Advanced Logistic Systems – Theory and Practice, Vol. 16, No. 2 (2022), pp. 54-59.
https://doi.org/10.32971/als.2022.013
THE USABILITY OF CRAFT LAYOUT DESIGN METHOD WITH THE
EXAMINATION OF SIMPLIFICATIONS
FRUZSINA BÓSA 1 ˗ PÉTER VERES 2 ˗ AZIZBEK KYDYKOV 3
Abstract: Logistics optimization has been gaining attention continuously by companies in recent
decades, but due to the current situation, such as the pandemic, recession, and supply chain
reorganizations, manufacturers have to resort to more serious tools. One of these is layout redesign,
which fundamentally changes a material flow process. One of the most well-known methods is the
CRAFT method, which uses two very serious simplifications. In this work, we will examine how
much difference the simplifications make in a system close to reality, and can we safely use the
CRAFT method.
Keywords: Layout planning, Layout redesign, CRAFT method
1. INTRODUCTION
Within the life of manufacturing companies, it is an inevitable event that some machines or
even entire production lines have to be reorganized from time to time. Ideally the
companies innovate, the technology improves, they have more machines, the product range
expand or change, or the production volume of existing products is increased. Logistics
processes are primarily meant to serve production processes; therefore logistics systems
must also keep up with the changing and developing technology [1], [2].
There are not always positive reasons behind the reorganization of plants. The past few
years have caused enormous problems for many companies due to the pandemic situation
[3], [4]. The supply chains were weakened, even several suppliers could not fulfil their
orders. Due to the weakened market and problems caused by the supply chain, it was not
possible or necessary to manufacture some products, which had a great impact on the
companies. Downsizing and reorganization came forward, and alternatives appeared on the
markets [5]. Even in the disadvantageous situation thus created, it was necessary to rethink
the function of the factories that had been worked until then, and one way to do this is to
transform the layout of the whole production and/or storage system. If production lines fail
due to a lack of labor or because they cannot produce due to lack of materials, and the
situation cannot be remedied in the long term, it may be worthwhile to consider and
implement other arrangements [6], [7].
In this work, we will use a simple example to test the efficiency of one of the best-
known layout organizing method: the CRAFT method. After a little introduction, we want
to answer if the simplification of the method is affecting the result and, if so, how much
deviation can we measure. If the deviation is not significant, the CRAFT method can be
safely used in its original form for layout planning.
1
BSc student., University of Miskolc, Institute of Logistics
fruzsibosa@gmail.com
2
PhD., University of Miskolc, Institute of Logistics
peter.veres@uni-miskolc.hu
3
PhD., Kyrgyz State Technical University, Institute of Logistics
azizbek.kydykov@gmail.com
The usability of CRAFT layout design method with the examination of simplifications 55
2. DESCRIPTION OF THE CRAFT METHOD AND THE TEST LAYOUT
We usually encounter the installation problem of objects with a given floor area and shape
as an internal plant layout task. The most often used solution method is the CRAFT
method, which classifies as a heuristic algorithm, meaning that there are no specific
mathematical relationships on the basis of which we can find an optimal solution. Based on
a guiding process, the possible versions must be examined, and then a solution must be
chosen from among them, which may not be the optimal solution, but it is suitable for us
given these circumstances [2], [6], [8].
In order to apply the method, we have to lay down the basic rules, which are as follows:
• The shape of the floor area of the workshop, in which the layout is carried out.
This cannot be changed or rotated, as this does not happen in reality either.
• The shape of the machines cannot change either, but they can be rotated, just like
in reality.
• The base area of the machines is taken into account, and the centre of gravity is
assumed to be the source and sink. This is the simplification we wanted to test.
• The movement between the machines takes place according to the Cartesian
coordinate system. In reality most of the routes in a factory are designed this
way, so this can generally be considered realistic [6], [9], [10].
First of all, we record the position of the starting layout, based on this the route matrix
between the individual points can be calculated [2], [6]:
(1)
where:
• xs is the x coordinate of the centre of gravity of the object
• ys is the y coordinate of the centre of gravity of the object
• i,j are the objects
The material flow matrix is considered as given, and it can be used to calculate the material
handling work [2],[6]:
(2)
After this, we have to look for all the possible exchanges, so that 1 object can be exchanged
with only 1 other object or set of objects at a time. We have to calculate every distance
matrix for all the possible exchanges. After we got the new material flow values, we pick
the smallest, and we fix this exchange. Now we repeat the previous set of instructions [8]:
• define the new set of exchanges of the objects (be careful not to get back a
previous configuration),
• calculate the new distance matrices,
• calculate the new material flow matrices,
• pick the lowest value of material flow number,
• fix that exchange of objects as the new starting position.
56 Fruzsina Bósa ˗ Péter Veres ˗ Azizbek Kydykov
We repeat these instructions until we have no more possible exchange, the material flow
numbers only get bigger (we pick the lowest number in all branches), or we think we find
an optimal solution [6],[8].
2.1. The testing layout
We create a simple, but somewhat realistic starting layout. There is no new equipment we
want to insert, we only need to optimize the current layout. The system consists of 6
machines (1-6) that can take 3 different shapes. From a raw material warehouse (Ra), a
finished product warehouse (Rk), an empty area (0) where objects can be placed freely, and
a closed area (black tile) where no objects can be installed. The red dots are the exit points
of the objects and the green dots are the enter points. The grey dots are the grid points, they
only here for visuals, and ease the computation. A grid in this layout is 4x4m=16m2.The
staring layout can be seen in Figure 1.
Figure 1. Starting layout of testproblem
The next important data is the material flow matrix. This doesn’t change over the whole
testing, but in the CRAFT method the enter and exit points are on top of each other, which
is in the centre of the objects. The material flow table can be seen in Table I.
Table I.
Material flow matrix, that represents the connection between the test objects
The usability of CRAFT layout design method with the examination of simplifications 57
3. TESTING OF THE CRAFT METHOD
We can represent the whole layout as coordinate points. In the CRAFT method, we
calculate the distance between two objects, by simple subtracting the X coordinates of the
two objects in absolute value, and do this for the Y coordinate also, and then add them
together. This eliminates the physical routes that connects them and ignore any other object
this route goes through. In a more realistic way, we can only use roads or routes between
objects. Table II shows the CRAFT method’s simplified distance matrix where object 7 and
8 are the raw material and finish product warehouses, and Table III shows the real distance
matrix, where all grid points are shown and calculated by Dijkstra algorithm [11], [12].
Table II.
Distance matrix in CRAFT method
Table III.
Distance matrix with realistic calculation
58 Fruzsina Bósa ˗ Péter Veres ˗ Azizbek Kydykov
For every step, and exchange of objects we have to calculate or modify the previous two
tables. This is very time-consuming for the realistic method, even for this simple problem.
Since the entire calculation of the two methods would take nearly 100 pages, we will only
show the final solution and the tree of exchanges. The total tree of CRAFT method can be
seen in Figure 2. In the nodes of the tree (squares) in the first row we can see the name or
the exchanges of the node, and in the second row there are the value of total material
handling work, where the lower is better.
Figure 2. Tree of exchanges of CRAFT Method
When we calculate the values of the realistic method, there are several sub-solutions,
because, in the CRAFT method, the orientation of object doesn’t matter, because we work
with their centre, but in the realistic method most of the time there are 2 or even 4 rotations
can be possible for every object. We can’t eliminate this problem automatically, because it
creates numerous similar layouts, so we address it by manually picking the most sensible
orientations for the objects. With this the trees are completely identical only the values
changes. Table IV contains the values compared between the two methods.
Table IV.
Deviation between the two methods
CRAFT Method Realistic method Deviation
Exchange Value of material Improvement from Value of material Improvement from between
name handling work (WL) Start layout (%) handling work (WL) Start layout(%) methods
Start layout 1980 0,00% 2080 0,00% 5,05%
1,2 excange 1696 14,34% 1764 15,19% 4,01%
14_5,6 1560 21,21% 1626 21,83% 4,23%
12_3,4 1592 19,60% 1688 18,85% 6,03%
123_5,6 1392 29,70% 1488 28,46% 6,90%
In Table IV we examined 5 important layouts (they are highlighted in Fig. 2), both the
improvements are similar, and the deviation between methods are in a very narrow value
limits (4…7%). With this finding we can assume, that the simplifications of CRAFT
method is acceptable, because we can find clear quasi-linear connection between the
The usability of CRAFT layout design method with the examination of simplifications 59
models. With this knowledge we can recommend the usage of the CRAFT method for
layout planning.
4. SUMMARY
In this paper, we proved that the CRAFT method does not affect the result as much as we
originally suspected, despite the fact, that it uses simplifications: the enter and exit points of
the installable objects are in the centre, and their transport path is practically a straight line.
From the data we can see that the connection between the CRAFT and realistic method is
quasi linearly proportional, so we can safely say that the CRAFT method can be
successfully applied to layout design tasks.
REFERENCES
[1] Cselényi, J. & Illés, B. (2004). Logisztikai rendszerek I. Miskolci Egyetemi Kiadó, Miskolc
[2] Cselényi, J. & Illés, B. (2006). Anyagáramlási rendszerek tervezése és irányítása, Miskolci
Egyetemi Kiadó, Miskolc
[3] Beyza, G. et al. (2022). The logistics service providers during the COVID-19 pandemic: The
prominence and the cause-effect structure of uncertainties and risks. Computers & Industrial
Engineering 165, paper: 107950. https://doi.org/10.1016/j.cie.2022.107950
[4] Nagy, G. et al. (2021). A világjárvány hatása a globális logisztikai folyamatokra.
Multidiszciplináris Tudományok: A Miskolci Egyetem Közleménye 11(4), 42-52.
https://doi.org/10.35925/j.multi.2021.4.4
[5] Meahjohn, I. & Persad, P. (2020). The Impact of COVID-19 on Entrepreneurship Globally.
Journal of Economics and Business 3(3), https://doi.org/10.31014/aior.1992.03.03.272
[6] Bóna, K. (2018). Az üzemi belső elrendezés tervezése. Akadémiai Kiadó, Budapest,
https://doi.org/10.1556/9789634543084
[7] Mahendra, S. (2012). Innovative practices in facility layout planning. International Journal of
Marketing, Financial Services & Management Research 1(12), ISSN 2277 3622
[8] Vivek, D. et al. (2016). Plant Layout Optimization using CRAFT and ALDEP Methodology.
Productivity Journal by National Productivity Council 57(1), 32-42. ISSN: 0032-9924
[9] Singh, S. P. & Sharma, R. R. K. (2006). A review of different approaches to the facility layout
problems. The International Journal of Advanced Manufacturing Technology 30, 425–433.
https://doi.org/10.1007/s00170-005-0087-9
[10] Vivek, A. D. & Chopade, K. (2005). Facility Layout Design by CRAFT Technique.
Proceedings of COMPUTIME, National Conference on Computational Methods in Mechanical
Engineering, Osmania University, Hyderabad
[11] Gass, S. & Fu, M. (2013). Dijkstra's Algorithm. Encyclopedia of Operations Research and
Management Science. ISBN 978-1-4419-1137-7 https://doi.org/10.1007/978-1-4419-1153-
7_200148
[12] Cormen, T. H. et al. (2001). Section 24.3: Dijkstra's algorithm. Introduction to Algorithms, MIT
Press and McGraw–Hill. 595–601. ISBN 0-262-03293-7.
Copyright of Advanced Logistic Systems: Theory & Practice is the property of University of
Miskolc and its content may not be copied or emailed to multiple sites or posted to a listserv
without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.
