Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
THE CASE STUDY OF BOTTLENECKS
IDENTIFICATION FOR PRACTICAL
IMPLEMENTATION TO THE THEORY
OF CONSTRAINTS
doi:10.2478/mape-2018-0051
Date of submission of the article to the Editor: 03/2018
Date of acceptance of the article by the Editor: 0
7
/2018
MAPE 2018, volume 1, iss ue 1, pp. 399-405
Prof. Wieslaw Urban
Patrycja Rogowska
Bialystok University of Technology, Poland
Abstract: According to the Theory of Constraints by Goldratt system bottlenecks are key to improving
productivity and profitability of the entire production system. An important issue in the practical application
of this theory is to identify bottlenecks in the system. The aim of this paper is to develop practical steps
to identify bottlenecks in the production system characterised by a homogeneous flow. The paper
includes a case study which shows six steps that allowing to specify not only the first bottleneck of the
system but to develop a comprehensive plan for removing bottlenecks.
Keywords: Theory of Constraints, bottlenecks identification, production flow
1. INTRODUCTION
The search for bottlenecks and focusing on relieving them is a leitmotif throughout the Theory
of Constraints by Goldratt (1990). Bottlenecks limit the capacity and throughput of the
production system, resulting in, among others, stagnant production, local accumulation of
stocks, and above all reduced productivity of the system as a whole. The identification of and
clearing bottlenecks are the basis for improvement of production and economic results
achieved by a company. But on the other hand, acting according to the Theory of Constraints
involves the need to solve a number of practical problems. One of them is the identification of
bottlenecks in the production system.
The purpose of this paper is to determine the procedure to be followed to identify a bottleneck
in the production system, as well as subsequent bottlenecks that should be subjected to
improvements in line with the approach proposed by the Theory of Constraints by Goldratt.
The research was conducted using a case study. The object of the study is the system of
production of building materials characterised by a homogeneous flow, i.e. variability of
product range that practically does not affect the characteristics of the product flow through
the manufacturing system. The identified steps to be used to define bottlenecks are also
important in utilitarian terms because there are few studies showing the practical applications
of the Theory of Constraints in relation to different configurations and characteristics of the
production system.
2. THEORY OF CONSTRAINTS AS THE BASIS OF MANUFACTURING SYSTEM
DEVELOPMENT
The guarantee of success of the organisation is the flexible adaptation of production to
changing market and customer requirements (Dulin et al., 2016). To enable a company to
quickly and fully meet the needs of potential clients it must have high efficiency with maximum
production flexibility (Jagodzinski, 2016). But it is not so easy, because variability of
Multidisciplinary Aspects of Production Engineering – MAPE vol. 1, iss. 1, 2018 400
production, scarcity of resources such as technology or personnel may constitute a bottleneck,
which limits the capacity of the production system (Lei and Li, 2017).
In the production process there are two types of bottlenecks differing in the level of their
utilisation. The first type is exceeding the level of utilisation of performance limits of the
constraint. An area of this type reduces the flow of goods and materials generating
performance limits in the whole production system. While the second type is 100% utilisation
of the resource capacity. Such a situation poses an extreme threat to the efficiency of
production processes. This area has the highest degree of exploitation (high risk of failure) as
well as affects the duration of the whole production process (Kolinski and Tomkowiak, 2010).
To efficiently identify, manage and eliminate production bottlenecks, you can use the
methodology of the Theory of Constraints (ToC), whose founder is an Israeli physicist Dr.
Eliyahu M. Goldratt (Saniuk and Saniuk, 2010). This tool is widely used in a production
environment, i.e. production planning, enterprise resource management, risk management,
project management, marketing and accounting (Szatkowski, 2014). Its basic premise is that
any process or system has one constraint that determines its performance. As mentioned
earlier, this concept gives scope for identification of constraints and their appropriate
management. The first step is to exploit the maximum production capacity of a bottleneck. If
the bottleneck work schedule continues to be a barrier to the company's production capacity,
the next step is to take action aimed at the reduction or complete elimination of this constraint
(Kedzierski, 2016). Keep in mind that attempts at eliminating the constraint in one place lead
to the identification of further constraints or the occurrence of new bottlenecks. Therefore it is
important to constantly improve the manufacturing process of products (Kasemset and
Kachitvichyanukul, 2010).
According to Goldratt (1990) a constraint is everything that prevents the organisation from
achieving its goal. The authors classify constraints as follows: market constraint, capacity
constraint, political constraint, raw material constraint, logistics constraint, behavioral
constraint and administrative constraint (Okutmus et al., 2015). Among them constraints in the
production flow capacity are typical production bottlenecks . Constrains identification is a
fundamental issue for ToC when it comes to its practical implementation in a real production
system. Alsmadi et al. (2014) suggest starting from financial record analysis whilst aiming at
identifying constraints, they are recognised when the operation of the system is tied to spent
costs. Another approach to look for constraints is by simulating a system model
(Golmohammadi, 2015; Costas et al., 2015). This approach refers for to capacity in
bottlenecks and exploits a multiproduct production/logistic system, the point is how to
determine the most favourable structure of the manufactured assortment. But also, an
interesting issue is how to identify constraints/bottlenecks in a homogeneous production flow.
3. INVESTIGATED MANUFACTURING SYSTEM
The studies company is a manufacturer of building materials. The main raw materials used for
production of silicate blocks are lime, sand and water. The key machinery of the company
includes silos, mixers, reactors, autoclaves and moulding presses. The plant produces 17
kinds of products. The production process of silicate blocks is carried out in five major steps:
1. Storage and mixing of raw materials − raw materials are stored in silos which are then
subjected to stirring in an appropriate proportion. 2. Transformation of the mixture in lime
hydrate − adding water to the mixture, which is then placed in reactors where slaking occurs.
3. Formation − presses form silicate blocks at a pressure. 4. Hardening − placing semi-finished
products in an autoclave at 200°C, where recrystallisation of the mixture occurs. 5. Quality
control and packing − subjecting blocks to quality control followed by stacking on pallets. The
company operates in three shifts, six days a week. The flow of materials and semi-finished
products takes place between the various processes in an automatic or semi-automatic
manner.
401 New Trends and Ideas in Technology and Engineering
Due to the uneven use of production resources, the company decided to analyse its production
capacity. The company is interested in improving the productivity of the system as a whole.
Below are the analytical research steps carried out in the company to build an appropriate
information framework that will clearly identify a bottleneck and subsequent bottlenecks,
including the directions of work on improvement of the system.
4. BOTTLENECKS IDENTIFICATION STEPS
4.1. Identification of the flow sequence (STEP #1)
The first step taken in order to identify a bottleneck is the observation of the company's
production system and the actual measurement of cycle times in every production process.
This action will allow to order processes occurring consecutively, which are not always in line
with the business records. Process observation was made for one type of product to be
manufactured. The analysis was extended to include the necessary information, i.e. the
quantity of available production resources for parallel processes and the number of pieces of
product in one process cycle, which indicates how many product units leave the process. Table
1 presents the results of the observation of the production system.
Table 1
Identification of the sequence of the production system flow and the number of parallel
processes
Process/operation
Process/
operation cycle
time
[sec]
Number of
available
productive
resources
Number of
pieces of
product in one
cycle
Formation 9 3 5
Placing semi-finished products on hardening trolleys 37 3 20
Internal transport of hardening trolleys 257 1 480
Hardening 25403 6 4320
Unstacking 382 3 4320
Palletising 130 1 240
Securing the product 140 1 160
Transportation to the warehouse 73 1 320
In the following studies a “production unit” is composed of 5 pieces of products manufactured
in a single cycle of the moulding machine.
4.2. Investigation of unit processes (STEP #2)
The intent of this step is to examine what time is needed for each production unit to pass
through each of the previously identified processes based on the quantity of available
production resources. This analysis will allow to determine which process or production step
is the most time-consuming. Fig. 1 shows the results of this stage.
Analysing Fig. 1 it can be concluded that the most time-consuming process is product
hardening. But the base times do not take into account unproductive losses occurring in the
industry. Therefore, the next steps to identify a bottleneck should take account of all types of
waste that occur in the examined production process.
Multidisciplinary Aspects of Production Engineering – MAPE vol. 1, iss. 1, 2018 402
Fig. 1. Process cycle times
4.3. Investigation of process cycle times with reference to production losses due to
product quality (STEP #3)
As already mentioned, important factors influencing the productivity of processes are
production losses. As a result of observation of the analysed manufacturing process, the first
type of waste concerning product quality was identified. This issue affects the process of
moulding products. Approx. 15% of products on each moulding press are damaged. The factor
causing the gaps is incorrect humidity of the raw material mixture. Taking into account this
factor, the time needed for a production unit to undergo each process is shown in Fig. 2.
Fig. 2. Process cycle times with reference to product quality
As a result of taking account of losses which occur during the moulding process, the time to
produce a unit product for 3 presses increased by 0.45 seconds. The total moulding process
time for the adopted production unit is 3.45 seconds. Other process times remain unchanged.
4.4. Investigation of process cycle times with reference to machine changeover time
(STEP #4)
Another unproductive loss affecting the identification of a bottleneck in the production system
is machine changeover time. The analysis of the investigated manufacturing process showed
that machine changeover takes place on moulding presses (moulding process).
33,08 2,68
4,90
0,15
2,71
4,38
1,14
0
1
2
3
4
5
time [sec]
process
production cycle time
33,08 2,68
4,90
0,15
2,71
4,38
1,14
0,45
0
1
2
3
4
5
time [sec]
process
production cycle time cycle time with reference to product quality
403 New Trends and Ideas in Technology and Engineering
Fig. 3. Process cycle times with reference to changeover time
The changeover time factor should be considered as Step 4 to enable the identification of
production system constraints.
When analysing enterprise data it was found that there are about 9 product moulding press
changeovers a month. The changeover time is 6 hours. An average batch manufactured by
the company is 750 hardening trolleys. The results of analysis are shown in Fig. 3.
Based on the aforementioned assumptions, changeover time per production unit is 0.6
seconds. So the total moulding process time for the adopted production unit is 4.05 seconds.
Other process times remain unchanged.
4.5. Organisation of working time for individual resources (STEP #5)
Another important step is to examine the extent of utilisation of the operation time for each
resource. The company on average manufactures approx. 121 hardening trolleys containing
a total of 5,184 production units per one production shift. If you know the duration of each
stage of the production process per production unit taking into account all the factors affecting
its duration, you can easily calculate the time of unused availability of all resources. The results
are shown in Fig. 4.
Fig. 4. Process cycle times with regard to unused availability
33,08 2,68
4,9
0,15
2,71
4,38
1,14
0,45
0,60
0
1
2
3
4
5
time [sec]
process
production cycle time cycle time with reference to product quality cycle time with reference to changeover time
33,08 2,68
4,90
0,15
2,71
4,38
1,14
0,45
0,60
1,51 2,48 2,88
0,66
5,41
2,85
1,18
4,42
0
1
2
3
4
5
6
time [sec]
process
production cycle time cycle time with reference to product quality
cycle time with reference to changeover time unused availability of the process
Multidisciplinary Aspects of Production Engineering – MAPE vol. 1, iss. 1, 2018 404
The average resource availability per production unit is 5.56 seconds. Analysing the above
data it is clear that the availability of resources is not fully utilised. The most effectively utilised
process is product hardening.
4.6. Analysis of the structure of process cycle times (STEP #6)
On the basis of the information in Step 5 it can be concluded which resources/processes
generate bottlenecks in the production system. It also shows the resulting type of cycle time
components. We identified 3 key bottlenecks in the production system in question, i.e.
hardening, product protection and moulding. The ToC logic tells to sequentially relieve
bottlenecks one after another, each time with a view to improving system capacity and
increase throughput. Table 2 shows a sequence of tasks that should be introduced to improve
the production system studied. When determining the degree of utilisation of the processes,
full scalability of the production capacity of processes was assumed, which in practice does
not always coincide with actual possibilities.
Table 2
Further improvement of the production system bottlenecks
Item
No. Improvement / Process Cycle time of the system after
relieving a bottleneck
Degree of
utilisation of
processes
1. Maximum utilisation of the availability of
hardening 4.90 56.2%
2. Increasing hardening production capacity
by supplying resources 4.38 62.9%
3. Increasing securing production capacity by
adding resources 4.05 68%
4. Improving changeovers and elimination of
defects in moulding 3.08 89.4%
However, steps towards the exploitation of bottlenecks should be preceded by steps to ensure
maximum utilisation of the production capacity of the hardening process. These actions should
be adapted so that the system worked in a rhythm equal the production cycle of the process
(Improvement 1 in Table 2 above). In the next step, the bottleneck which is product hardening
should be cleared through the supply of resources. After the removal of this constraint, product
safety should be taken into account. The data in Table 1 indicate that the process is performed
by a single worker. Therefore, in order to relieve the bottleneck, the number of employees
dealing with securing finished products should be increased. The last process important due
to its time-consuming nature is moulding. To increase throughput, losses generated in this
process should be eliminated and the moulding press changeovers should be improved.
5. CONCLUSION
The identification of bottlenecks and their elimination have a significant impact on the efficiency
of the production system of the company. The first course of action was to examine how the
analysed production system actually works and the cycle times of the examined processes.
Another important factor that has a significant impact on the production system is to identify
all types of waste generated in the process. In the studied company, 3 types of waste were
identified, i.e. due to manufacturing defects, machine changeover time and improper
organisation of shift work resulting in underutilisation of the available processes. These are
typical losses so each next step (2 to 5) should be devoted to each of them individually. This
results in a graphical presentation of the time structure of individual processes, which allows
you to draw conclusions as to existing bottlenecks and the sequence of actions to be taken to
eliminate them.
The proposed production process bottleneck identification logic allows to indicate limitations
of the system and gives guidelines on how to improve the direction of the production system.
405 New Trends and Ideas in Technology and Engineering
ACKNOWLEDGEMENTS
Research have been carried out in the framework of work S/WZ/1/2015 and financed by
Ministry of Science and Higher Education from the funds for science.
REFERENCES
Alsmadi, M., Almani, A. and Khan, Z. (2014). Implementing an integrated ABC and TOC approach to
enhance decision making in a Lean context: a case study. International Journal of Quality &
Reliability Management, 31(8), pp. 906–920.
Costas, J., Ponte, B., de la Fuente, D., Pino, R. and Puche, J. (2015). Applying Goldratt’s Theory of
Constraints to reduce the Bullwhip Effect through agent-based modeling. Expert Systems with
Applications, 42, pp. 2049–2060.
Dulina, L., Mleczko, J. and Mičieta, B. (2013). Identification of Bottlenecks in the Unit Make to Order
Production. Applied Computer Science, 9(2), pp. 43–56.
Goldratt, E.M. (1990). What is This Thing Called Theory of Constraints and How Should it be
Implemented? New York: North River Press.
Golmohammadi, D. (2015). A study of scheduling under the theory of constraints. International Journal
of Production Economics, 165, pp. 38–50.
Jagodziński, J. and Ostrowski, D. (2016). The production process optimization based on the principle of
continuous improvement on the selected example. Zarządzanie, 24(2), pp. 201–214, (in: Polish).
Kasemset, C. and Kachitvichyanukul, V. (2010). Bi-level multi-objective mathematical model for job-shop
scheduling: the application of Theory of Constraints. International Journal of Production
Research, 48(20), pp. 6137–6154.
Kędzierski, P. (2016). Theory of Constraints in production planning. Logistyka Produkcji, 1(21), pp. 43–
45, (in: Polish).
Koliński, A. and Tomkowiak, A. (2010). Using the analysis of bottlenecks in production management.
Gospodarka Materiałowa i Logistyka, 9, pp. 16–21, (in: Polish).
Lei, Q. and Li, T. (2017). Identification Approach for Bottleneck Clusters in a Job Shop on Theory of
Constraints and Sensitivity Analysis. Journal of Engineering Manufacturing, 231(6), pp. 1091–
1101.
Okutmus, A., Kahveci, A. and Kartašova, J. (2015). Using theory of constraints for reaching optimal
product mix: An application in the furniture sector. Intellectual Economics, 9, pp. 138–149.
Saniuk, A. and Saniuk, S. (2010). Theory of constraints applied to resources management in micro and
small-sized enterprises. Ekonomiczne Problemy Usług, 50(585), pp. 355–363, (in: Polish).
Szatkowski, K. (2014). Modern production management. Warsaw: Wydawnictwo Naukowe PWN, (in:
Polish).
