, we can see that elimination bottleneck in the manual assembly workstation can be eliminated by introducing collaborative workplace. In the evaluated case the costs of workplaces of entire production system have reduced by
8.34 %, the number of finished products has increased by 3.83 %, elimination of production system bottleneck decreased the remaining product in system by 93.67 %.
A detailed time study of the collaborative workplace confirms that all workers have an associated work rhythm that is not necessarily always the same for all workers. Since, we are talking
about a collaborative workplace, where the robot cooperates directly with the workers, adjusting
the processing time of the collaborative operation makes sense if this time is within the estimated
time of the workplace, and it does not negatively affect the rest of the system utilization. It should
be added that each worker has his own preferences regarding of the assembly position, both the
worker and the robot (ergonomics and positions studies). Different workers feel more comfortable at different robot speeds. It makes sense to take all technological and sociological influences
into account as much as possible when planning collaborative workplaces, thus ensuring maximum production system capacities.

5. Conclusion

In our research work, we have focused on presenting the impact of collaborative workplaces on
the entire production process capacity, which is positive with the presented results. We presented
various simulation models, both manual assembly workplaces, and the introduction and impact
of collaborative workplaces on production capacity. A detailed time study of the assembly time
impact of both the real-world collaborative workplace and the simulation model was presented.
The presented results showed a positive degree of correlations and the specificity of the use of
both approaches to achieve effective capacity planning. Of course, the results and findings, along
with positive answers to the initial research question, raised many questions about how to optimally construct and prepare a collaborative workplace that could fully utilize both worker and
robot capacities and effectively consider both technological and sociological aspects. In the future
research work, we will focus on a detailed study of the technological and sociological aspects of
collaborative workplaces and their correlation. Even though the presented research work deals
with assembly line production, collaborative workplaces, with their great flexibility, can be used
in different types of production at different workplaces. However, as can be seen from the results,
their justification in relation to capacity utilization needs to be studied in detail in future.

Acknowledgement

The authors gratefully acknowledge the support of the Slovenian Research Agency (ARRS), Research Core Funding No.
P2-0190. The authors acknowledge the use of research equipment system for development and testing cognitive production approaches in industry 4.0: Collaborative robots with equipment and Sensors, hardware and software for ergonomic analysis of a collaborative workplace, procured within the project "Upgrading national research infrastructures – RIUM", which was co-financed by the Republic of Slovenia, the Ministry of Education, Science and Sport and the
European Union from the European Regional Development Fund.

References
[1]
[2]

Moniz, A.B., Krings, B.-J. (2016). Robots working with humans or humans working with robots? Searching for
social dimensions in new human-robot interaction in industry, Societies, Vol. 6, No. 3, Article No. 23, doi:
10.3390/soc6030023.
Michalos, G., Makris, S., Tsarouchi, P., Guasch, T., Kontovrakis, D., Chryssolouris, G. (2015). Design considerations
for safe human-robot collaborative workplaces, Procedia CIRP, Vol. 37, 248-253, doi: 10.1016/j.procir.2015.
08.014.

Advances in Production Engineering & Management 16(4) 2021

441

Ojstersek, Javernik, Buchmeister
[3]
[4]
[5]
[6]
[7]
[8]
[9]

[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]

442

Nahavandi, S. (2019). Industry 5.0—A human-centric solution, Sustainability, Vol. 11, No. 16, Article No. 4371, doi:
10.3390/su11164371.
Özdemir, V., Hekim, N. (2018). Birth of Industry 5.0: Making sense of big data with artificial intelligence,“The
Internet of Things” and next-generation technology policy, OMICS: A Journal of Integrative Biology, Vol. 22, No. 1,
65-76, doi: 10.1089/omi.2017.0194.
Gualtieri, L., Rauch, E., Vidoni, R. (2021). Emerging research fields in safety and ergonomics in industrial
collaborative robotics: A systematic literature review, Robotics and Computer-Integrated Manufacturing, Vol. 67,
Article No. 101998, doi: 10.1016/j.rcim.2020.101998.
Leber, M., Bastič, M., Moody, L., Schmidt Krajnc, M. (2018). A study of the impact of ergonomically designed
workplaces on employee productivity, Advances in Production Engineering & Management, Vol. 13, No. 1, 107-117,
doi: 10.14743/apem2018.1.277.
Mateus, J.E.C., Claeys, D., Limère, V., Cottyn, J., Aghezzaf, E.-H. (2019). Ergonomic and performance factors for
Human-robot collaborative workplace design and evaluation, IFAC-PapersOnLine, Vol. 52, No. 13, 2550-2555, doi:
10.1016/j.ifacol.2019.11.590.
Lasota, A.M. (2020). A new approach to ergonomic physical risk evaluation in multi-purpose workplaces, Tehnički
Vjesnik – Technical Gazette, Vol. 27, No. 2, 467-474, doi: 10.17559/TV-20180312131319.
Ojstersek, R., Acko, B., Buchmeister, B. (2020). Simulation study of a flexible manufacturing system regarding
sustainability, International Journal of Simulation Modelling, Vol. 19, No. 1, 65-76, doi: 10.2507/ijsimm19-1-502.
Ojstersek, R., Buchmeister, B. (2020). Simulation modeling approach for collaborative workplaces’ assessment in
sustainable manufacturing, Sustainability, Vol. 12, No. 10, Article No. 4103, 18 pages, doi: 10.3390/su12104103.
Nikolakis, N., Maratos, V., Makris, S. (2019). A cyber physical system (CPS) approach for safe human-robot
collaboration in a shared workplace, Robotics and Computer-Integrated Manufacturing, Vol. 56, 233-243, doi:
10.1016/j.rcim.2018.10.003.
Koppenborg, M., Nickel, P., Naber, B., Lungfiel, A., Huelke, M. (2017). Effects of movement speed and predictability
in human-robot collaboration, Human Factors and Ergonomics in Manufacturing & Service Industries, Vol. 27, No.
4, 197-209, doi: 10.1002/hfm.20703.
Wang, Y., Cen, H.J., Yang, O. (2018). Optimal configuration for workshop manufacturing system under dual
resource constraints, International Journal of Simulation Modelling, Vol. 17, No. 1, 180-189, doi: 10.2507/ijsimm17
(1)co5.
Gualtieri, L., Rauch, E., Vidoni, R. (2021). Methodology for the definition of the optimal assembly cycle and
calculation of the optimized assembly cycle time in human-robot collaborative assembly, The International Journal
of Advanced Manufacturing Technology, Vol. 113, No. 7, 2369-2384, doi: 10.1007/s00170-021-06653-y.
Ojstersek, R., Tang, M., Buchmeister, B. (2020). Due date optimization in multi-objective scheduling of flexible job
shop production, Advances in Production Engineering & Management, Vol. 15, No. 4, 481-492, doi: 10.14743/apem
2020.4.380.
Trstenjak, M., Ćosić, P., Antolić, D. (2019). Workpiece classification criteria in automated process planning,
Tehnički Vjesnik – Technical Gazette, Vol. 26, No. 1, 256-262, doi: 10.17559/TV-20180215105405.
Zheng, Z.L., Bao, X. (2019). The investment strategy and capacity portfolio optimization in the supply chain with
spillover effect based on artificial fish swarm algorithm, Advances in Production Engineering & Management, Vol.
14, No. 2, 239-250, doi: 10.14743/apem2019.2.325.
Katić, I., Berber, N., Slavić, A., Ivanišević, A. (2020). The relations between investment in employees’ development
and organizational productivity and service quality, Tehnički Vjesnik – Technical Gazette, Vol. 27, No. 4, 10771083, doi: 10.17559/TV-20181121101314.
Kunpeng, Y., Jiafu, S., Hui, H. (2017). Simulation of collaborative product development knowledge diffusion using
a new cellular automata approach, Advances in Production Engineering & Management, Vol. 12, No. 3, 265-273,
doi: 10.14743/apem2017.3.257.
Fechter, M., Seeber, C., Chen, S. (2018). Integrated process planning and resource allocation for collaborative robot
workplace design, Procedia CIRP, Vol. 72, 39-44, doi: 10.1016/j.procir.2018.03.179.
Rega, A., Vitolo, F., Di Marino, C., Patalano, S. (2021). A knowledge-based approach to the layout optimization of
human–robot collaborative workplace, International Journal on Interactive Design and Manufacturing, Vol. 15, No.
1, 133-135, doi: 10.1007/s12008-020-00742-0.
Mateus, J.C., Claeys, D., Limère, V., Cottyn, J., Aghezzaf, E.-H. (2019). A structured methodology for the design of a
human-robot collaborative assembly workplace, The International Journal of Advanced Manufacturing Technology,
Vol. 102, No. 5, 2663-2681, doi: 10.1007/s00170-019-03356-3.

Advances in Production Engineering & Management 16(4) 2021