1. INTRODUCTION
The
boost of the domestic market on the past years, leveraged by government
programs and major sport events, increased the interest in investments of
foreign countries in Brazil. In the construction machinery market, the main
investments are from Asian origin. Led by China in terms of investment (Table
1), which defines Brazil as a strategic partner, as it has 1.4 billion
inhabitants and US $ 2.4 trillion of reserves, thus having much need for food,
oil, iron ore , energy, etc. (Estado de São Paulo - February 11th, 2011).
Table 1: Recently Installed in
Brazil
|
Recently installed in
Brazil |
||
|
Companies |
Investment |
Origin |
|
San Heavy Industries |
U$ 100 Millions |
China |
|
Liu Gong Machinery |
U$ 120 Millions |
China |
|
XCMG |
U$ 1 Billion |
China |
|
Hyundai |
U$ 150 Millions |
South Korea |
|
Doosan |
U$ 35 Millions |
South Korea |
Source: Estado de São Paulo (February 11th,
2011)
On
the other hand, there was also stimulating competitiveness, since Chinese
companies have segmentation strategy of low costs. The need to reduce
manufacturing costs, to increase profitability makes companies seek,
increasingly, lean production systems. However, the decision of which method to
use and whether it will meet the needs is still a paradigm for many companies.
The
decision-making process appears as an unexplored factor in the work of
spreading the lean production model. Toyota is recognized as the worldwide
pioneer in this method, by creating a focused production system for the
continuous improvement of productivity indicators and quality. The companies
using lean manufacturing approach, in general, have a competitive advantage
over those using traditional approaches, as it has been realized by the
industry and also by the academy, not only in developed countries but also in
emerging countries (TORRES JUNIOR, 2010 ).
To
obtain a progressive removal of losses, it takes a close relationship between
customer and suppliers, whether external or internal, meaning that the focus of
a company is to become lean and healthy supply chain, through a win
win-relationship with everyone involved.
The
target company in this study is originated in Japan and has been in Brazil for
more than 40 years as part of machinery section for Construction, Mining and
Forestry. The objective of this study is to analyze the overall results of the
kaizen philosophy implementation at in-house logistics in a construction
machinery factory. Our target is the production efficiency and its influences
with the process in order to answer the following research problem: "The
application of kaizen system, in addition to the parts concentration and the
representativeness analysis of travel in the picking contribute as tools for
improving the efficiency of an assembly line?"
Eliminating
the losses in the production process, it is expected to obtain gains in the
flow of the studied assembly line, making it more efficient. The research
methodology adopted in this study is action-research, which is a type of
research with empirical basis, to solve a collective problem and that
researchers are involved in a cooperative and participative way.
It is
known that the methods and concepts used in this article can be used in any
companies that have a production line with a big volume of control items. This
work is structured as follow: The first session deals with the introduction and
the context in which the company under study is inserted. The second section
presents the theoretical basis of the topics discussed. The third section
presents the proposed method for the study.
The fourth section describes the results obtained after the
implementation of activities. The conclusion of the research together with his
closing statements is presented in the fifth section.
2. THEORETICAL REFERENCE
Among
the many changes that have occurred in automakers over the years, most of them
refer to the mass production system, which begins with the creation of the
first mass production model named, by the American businessman Henry Ford,
Fordism in 1914.
In
the mid-70s, facing the need for flexibility production, there is a unique and
new way of managing the work process: the Toyotism. There, the workers become
multifunctional specialists. He raised the productivity of Japanese auto
companies by eliminating losses and it is now considered a model adapted to
flexible production system. Among its features are: the existence of a
cooperative relationship between managers and workers, that is a horizontal
administrative hierarchy; rigid quality control.
According
to Pergher (2011), “The Toyota Production System (TPS) is one of contemporary
approaches to widespread production engineering in the industrial context,
which proposes improvements in processes, by eliminating the losses."
Shimokawa
and Fujimoto (2011) bring the Toyota production system is based on two basic
concepts: cost savings through the elimination of losses and the recognition of
the Japanese diligence.
The
Kaizen system is premised on continuous improvement and its philosophy is an
important resource in the relentless pursuit of improvement in production and
administrative processes, making them leaner and faster. Monteiro (2012) points
out that "The Japanese word Kaizen means" continuous improvement
"results from combining two words Kai and Zen, which means" change
"and" better ", respectively".
Considered
as the main cause of the Japanese success, the Kaizen has been spread to many
countries. Companies around the world have used this management model in their
processes, seeking to improve the quality of its products and services to meet
customer needs, with reducing of time and operations cost, thus obtaining the
desired profit. (MEDEIROS, 2012).
Peinado
and Graeml (2012) cite the importance of the analysis, improvement and review
of logistics processes of the organizations to serve their strategy even
better. The competition for greater
efficiency has led some companies to update and use new ways to conduct their
business, having as one of its main objectives the seeking for improvements by
developing new business models or even adopting intensive technology in improving
processes, products and services.
Ramos
(2014) brings the term lean production as a way to do more with less, in order
to organize in the best way the production processes and build strong
relationships with customers and suppliers, by using the pulled production
system. In this system, production happens from the customer's request, making
the raw material to be purchased as needed for production, avoiding stocking of
finished product. In addition, Saurin and Ferreira (2008) classify it as a new
industrial organization system, inspired by the Toyota Production System (TPS),
which aims to eliminate any losses of the production system, enabling high
quality services and products at the lowest cost possible and attending more
effectively to customer needs.
In
addition to this already mentioned, Hayes and Wheelwright in 1984, were the
first to mention the term World Class Manufacturing, when describing developed
capabilities by Japanese and German companies in competition for export
markets. In 1986, Schonberger also cited such term in his book World Class
Manufacturing taking the idea that adopting Just-in-Time and Total Quality
practices, any company could reduce their lead time and become a World Class
Manufacturing (CUTS, 2010). Thereafter, the two concepts have been adopted and
expanded by several authors.
The
main motivation of Lean Manufacturing methodology is to seek the time reducing
between customer order and delivery through losses elimination. It promotes the
identification of what adds value (and that does not add) from customer's
perspective -The interconnection of necessary steps and goods production on the
value flow, so that this proceed without interruptions, detours, returns, waits
or losses - and the operation of this pulled demand flow (SILVA, et al., 2011).
The
industries are going through a critical review of their own operating
standards, regarding to its production planning and inventory control
processes, so that it can have a production management even more detailed
(DIAS, 2008).
Whenever
a company seeks to reduce its operating costs and improve productivity, it
carefully reconsiders their picking activity. The picking consists in
collecting the right products, in the right quantity, to meet the needs
expressed by customers. So it is from picking that begins in customer service
and therefore this activity is aimed for much attention. The faster processes picking, the faster it is achieved to deliver to the
customer (time); As more efficient the picking, the lower the cost to the customer;
As more effective the picking, without mistakes, the greater is the quality of
delivery (CARVALHO, 2010).
Tompkins
(2010) observed (Figure 1) that travel time represented 50% of the picking
activity. The travel time is a linear function of distance traveled in a pick-to-part
system, where the operator travels to the inventory.
%REQUEST -
PICKING TIME
Figure 1:
Typical Distribution picking time
Source: Tompkins, 2010
Consequently,
the travel distance is considered as one of the main objectives of the
logistics processes optimization studies. Routing policies aim to minimize this
distance in different contexts. (PUREZA; LAZARIN, 2010).
Pereira
(2012) provides that "for routing being one of the most important and
complex issues belonging to the distribution logistics, heuristics alternatives
tothe problem are constantly presented ". Due to that, optimal methods
have some disadvantages such as, depending on the layout it might not be an
optimal solution to the routing problem, the optimum paths may seem illogical
for operators that change their itinerary, an optimal method cannot take into
account congestion in the paths, in which can be done with a heuristic model.
Relating to that, according to Medeiros (2011) it has more approximate to the
real results and require less computational effort, as compared to the exact
methods.
Bartholdi
and Hackman (2011) bring the travel time during the picking activity as a loss
because it costs in terms of manpower and does not add value, generating the
need for path analysis to be performed for all the activity in a production
line, in order to achieve the aim of improving the process. In addition,
Tompkins (2010) noted that such trips accounted for 50% of the time of picking
activity, presenting another indication of the importance of the review of such
parameter.
3. RESEARCH METHOD
The
research methodology adopted in this study is action-research, in which
according Thiollent (2011, p.14), can be defined as:
"... A kind of empirically based research that is designed and
built in close association with an action or resolution of a collective problem
and in which researchers and representative participants of the situation or
problem are involved in a cooperative way or participative . "
According
to the policies of the studied target company, including the terms of lean
manufacturing quoted in this work will be deleted "World class
manufacturing" and "Lean Manufacturing" because both derive from
the same method and only have been adapted to other realities to be implemented.
Based
on what Tompkins (2010) noted, in this work will be studied in loco the macro
production process from parts receiving to their final use on the assembly
line, prioritizing the handling and transportation time analysis, because it
represents 50% from a picking process, relating that is the source that will
bring the best results related to the improvement of process efficiency.
Before
executing the activities, it will be measured the section production efficiency
through a management software (Hyperion workspace) used in the studied company.
This software makes the monthly measurement of employee’s shifts time related
to the production indication, which prevents any manipulation of the final
results.
First
of all, it will be made a chronological measurement of the times looking at the
activities that do not add value to the process, by using a calibrated
stopwatch. It was created a form containing the process segmentation, mainly
interested in the transportation and handling times. It will be made Ishikawa
diagram, also known as cause and effect diagram to find the possible causes of
these problems, in order to gather similar processes as proposed in lean
production, that is built in the assembly process and internal logistics parts
preparation to reach the target sectionin thecorrect form, amount and time. To
assist in methodology, it was created a 5W1H to define the activities,
responsible people, and the purpose thereof.
Table
2: 5W1H - Implementation of the Kaizen philosophy on the assembly line
As
proposed by Ramos (2010), storage does not add value to the product, but
contributes to the whole logistics system could comply with the proposed value.
With this, it is aimed the gathering of stock in one place, improving parts
control and reducing unnecessary handlings. To build a new domestic
distribution center, called kitting, picking method will be defined as well as
its routing.
It
will be deployed S-Shape routing process, where the operators get into the
corridors, in which there are parts in their respective locations to be
separated as increasingly order of locations and comes out the opposite side
necessarily. He goes on before finishing the kits separation. This strategy is
the simplest, but based on what was done by Medeiros (2011), it will be the
most appropriate method to be implemented by its convenience and for achieving
an average solutions close to 5% of the optimal solution, which can confirm
their usage in practice.
During
the unification of inventories, a survey will be raised based on the movements history
in order to organize continuously and increasingly the parts that have higher
inventory movements, and to position them to reduce handling during the picking
process. The parts supply method will be studied with the participation of
those involved in the process by exposing problems and improvement suggestions.
After
implementing the presented methods, the results will be monitored in accordance
to the section production efficiency measurement through management software (Hyperion
workspace), where it will be seen within six months. It is known that the
methods and concepts used in this article can be used in any companies that
have a production line with a big volume of controlled items.
4. RESEARCH
4.1.
Manufacturing
section Profile
The
assembly section, object of study, has a total area of 3.072m². It has
rectangular shape with 96m long and 32 meters wide. The assembly line itself is
the continuous type and has a length of 75m in which are produced 6 different
models of construction machinery with about 30 different options.
The
stock parts having a mass of over 500g and / or bigger pieces than 50 mm in
length are arranged next to the assembly line in shelves door pallets type,
arranged as closest as possible to the work station where the parts are
assembled (Figure 2).
FIGURE 2:
Assembly Industry Layout
The
shelves are arranged in two categories: the first one called SA (Sub Assembly),
making reference to the racks located on the sub assembly components which
supply the main assembly line, and AL (assembly line), which references the
shelves located on the Assembly Line.
Smaller
parts, that weigh less than 500g and have less than 50 mm are stocked in a
warehouse with shelves rack Flow type called "Kitting" and subsequently
supplied in plastic boxes per lot in supply cars. Those cars are removed by the
assembler himself when occurs a change in the machine model to be produced.
The
"Kitting" stock layout has a total area of 1536m². It is
rectangular shape with 96m long and 16m wide. It has three aisles of parts
separation, called K1, K2 and K3 with twenty-four shelves each. All shelves
have the capacity to address 45 and 180 positions / box, which allows them to
be stored 12,960 positions / box.
FIGURE 3: Kiting of Industry Layout
4.2.
Storage
Policies
When
a new item arrives for the first time in the receiving dock, there is no
defined location. Firstly, it will be classified according to the machine model
to be used, and it will be stocked in "pallets on the floor" at the
beginning of the assembly line. Then it should be allocated in Pallet holder
closer to the place of use, or if the characteristics of the parts fit the
Kitting section, it will be stored and then supplied as addressed.
The
Picking system in assembly-line section is the pick-to-part type, that is a
manual system where the operator travels to separate the items. The most
critical point in the case of the layout is shown in figure 3, since in this
case the "operator" is also the assembler.
There
is not a particular routing policy for pallet holder on the assembly line,
since the assemblers perform the pickings in the order they feel better,
moreover, physical boundaries, along with the need to bring closer the parts to
the usage location. This process makes that there are many parts of different
models, different sizes and characteristics on the same shelf, including
hindering the best use of physical space.
In Kitting
section, the system is also pick-to-part, but done as addressing system in
ascending order of leasing parts, following the flow as follows:
Figure
4: Kitting section parts separation Flow.
The section's work is the semi-indirect type and its
efficiency is not controlled
5. DIAGNOSIS
Through
the Integrated Management System "Hyperion Workspace", software used
at the assembly plant, it was observed the current efficiency of Assembly Line
(F50) through the history of the last three months:
Figure 5:
productive efficiency graph in the assembly section
Source:
Hyperion Workspace - 12.15.2013
Through
cronoanalysis, the assembly process was characterized in two parts: The first
of the activities that add value to the assembly, and the second of activities
that do not add value to the assembly (abnormal).
Figure 6: assembly process time graph x type of
activity
Activities,
that add value to the assembly, make reference to increase the capacity and
efficiency of the productive section, which is the actual assembly line. In
this way we can reduce the time on activities that add value to assembly, with
investment in new technologies and more productive tools, but cannot eliminate
work steps.
Now,
the activities that do not add value to the assembly process, are characterized
by resolution of defects, overproduction, waiting handling, extra processing,
under-utilization of people and especially transportation. To identify the
possible causes of the problem of the activities that do not add value, it was
performed an Ishikawa diagram, where he highlighted three items: unpacking
operations, several machine models in just a shelf and transporting and
handling.
Figure 7: Ishikawa Diagram of activities that do not
add value to the assembly.
It
was stratified the possible causes of activities that do not add value to the
process through the data collected in cronoanalysis, where it was observed that
55% of these activities were characterized by transport and handling and
unpacking 35% and 10% others as graph below:
Figure 8: activities that do not add value in the
Assembly Line (NVA).
It
was analyzed the changes performed on the Assembly Line related to the
pick-to-part, where it was found the following flow:
Figure 9: Assembly Process Flow
Through
the process flow analysis it can be seen that the handling during the assembly
process is very intense, and is difficult to measure exact distances, since the
Assembly Line is the continuous pulled type. In that way, the proposal is to
reduce the changes to get parts on the shelves and parts from the Kitting
section, eliminate unpacking and disposal of the assembly process.
The Kitting
process flow is defined by addressing sequence, but the addressing is defined
only by the availability of physical space in stock, so that the movements may
be higher in most consuming items, as follows:
TABLE 3: Example
of monthly consumption of parts list and addressing.
This
table shows that in many cases the parts consumption is large and the distance
to be covered during their separation is larger than that in parts with a lower
consumption. The proposal is to reduce handling during the picking process as
suggested by Ramos (2010).
6. RESULTS
As
proposed in the reference, it was created a single location for all parts that
reaches the plant, enabling them to be delivered and it was created
standardized way to delivery items to assembly line. In the new method of
delivering parts, the parts are allocated in a car named Kit machine as
delivery addressing (model, serial number and line on), where is supplied the
required amount to assemble a machine.
According
to the machine production schedule, Kit machine is positioned and pulled few
meter away from the machine, as shown below:
Figure 10: Assembly Process Current Flow.
With
the new flow in the assembly line, the assembler does not make handling to the
shelf to remove parts, and also does not perform the parts unpacking. After
executing the activities, measurement and assembly section efficiency results monitoring
was performed, as the graph below:
Figure
11: Production efficiency in the assembly industry
Source: Hyperion Workspace 01.03.2015
As
noted in graph 4, after the implementation of new parts supply flow, and
consecutively with a significant reduction of activities that do not add value
to the assembly process, there was a significant increase in the efficiency of
the assembly line section (F50).
The
efficiency loss in February 2014 is acceptable since at this month there was
the implementation of a new machine model in the industry, which can also
reveal that in this particular case, the assemblers adaptation curve. For the
new model. obtained a significant improvement, as previously entailed about
four months of falling efficiency. With the new process flow, when occurs a
development, the most significant impact is applied to Kitting section, which
in these cases, they should prepare spaces and create new picking sequences.
It
has expanded the Kitting section to aggregate the parts that were housed in the
pallets holder at assembly line. For a leaner picking, it was made up the
reorganization of the parts allocation, in accordance with criteria as shown
below:
Table 4:
Classification of parts
|
CORRIDOR |
CLASSIFICATION |
AVERAGE MOVEMENTS
/MONTH |
|
K1 |
HIGH
MOVEMENT PARTS |
>
1.000 |
|
K2 |
MEDIUM
MOVEMENT PARTS |
501
A 1.000 |
|
K3 |
LOW
MOVEMENT PARTS |
<
500 |
It
was raised the history over a period of six months to relate the average item
movements / month. Thus the items cited as examples in Table 3, were relocated
as shown below:
TABLE 5: New
allocation of parts
|
MONTLY
AVERAGE CONSUMPTION |
PAST LOCATION |
ACTUAL
LOCATION |
|
|
0102480820 |
1816 |
2K125E1 |
2K102B2 |
|
0102481230 |
1816 |
2K318B1 |
2K106A3 |
|
0102480818 |
1312 |
2K113A5 |
2K107D2 |
|
0102480616 |
1257 |
2K102C3 |
2K107E1 |
|
0102480825 |
978 |
2K202B7 |
2K201B1 |
|
0102481025 |
908 |
2K317B2 |
2K202C4 |
|
1243030158 |
771 |
2K227B3 |
2K207B1 |
|
0102480620 |
728 |
2K218A3 |
2K211B3 |
|
0102480630 |
584 |
2K324D1 |
2K213A5 |
|
1342711310 |
454 |
2K101A2 |
2K306B4 |
|
1246232510 |
227 |
2K116D4 |
2K310C2 |
|
1246232580 |
227 |
2K317D3 |
2K325A4 |
In this way, the distance traveled
by the logistics operators is lower during the picking, improving the process
flow and reducing 32% of the kit’s separation average time. The method used for
flow rack type shelves in Kitting section, was extended to the pallets holder
from the Assembly Line, the final layout of the sections obtained the following
organization:
Figure
12: New assembly section layout and Kitting
According to the new arrangement of
the Assembly Line, it was obtained some improvements in processes that involves
Assembly Line flow, as follows:
Figure 13: Improvements implemented in Assembly Line
The
“kit machine” now effects the parts supplies that previously were made by
hydraulic pallet trucks, with the amount depending on the receiving. It
contains all the parts to assembly a machine. However transporting the kit
machine is made through an electric tug, where you can transport up to 4 units
of the kit machine per trip.
The
handling during the picking process done by the assembler in which previously
was from 3m to 18m, with the added kit machine on the back side of the machine
on Assembly Line, it was increased to a maximum of 1m, reducing to almost zero
the assembler’s handling.
In
terms of layout, it was obtained great advances in relation to 5S and
principally the production flexibility, which, currently, with the development
of new models, it is possible further adaptation from assemblers for items
identification of each model and the increase in the mix of machines produced
on one line.
Upon
execution of the activities that were proposed, the sections involved had some
major changes, as Table 6:
TABLE 6: Major
internal changes.
7. CONCLUSION
Some
implemented actions reflected significantly on direct results in the assembly
section efficiency indicators, while others only have noted on other section
indicators. The main benefit of the subject company of this study was the
improved production efficiency of an assembly line, as shown in Figure 4, which
obtained an increase of 50% to 85% of productive efficiency during the period
of eight months. Another important issue to be mentioned is related to models
flexibility to be produced on the assembly line, since the current flow for any
model is the same, that is, when there is the implementation of a new machine
model, all the logistical effort is to structure the section where are found
the parts in stock and kit machine supply.
Regarding
the analysis for routing the Kitting section, there was an average gain of 32%
in the total process time and thus strengthen in terms of availability of
skilled labor, the new picking process for the pallets holders that previously
were located next to the assembly line, not increasing the total picking time.
This
paper aims to identify the influence of implementation of lean internal
logistics on the productive efficiency of an assembly line. We observed that as
quoted by Tompkins (2010) actually transport offered great influence on
activities that did not add value to the process, in this way, with the
elimination of losses and with a new flow definition, process became leaner.
Another issue to be emphasized is that the parts were stored in a single area
where it was improved the usage of physical space and speed process.
This
action-research work has shown that the influence of the implementation of lean
internal logistics has brought benefits in all processes involved in the study,
but it is worth noting that activities have brought great benefit to the
studied company, may or may not bring benefits to other companies. In that way
the concepts and activities realized can be study objects in researches.
8. REFERENCES
ADES, C. (2011) Investimento no setor de
construção civil. Estado de São
Paulo, São Paulo.
ASSUNÇÃO, D. S. et al. (2013) Modelo de Gestão KAIZEN e Sua
Aplicação no Setor de Fertilizantes. In: 4th International Workshop
Advances in Cleaner:" Integrating Cleaner production into Sustainability
Strategies. p. 1-7.
BARTHOLDI III, J.
J.; HACKMAN, S. T. (2011) Warehouse & distribution science: release 0.92. Atlanta,
GA, The Supply Chain and Logistics Institute, School of Industrial and Systems
Engineering, Georgia Institute of Technology.
CARVALHO, José Crespo de. (2010) Logística e Gestão Logística. In Logística e Gestão da Cadeia de
Abastecimento. Sílabo.
CORTEZ, P. R. L.; et al. (2010) Análise das relações entre
o processo de inovação na engenharia de produto e as ferramentas do WCM: estudo
de caso em uma empresa do setor automobilístico. XXX ENGEP–Encontro Nacional
de Engenharia de Produção. São Paulo.
LAZARIN, D. F. (2010) Roteamento
dinâmico de veículos: análise do impacto em atividades de prestação de
serviço.
MEDEIROS, W. J. N. (2012) Abordagem heurística paralelizada para a resolução do problema do
roteamento de veículos capacitados com base na estratégia, cluster first
route second.
MONTEIRO, M. M. (2012) Pull Flow na Indústria Automóvel
KaizenInstituteConsultantGroup. Tese de Doutorado. Universidade do Porto.
NUNES, F. L.; MENEZES, F. M. (2015) Sistema hyundai de
produção e sistema toyota de produção: suas interações e diferenças. Revista
Acadêmica São Marcos, v. 4, n. 2, p. 101-120.
PEINADO, J.; GRAEML, A. R. (2007) Administração da produção: operações industriais e de serviços.Curitiba
:UnicenP.
PEREIRA, L. G. Simulação
estocástica e métodos heurísticos paralelizados para a resolução do problema de
roteamento de veículos capacitados com base na estratégia, cluster first route
second.
PERGHER, I.; et al. (2011) Discussão teórica sobre o
conceito de perdas do Sistema Toyota de Produção: inserindo a lógica do ganho
da Teoria das Restrições. Gestão & Produção, v. 18, n. 4, p. 673-686.
RAMOS, M.; Eidt, A. (2015) A filosofia Kaizen no processo produtivo da empresa Construblocos
Scherer.
RAMOS, T. (2010) Gestão
da Armazenagem e dos Stocks na Gestão da Cadeia de Abastecimento. Logística e Gestão da Cadeia de Abastecimento.
SAURIN, T. A.; FERREIRA, C. F. (2008) Avaliação qualitativa
da implantação de práticas da produção enxuta: estudo de caso em uma fábrica de
máquinas agrícolas. Gestão & Produção, v. 15, n. 3, p. 449-462.
SHIMOKAWA, K.; FUJIMOTO, T. (2011) O nascimento do Lean. Conversas
com TaiichiOhno, EijiToyoda. São Paulo: Leaninstitute Brasil, Bookman.
SILVA, I. B.; et al. (2011) Integrando a promoção das
metodologias Lean Manufacturing e Six Sigma na busca de produtividade e
qualidade numa empresa fabricante de autopeças. Revista Gestão &
Produção, v. 18, n. 4, p. 687-704.
THIOLLENT, M. (2011) Metodologia da pesquisa-ação. In: Metodologia
da pesquisa-ação. Cortez.
TOMPKINS, J. A.
et al. (2010) Facilities planning. John Wiley & Sons.
TORRES JR., A. S. (2010) Metadecisão no modelo de gestão
Toyotista. Revista de Administração Mackenzie, v. 11, n. 6.