15th International Conference on Industrial Manufacturing and Metallurgy IOP Publishing
IOP Conf. Series: Materials Science and Engineering 966 (2020) 012091 doi:10.1088/1757-899X/966/1/012091
Organizational factors for manufacturing defects reduction in
small batch production
I V Ershova1, 2, S G Baranchikova1, E V Cherepanova1 and N N Chistyakov1
1
Institute of New Materials and Technologies, Ural Federal University named after
the first President of Russia B N Yeltsin, 620002, 19, Mira street, Yekaterinburg,
Russia
E-mail: 2 [email protected]
Abstract. The study is devoted to the defects organizational causes analysis in the machine
building products production process. The companies with the small batch production were
taken into consideration. Factors related to the time norms fulfilling percentage, the work
rhythm, the consolidation of operations at one workplace, the new technologies introduction
and equipment failures were chosen. A preliminary factors selection according to independence
criteria allowed us to build a reliable regression model for the defects rate calculating. The
main factors influencing the defects rate in small batch production are the time norms fulfilling
percentage, the rate of consolidation of operations at one workplace, and the number of new
technological processes introduced. This allowed us to offer recommendations on the
machining sites work organization. The final operations and productswhich are made from
expensive metals should be carried out at workplaces with a low rate of operations
consolidation. Also, it is important that the wage system should not contribute to over-
fulfillment of norms. The maintenance and repair system should prevent malfunctions during
operations fulfilment.
1. Introduction
Product quality management concepts change according to the production development. The
traditional quality assessment methods are based on large statistical samples of indicators. Application
of Shuhart control cards is described, for example, in [1]. In [2] the shortcomings of Shuhart maps are
identified and improvements for this method are offered. The statistical concept of ‘six sigma’ is
widely used to improve the processes quality and to identify the reject causes [3].
This does not applicable to small-batch manufacturing and customer-oriented production. This
conclusion is made in the article of [4] and confirmed by Russian investigators [5]. In the context of
uncertainty in small-batch manufacturing, it is difficult to determine the reject causes from the product
sample. The authors [6] propose to use a combination of statistical and Bayesian approaches. This
requires the use of the aprioristic judgment on the reject reasons.
The reject reasons can be technical, organizational and connected with human resources. Technical
reasons related to equipment and tool deterioration are quite widely considered (for example, in [7])
and are not the subject of the research. The human factor is also studied in detail. Most often, the
influence of the human factor is connected with the workers’ motivation [8]. Other authors consider
the relationship between the quantity of defective products, caused by the workers’ errors, and the
complexity of design and technology [9]. This is relevant for labour-consuming processes such as
assembly. These factors are also important for metalworking but to a lesser extent.
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
15th International Conference on Industrial Manufacturing and Metallurgy IOP Publishing
IOP Conf. Series: Materials Science and Engineering 966 (2020) 012091 doi:10.1088/1757-899X/966/1/012091
The organizational factors are in the greatest interest of researchers. In [10,4] the products quality
is considered as a consequence of the industrial system construction quality, including production
logistics, maintenance, management and control methods. This is a comprehensive approach for the
entire production system. The quality issues of raw materials and components, which are also related
to the control factors, go beyond the scope of the production process and are not considered in the
work. They are discussed in other investigations, for example [11].
The purpose of the investigation is to determine the organizational factors of the reject during the
metalworking operations at a particular workplace in small-batch manufacturing.
2. Materials and methods
The investigation is based on actual operation data of the mechanical processing workshop during the
year. The main method of research was regression statistics. The choice of the reject factors was based
on an analysis of researchers' work on the subject and the expert assessments of the reject.
The main factors selected for the analysis are:
• Index of labour performance (Х1);
• Amount of the overtime work in the workplace per month (in hours) (X2);
• Amount of new technology in the workplace, previously not performed here (X3);
• Index of the fixed operations (the amount of different operations completed at the workplace)
(X4);
• Number of equipment malfunctions (X5).
The investigated function (Y) is the share of reject in the total volume of production.
The data for analysis are shown in Table 1.
Table 1. Data for analysis.
Month Y Х1 Х2 Х3 Х4 Х5
1 3.9 1.071 8 1 23 0
2 3.7 1.039 0 0 24 0
3 3.8 1.055 0 0 24 0
4 4.3 1.108 2 0 26 0
5 6.5 1.203 8 1 34 1
6 7.1 1.222 10 2 37 1
7 8.2 1.251 16 1 43 2
8 6.7 1.211 12 0 39 0
9 7.2 1.232 16 2 41 1
10 8.5 1.267 16 2 45 2
11 4.4 1.112 0 0 26 0
12 9.1 1.327 20 5 49 2
The importance of the ‘overtime accounting’ factor is noted in [12] as a part of the optimal
production program model creation. The need to take into consideration the change in technology is
described, for example, in [13, 14]. The remaining factors are selected on the basis on the expert
estimates of managers.
The data was processed using the Microsoft Excel.
3. Results and discussion
On the first stage of the investigation, a multi-factor regression model was used. The influence of all 5
factors was taken into consideration in the model. This model shows a high determination coefficient
(R2 = 0.99). But the statistical significance coefficients evaluation of the multi-factor model showed
that two factors from the considered set (Х2 and Х3) have an unacceptably high level of P-values
(significantly higher than the normative level of 0.05) and therefore are not statistically significant.
The second stage of the investigation, these factors (X2, X3) were excluded from the multi-factor
model.
2
15th International Conference on Industrial Manufacturing and Metallurgy IOP Publishing
IOP Conf. Series: Materials Science and Engineering 966 (2020) 012091 doi:10.1088/1757-899X/966/1/012091
As a result, the model became:
Y = –7.1999+8.2358X1+0.0972X4+0.4146X5 (1)
Model coefficients are statistically significant. The determination factor has a high value. Fisher 's
test value is improving.
The model reflects the share reject dependence from: the index of labour performance, the Index of
the fixed operations and number of equipment malfunctions. At the same time, if independent
parameters are close to 1 (or 0 in the case of equipment failures), then the reject share according to this
model will be extremely small. Therefore, this formula is not effective when used in mass and large-
scale production. Because in these types of production fixed operations index does not change from
month to month, and the labour performance index is more stable, as most actions of the worker are
automated. The formula could not also be applied in single production because in single production
usually each operation performs for the first time and once. It can be called a ‘permanent
implementation’. In addition, time standards in single production are calculated by extended methods
and often do not correspond to reality. The model can be used only in small-scale and medium-scale
production.
Using this model, the following recommendations for workplace production can be formulated:
1. Final operations have to be carried out at workplaces with a small index of the fixed operations.
2. The products from expensive materials have to be produced on the workplaces with a small
index of the fixed operations.
3. The remuneration system should not encourage over-performance.
4. The repair, maintenance and operation of the equipment have to prevent problems during
operations.
Thus, it is more efficient to have one pre-operation equipment working in 2–3 shifts and several
workplaces for final operations working in one shift.
Similar situation is with material. It is more efficient to have one equipment, rolling black metals,
working in 2–3 shifts, and several workplaces, rolling non-ferrous metals and alloys, working in one
shift.
At the same time, both the cost of the material and the processing stage must be taken into account.
During the production reorganizing, it is necessary to factor into the reduction of costs and the
constant quality improvement, which are necessary in the strict competition conditions. Therefore,
special creation of several workplaces for final operations may be economically unprofitable. This
requires additional calculations.
The remuneration system has to reduce the share of reject and the corresponding costs associated
with the defective products production. According to the mathematical model, the proportion of reject
depends on the rate of standard time implementation. Accordingly, the remuneration system has not to
encourage the overtime work. The Russian enterprises in metal-working industries very often use the
piece rate system. It is necessary to provide transition to time system with normalized task.
The share of reject is affected by equipment problems. planning and preventive systems of
equipment repair and maintenance can be used for equipment problems preventing [11]. In order to
update the equipment maintenance and repair system, it is necessary to develop the scheduled repair
system for modern CNC machines, as well as to create internal standards for maintenance and
operation of the workplaces. The standards can be based on the 5S system [12].
Since the first three recommendations are limited in application, more attention should be paid to
the fourth recommendation – preventing equipment problems by improving the equipment repair and
maintenance system.
4. Conclusion
A machine-building enterprise wishing to reduce the reject should pay attention to the system of
maintenance, repair and operation of technological equipment; to the basis of the defect-free labor
stimulation system and the organizational principles of products movement along the technological
3
15th International Conference on Industrial Manufacturing and Metallurgy IOP Publishing
IOP Conf. Series: Materials Science and Engineering 966 (2020) 012091 doi:10.1088/1757-899X/966/1/012091
route. That would help the enterprise successfully compete and introduce innovative technologies and
equipment.
References
[1] Morozova A E, Yurakov N S and Yurakova T G 2018 Application of Shuhart control cards for
statistical control of parts quality Modern materials, techniques and technologies 6 (21) pp
68–72
[2] Klachkin V N and Kravtsov Yu A 2016 Detection of violations in multidimensional statistical
control of technological process Software products and systems 3 pp 192–7
[3] Boccarusso L, Fazio D D , Durante M, Langella A and Capece Minutolo F M 2019 CFRPs
drilling: Comparison among holes produced by different drilling strategies Procedia CIRP
79 рp 325–30
[4] Colledani M, Tolio T, Fischer A, Iung B, Lanza G, Schmitt R and Váncza J 2014 Design and
management of manufacturing systems for production quality CIRP Annals - Manufacturing
Technology 63 (2) pp 773–79
[5] Iudin S V 2019 Problems of quality control and reliability of small-scale products of weapons
and military equipments News of Tula State University. Technical science 12 pp 88–92
[6] Iudin S V, Protasiev V B, Podkopayev R Y and Iudin A S 2018 Methodology of calculation of
information plans of statistical acceptance control on the basis of Bayesian approach Co-
temporary knowledge-intensive technologies 11 pp 90–4
[7] Pearn W L and Hsu Y C 2007 Optimal tool replacement for processes with low fraction
defective European Journal of Operational Research 180 (3) pp 1116–29
[8] Gorlenko O A and Mozhayeva T P 2017 Quality assurance of machine-building production on
the basis of human resources management News of Volga STU 9 (204) pp 118–20
[9] Su Q, Liu L and Whitney D E 2010 A systematic study of the prediction model for operator-
induced assembly defects based on assembly complexity factors IEEE Transactions on
Systems, Man, and Cybernetics Part A: Systems and Humans 40 (1) pp 107–20
[10] Inman R R, Blumenfeld D E, Huang N, Li J and Le J 2013 Survey of recent advances on the
interface between production system design and quality IIE Transactions (Institute of
Industrial Engineers) 45 (6) pp 557–74
[11] Mitin A 2013 Factors influencing quality in mechanical engineering Standards and quality 3 pp
66-9
[12] Chiu S W, Wu H Y, Chiu Y –S P and Hwang M H Exploration of finite production rate model
with overtime and rework of nonconforming products Journal of King Saud University -
Engineering Sciences 30 (3) pp 224–31
[13] Zhilina S, Kapitanov N, Osochenko O, Kochedykova I and Vazhdayeva A 2016 The dedicated
algorithms for design and technological preproduction and information support of production
process News of the Samara Scientific Center of the Russian Academy of Sciences 4 (3) pp
638–44
[14] Edvabnik V G, Fartishev Y M, Gautsel A K and Kuznetsov M M 2013 The assessment of the
stability of production processes in the conditions of the single and small-batch production
INTEREKSPO GEO-SIBIR 5 pp 195–200
