PRODUCTION
BATCH SIZING AND INVENTORY LEVEL CONTROL USING SIMULATION SOFTWARE
Bruna Christina
Battissacco
Universidade de
São Paulo, Brazil
E-mail: brunacb@usp.br
Walther Azzolini
Junior
Universidade de
São Paulo, Brazil
E-mail: wazzolini@sc.usp.br
José Henrique de
Andrade
Instituto
Federal de São Paulo - IFSP, Brazil
E-mail: jose.andrade@ifsp.edu.br
Maicom Sergio
Brandão
Universidade
Federal de São Carlos, Brazil
E-mail: maicom.brandao@usp.br
José Marcelo Barbosa
Palma
Núcleo de
Estudos e Pesquisa em Gestão (NEPGEST), Brazil
E-mail: jmarcelo.palma@gmail.com
Submission: 2/24/2021
Accept: 8/10/2021
ABSTRACT
Computer simulation has great application potential in the area of Production Engineering as a tool to support decision making as it allows to simulate the functioning of a real system through logical relationships, in order to observe its behavior under different scenarios. Which could not be practiced in the real system. In line with this aspect, the present work aims to present a report on the application of simulation for the design of production batches and inventory control, highlighting the process necessary for the construction of the generated simulation model, as well as the challenges and opportunities observed. In order to achieve the proposed objective, a literature review was carried out on the topics of interest; the choice and understanding of simulation software and; the survey of data from a large auto parts manufacturer located in the interior of the state of São Paulo. The main results were an increase in production volume from the inclusion and dimensioning of a buffer after the assembly process and a balance between the number of items of each component of the product structure in the supply with the real capacity of manufacturing cell processing. Cabe ressaltar que o modelo de simulação produto da pesquisa deste artigo pode ser aplicado como um sistema de apoio à decisão do gestor para a elaboração do planejamento estratégico e do planejamento operacional com o propósito de melhorar a capacidade de análise e decisão. It is important mentioning that the simulation model in this article can be applied as a support system decision for the preparation of strategic planning and operational planning, with the purpose of improving the analysis capacity for decision-making.
Keywords: Logistics; Cellular manufacturing; Production Management; Simulation
1.
INTRODUCTION
The crescent use of technology as a support for the
process of productive systems management have shown great potential. It is able
to assist decision making, increasing business results accuracy, therefore
contributing to efficiency and overall accuracy.
In the emerging context and progressive consolidation of
Industry 4.0 (I4.0) and associated paradigms, finding ways to perfect and
integrate production management, and making adequate use of the available
technologies will be both a challenge and an advantage for the companies that
aim to acquire competitive advantages in this new context.
Industry 4.0 is a term conceived in the 2011 Hannover
Fair as a part of a long-term strategy from Germany to strengthen their
manufacturing sector competitiveness (Liao et al., 2017), and to promote the
emergence of dynamic value chains, real time optimized and auto-organized based
on criteria as cost, availability and resource consumption (Kagermann
et al., 2013). Thus, it could provide the necessary flexibility to companies
answering the market’s demands, the crescent product personalization, smaller
product life cycle, as well as the crescent complexities of products and
production (Hirsch-Kreinsen, 2016).
The main technologies and development tendencies that are
driving these innovations in the I4.0 include the following themes: Green
Information Systems, big data, autonomous robots, horizontal and vertical
integration between systems, cyber security, virtual and augmented reality,
industrial IoT, additive manufacture, cloud technologies, modeling and
simulation, this last one being a key tool for I4.0 solutions by allowing
improvement evaluation in a highly complex and dynamic scenario (Goienetxea Uriarte, 2018). The
simulation is defined as the procedure of designing a model of a real or
hypothetical system to describe and analyze the system's behavior (Scheidegger et al., 2018).
Simulation is a technology used to develop exploratory
and planning models to optimize decision making, as well as to project and
operate complex production systems (De Paula Ferreira, 2020; Goienetxea Uriarte, 2018).
Academic studies about risks, costs, receipt potential,
and implementation barriers of I4.0 are scarce (Kagermann
et al., 2013). Lugert et al (2018) suggest that
simulation techniques would be useful to address those issues since they offer
the possibility of evaluating multiple I4.0 scenarios through exploratory and
complex systems planning methods that can help in the partial treatment of the
mentioned problems. Computational simulation, even preceding I4.0, regains
relevance since allows the use of computational techniques to simulate the
operation of a system from mathematical models. This method comes up as an
important tool for it allows the simulation of a real system through logic
relations, in order to observe its behavior under different scenarios - which
couldn’t be practiced in the real system (Morabito & Pureza,
2010).
Another reason for using simulations is the high costs
associated with experiment development in the real system or with building a
physical model (Scheidegger et al., 2018).
Particularly, simulations have performed a significant
role in assessing the project and the operational performance of the
manufacturing systems. Successful applications of simulation in various
real-life problems proved their efficiency in approaching several challenges in
the manufacturing sector (Negahban, 2014).
Soares et al. (2011) point out that in many studies it is
perceived the benefits of simulation, either for previsions or analysis of
unexpected events, individual or simultaneous, generated by changes in a
productive process. Other benefits of simulation include providing a systemic
vision of such changes, which are hardly obtained through conventional analytic
studies. In this sense the present article considers the coronavirus (Covid-19)
pandemic period and its impacts on scale economy since social isolation is
pointed by the World Health Organization (WHO) as an efficient preventive
measure to control and contain the sickness, negatively affecting the offer and
demand conditions.
In this way, the companies found themselves with their
operations totally or partially interrupted, disarticulating the commercial and
personal relations and consequently the supply chain, turning explicit the
necessity of integration and review of the operation mode practiced.
Due to this context and given of the computational
simulation potential, the objective of the current work is presenting an
application report of simulation for the production lots dimensioning and
inventory control, highlighting the necessary process to constructing the
generated simulation model also the challenges and opportunities observed.
In order to reach the proposed objective, a literature
review was made regarding the themes of interest, the choice and comprehension
of a simulation software, the data survey from a large auto parts factory
located in the interior of São Paulo, and the construction and application of a
simulation model for the studied case.
This work is structured in five sections. Beyond the
present introduction, section two presents the theoretical references utilized
to base the study, section three presents the methodological aspects, followed
by data analysis and discussion (section four) and conclusion (section five).
In the end, the bibliographical references are presented.
2.
THEORETICAL REFERENCES
2.1.
Computational Simulation
Simulation is a type of
mathematical modeling that aims primarily to portray the dynamics of an
existing or planned system for solution evaluation. This technique is
recommended to approach a higher detail level on the process (Peixoto &
Pinto, 2006). The simulation consists of one of the 4.0 Industry tools,
providing more precise and agile information (Gaziero
& Cecconello, 2019).
Besides definitions of
computational simulations, Gaziero and Cecconello (2019) emphasize that simulation can be used to
deal with uncertainty and to create dynamic visions, taking into account stock
levels, waiting times, and resource utilization for different scenarios,
supporting decision making. Authors as Biswas and Narahari
(2004) highlight the importance of utilizing the simulation for analyzing
product flow and supply chain information in the decision-making process. It is
useful because of the large-scale information from supply chains involving a
hierarchical structure of decisions with dynamic interactions between two
organizations. In this direction, Hernandez and Librantz
(2013) approaches, through a simulation model, the possibility of reducing
stock maintenance costs by using a new planning strategy.
Corroborating with the
authors, Silva (2005) supported the following hypothesis: the combination
between simulation and optimization has great value on decision-making support,
bringing significant advantages on elaborating a manufacturing cell with lower
production costs.
According to the author,
the utilization of Activity Based Management (ABM) / Activity Based Costing
(ABC) made it possible to combine necessary information for obtaining more
precise costs. Therefore the results of a cell also triggered off an
“optimized” production cell. For Negahban and Smith
(2014), some of the factors that contribute to the growth of successful
publications and applications of simulations are: the incorporation of
optimization algorithms in simulation software packages, the reduction of
variability and other efficiency increasing techniques. Such factors increased
the tool’s credibility between researchers and professionals.
2.2.
PRODUCTION DIMENSIONING
2.2.1.
Production lot
dimensioning and inventory control
In the work of Catelan et al. (2020), the problem of lot dimensioning is
pictured as basically determining the number of items to be produced in each
period of limited time. Variations of machinery quantities involved in the
process will consider both resource restrictions satisfaction and demand
attendance.
According to the
authors, the lot dimensioning problem has an economic origin and involves
production costs, stock, and machine preparation. Besides, dimensioning is a
tactical and operational competence that interprets strategic planning
decisions and transforms them into real production plans. Oliveira and Santos
(2017) argue that decisions on lot sizing are necessary to improve stock
management and to reduce costs. High stock levels grow costs like maintenance,
and low stock levels could delay demand attendance.
It stands out that the
programming problem is in determining production lot sequencing to minimize the
time and cost generated by lines set up in the production line. In this
exchange of products, there is dependency between the previous items the setup
time is considered dependant to the sequence and/or
cost structure, entailing a simultaneous decision on sizes and sequences of
production. According to Glock, Grosse and Ries (2014
apud Oliveira & Santos, 2017, p. 4) “the problem of lot dimensioning (LD)
consists in determining the optimal size of production lots in order to
minimize costs and attend the clients demand, receiving special attention of
researchers due to its importance for global economy”. Between the outstanding
authors on literature regarding dimensioning and sequencing of production lots,
exemplified by the General Lot-sizing and Scheduling Problem (GLSP) are: Drexl and Kimms (1997),
Fleischmann and Meyr (1997), Haase
and Kimms (2000), Meyr
(2000), Allahverdi et al. (2008), Araújo et al. (2008), Jans and Degraeve (2008),
Toso, Morabito and Clark (2009 apud Junqueira & Morabito, 2018), Ferreira
et al. (2010),
Clark, Almada-Lobo and Morabito (2010 apud Junqueira
& Morabito, 2018).
Taking into account the
variables used to calculate depletion time and methods for production lot
dimensioning, two methods are analyzed as to their implications, respecting
cases and peculiarities which would foment different models. That is, the
Economic Lot of Shopping (ELS) or the model (Q, r).It must be highlighted that
the leveling between produced and consumed value due to the number of
competitors and today’s uncertainty of demand became a challenge for
industries.
The model of Wanke and Saliby (2005) presents
a solution of the inventory control (Q, r) model with uniform distribution of
demand and lead time.
The model (Q, r)
considers the stock of new products and the long term learning of the
characteristics on demand distribution in lead-time supply.
Deals with lot sizing
and request point for uniform supply lead-time and demand, in a more applicable
form than normal distribution due to presenting situations where any result
have the same probability of occurrence.
Starting from this
premise, Wanke and Saliby
(2005), calculating the values of the function of probability density of supply
lead-time demand, demand projection and the respective variance of demand on
response time (X), can be done through the mathematical expressions (1), (2)
and (3).
(1)
(2)
(3)
In the expressions dm
corresponds to daily minimum demand and dM
to daily maximum, tm represents minimum supply lead-time in days and
tM maximum supply lead-time, also in days.
The model treats the delimitation of
inferior and superior limits both according to demand and resupply lead-time
variability.
Regarding the product between demand
and supply lead-time, the products of the coordinates of these vertices are tm
* dM and tM
* dm, and the integration regions also depend on the size of the reorder
point.
Depending on the size, according to Wanke and Saliby (2005) the level
of service in the cycle must be calculated in one of the three possible
integration regions:
1. Integration Region 1: 𝑟 ≤ 𝑡𝑀 × 𝑑𝑚
2. Integration Region 2: 𝑟 ≥ 𝑡𝑚 × 𝑑𝑀
3. Integration Region 3: 𝑡𝑀 × 𝑑𝑚< 𝑟 < 𝑡𝑚 × 𝑑𝑀
The model of Sarkar et
al. (2019) presents the setup time reduction and of the impact of security
factors on stock resupply management on supply chain.
In other words, the
supply chain model for reorder point considers the productive unit and retail
to reduce lead-time and setup time, in order to obtain their impacts on total
costs when lead-time demand is stochastic.
With lead-time depending
on lot size and consisting in production time and setup time, with free
distribution.
The model proposed by
Sarkar considers a Two-echelon supply chain: provider or manufacturer and
retailer that attend the market. And detects the event of manufacturing process
anomaly through the decision variable 0.
The vendor (1st
level): manufacturer works with stochastic lead-time (lead-time L(P,Q)) due to
productivity oscillation (there is variability between the productive rates Pmin and Pmax).
The buyer (2nd
level: retailer) operates through the system (Q, S). A system of continuous
revision. In this case the retailers always acquire a lot with equal amounts of
the product in all dispatches, according to the resupply period.
However there is an
impact related to lot size and consequently to production, being considered by Sarkar
et al. (2019):
a) Average (DL(P, Q));
b) Standard Deviation (𝜎√𝐿(𝑃,𝑄));
c) Lead-time depends on
setup time and handling time. Both represented by parameter 𝑡𝑆.
d) The Lead time crashing
cost also reduces 𝑡𝑠.
e) Setup crashing cost also
reduces 𝑡𝑠.
f)
The moment also considers the possibility of investments on minimizing
setup cost and parts with manufacturing defects.
g) Retailer: there are
delayed requests from the manufacturer. In this case there are security factors
(𝑘1 and 𝑘2) for the lots and the security
stock (S).
The model
considers for the vendor cost function (manufacturer) that the unitary
production cost is in function of the production rate P ($/time) according to
function Cp (P) presented by expression (4).
(4)
𝑃𝑚í𝑛 > 𝐷.
Which
impacts on production lot size. Meaning that the bigger the lot more possible
for the production rate to rise which is relevant to the factory since it can
produce a surplus lot if the demand is inferior to the amount produced in the
considered time period.
From the
production rate it is also determined the average inventory, being applied the
storage cost according to expressions (6) and (7).
(5)
(6)
With hv for vendor and hb equal to ($ / unit / time unit) buyer.
Another crucial point is in varying
the storage cost.
The mathematic expression (7),
according to the authors, defines the total vendor cost and (8) the total cost
of the supply chain.
(7)
(8)
In the expressions
(7) and (8) there is an impact from the parameters Av (cost by setup) and from
the amount of defective parts. The total vendor cost and the supply chain cost
both rise from a higher lot size.
This dynamic
of gains and losses in business and the need for allocating resources in the
best possible way converge on the current problems of the decision-making
process on costing systems. Besides, for Mantovani et
al. (2019), the adequate management of processes inherent to logistics allows
companies to obtain competitive advantages allied to cost reduction.
2.2.2.
Cell Layout
Similar to
the present study, Soares et al. (2011) performed a case study in a
manufacturing cell of a company from the automotive sector. Among several
studies on computational simulation, restructuring cellular layout seems useful
to reduce in-house stock levels, increase resource productivity, reduce lead
time, and adequate workforce on the production cell. Decker Junior et al.
(2020) highlight that choices of layout configuration are fundamental to make
feasible and raise a company's competitiveness. Thus, a cell is projected to
attend a determining part (from raw material to finished product), using
similar machines and tooling. Upon contemplating the manufacturing process and
the material storage, there is special attention to the manufacturing cell, which
according to Mancio and Sellitto
(2017), mixes characteristics and layout advantages of processes and products,
usually with a substantial reduction in crossing time and in-process inventory.
Similar parts and pieces are grouped in number and volume of minimum
production, for example, in families or subgroups organized by analytical
methods.
Regarding
the theoretical framework from Mancio and Sellitto (2017), cellular layouts are best suitable for
medium volume and medium variety operations. However, the efficiency of
cellular manufacturing will depend on the adequacy of the heuristics used in
its planning. And according to Ekren and Ornek (2008), when the production sequencing suffers
frequent and substantial changes it is encouraged to incorporate the virtual
cellular layout concept. For Carvalho et al. (2019), through the implementation
of a manufacturing cell, it is possible to do project a reduction in production
time (in days), the potential gain in production time (in %), and the quantity
of produced parts (by day). Besides the improvement in stock turnover and
reduction in people involved in manufacturing. In other words, the
manufacturing cell performance can be measured in multiple ways, such as:
productivity, WIP levels, crossing time, average of delayed work, average of
anticipated work, average time of work on waiting list, machine utilization,
operator utilization, material handling costs, setup costs and inventory costs
(Soares et al., 2011).
The
relevance of the considerations and analysis of cellular layout are imperative
to achieve less waste and to optimize the activities which aggregate value
through the productive chain.
3.
RESEARCH METHOD
The study
contemplated the application of the simulation and modeling method in a
manufacturing cell of a big auto parts manufacturer located on the interior of
the São Paulo state, which will be referred as Company Alpha. For that, it was
considered the simulation stages described by Chwif
(1999), which contemplates the model’s conception, its implementation and
analysis of the results (Figure 1).
Figure 1: Life cycle of a
simulation model
Source: Chwif,
1999, p.10
3.1.
Conception
The first
stage on constructing a simulation model involves its conception. Thus, it was
defined as the objective of the problem representing a production system in a
simulation software and the creation of improvement scenarios in lot
dimensioning and inventory control.
For that, it
was necessary to understand what parameters were fundamental for modeling, so a
bibliographic research was performed. In the research, the authors consulted
journal databases and scientific publications with recognized sources, national
and international. In this first stage there were identified system parameters
as depletion time, flow time and average stock. Once identified the parameters
and information necessary to describe the system, the first abstract model was
elaborated from the processes taking place inside the manufacture cell being
studied, involving machining and component assembly, as well as material flow.
From this
abstract model, new information’s were added relative to the studied case, that
is, the input data collected. Choosing a specific case for application helps on
approximating theory and practice. And in this case allows an interesting role
of, not only creating an abstract model to evaluate the behavior of the
parameters obtained from theory, but also that the application for more than
one specific reality allows to explore atypical events in dynamic environments.
For example, upon demand unpredictability in order to understand the behavior
of production processes and stock management.
It stands
out that there is no pretense on exhausting the theme due to being a punctual
and robust case, but to foment important analysis to future studies. Therefore
much attention and care are given to generalizations or simplifications, and in
no moment the scientific rigor is unvalued, being necessary to validate the
case study (Yin, 2001; Ventura, 2007).
Some
examples of utilized input data: number of suppliers, processing time of
machine components and number of hours in a work shift. After collection and
input of data, the next stage contemplated the model’s implementation.
3.2.
Implementation
The
implementation stage, as mentioned by Chwif (1999),
involves the conversion of the model to a simulation language, which is
commonly deeply connected to computational language. For that, the software
Plant Simulation was utilized, with its respective programming language. Plant
Simulation was developed by Siemens PLM Software and utilized to help on
modeling, simulation, analysis and visualization of production processes,
material flow and logistic operations. The programming language utilized by the
software is SimTalk II.
3.3.
Analysis
Finally, the
last step involved in the method was analyzing from the experimental or
operational model. As the name suggests, this model is the result of the
computational model validated by the consistency on representing the studied
process and carries great value to creating scenarios (Chwif, 1999).
Five
scenarios were explored in this step, altering variables like running machine
speed, buffer presence and the increase of the dimension on the process running
machine. From the obtained results for flow time and produced volume, it was
indicated in which context the best result would be obtained, contributing to a
more rationalized decision making.
4.
RESULT ANALYSIS AND DISCUSSION
4.1.
Model construction and
generated scenarios
Figure 2 represents in scale the manufacturing cell used
by this study with the work objects distributed on the frame of the Plant
Simulation software from Siemens (Version 14.1, proprietary license from the
School of Engineering of São Carlos (EESC/USP).
The
constructed manufacturing cell model produces a subset of the product
manufactured by Alpha company. The subset is assembled on workstation 3. The
assembly is executed from the union of 8
components.
Figure 2: Layout of each
Manufacturing Cell.
Source: Authors (2020).
The supply
of components in the manufacturing cell is executed by the supply system A1,
A2, A3, A4, A5, A6, A7 and A8. In order to replicate the system, the authors
developed a routine in SimTalk II language,
considering two subset models according to the material structure of the
product on Table 1.
The
production flow in the manufacturing cell, study object of this work, does not
represent a problem with setup times dependent on the sequence, which motivated
the authors to consider the model from Wanke and Saliby (2005) to scale the buffer after the assembly
process.
In function
of the foreseen demand of products A and B, considered by this study as 856
units per day each, and the planning horizon considered as 20 working days with
an uninterrupted operation turn of 8.5 hours per day, the available amounts
available for cell processing are found on Table 1. Figure 3 represents the saw
tooth graph of the supply in question.
Table 1: Bill of Materials.
|
Supplier |
Item |
Quantity |
Daily Supply |
|
A1 |
1 |
1 |
1.712 |
|
A2 |
2 |
6 |
10.272 |
|
A3 |
3 |
1 |
1.712 |
|
A4 |
4 |
1 |
1.712 |
|
A5 |
5 |
1 |
1.712 |
|
A6 |
6 |
3 |
5.136 |
|
A7 |
7 |
9 |
15.408 |
|
A8 |
8 |
2 |
3.424 |
Source: Authors (2020).
Figure 3: Saw tooth Graph from Item
7 (Supplier A7).
Source: Authors (2020).
Of the eight
components from BOM only the Supply A1 component has a machining process, which
can be executed in two similar machines with processing time of 17.88 seconds
per item on both machines, as indicated by Figure 2.
After
machining on machines 1 or 2 the item is transferred to assembly, workstation 3,
through running machine E1. The assembly time is 15.5 seconds by set.
The
production flow follows from the balancers (workstation 4) to stations 5, 6, 7,
8, 9 and 10. Regulator (15.78 seconds per item), SAX (load measuring machine -
18.69 seconds per item), engraving the laser identification and oiling of the
set (13.03 seconds per item), final inspection and packing (12.83 seconds per
item).
With the
model design finished, with the correct position of the machines on the frame
according to the layout of the cell used by the study, five simulation
scenarios where generated:
· 1º scenario – velocity of 1 m/s on the running
machine C1 without assembly buffer and the number of items on C1 without
definition;
· 2º scenario – velocity of 1 m/s on the running
machine C1 with assembly buffer and the number of items on C1 after buffer
equal to 6;
· 3º scenario – velocity of 3 m/s on the running
machine C1 with assembly buffer and the number of items on C1 after buffer
equal to 6 with a raise on volume of items during supply, according to Table 2;
· 4º scenario – velocity of 3 m/s on the running
machine C1 with assembly buffer and the numbers of items on C1 after buffer
equal to 6;
· 5º scenario – increase on the extent of the
running machine C1 from 3 to 3.5 meters and raising the velocity from 3 to 4
m/s, with inclusion of buffers on the regulators and the supply buffer raising
the amount of items in the running machine after the buffer from 6 to 7.
The best
result, according to Table 2, is scenario 4 with a volume of 29.576 items and
an assembly flow until the packing of 21.308 seconds. The graph on Figure 3
exhibits the increase on produced volume in addition to the reduction on flow
time on assembling the kit’s package.
According to
Figure 2 the workstation 3 (Assembly) when liberating the set its operation
becomes restricted due to the velocity of the running machines and the
transport tables to the balancing machines. Before the inclusion of the buffer
after assembly, due to the lack of space for set movement, limiting the space
on the running machine C1, and the transporting table do not perform the
transport operation, interrupting assembly while waiting for space liberation.
Table 2: Supply volume of the items
on each scenario.
|
|
|
|
Production
Lot by product |
|||
|
Scenario |
Produced
Volume |
Flow
Time |
1st
week |
2nd
week |
3rd
week |
4th
week |
|
1 |
26.692 |
23,616 |
856 |
650 |
650 |
650 |
|
2 |
26.679 |
23,615 |
856 |
650 |
650 |
650 |
|
3 |
29.572 |
21,310 |
856 |
856 |
856 / 650 |
650 |
|
4 |
29.576 |
21,308 |
856 |
856 |
856 / 650 |
650 |
|
5 |
29.574 |
21,311 |
856 |
856 |
856 / 650 |
650 |
Source: Authors (2020).
Figure 4: Variation of produced
volume against reduction of flow time.
Source: Authors (2020).
The
inclusion of a buffer regulates the flow so interruptions in the assembly won’t
be necessary, maintaining a more continuous production flow and raising
produced volume as shown in Table 2.
From the
construction model and flow reconfiguration with the buffer inclusion after
assembly the authors applied the lot dimensioning procedure from the authors Wanke and Saliby (2005) to scale
the item volume to be kept on the assembly buffer in order to maintain
continuous flow without interruption with minimal stock.
The buffer
calculation after assembly was accomplished according to expressions 2 and 3.
Tables 3 and 4 show the results.
Table 3: Lot dimensioning assembly
buffer.
|
Minimum daily demand |
|
591 sets/day |
|
Maximum daily demand |
|
856 sets/day |
|
Lead time of minimum resupply |
|
0,58 days |
|
Lead time of maximum resupply |
|
1,30 days |
|
Variance |
|
1754,16 |
|
Standard Deviation |
|
41,88 |
Source: Authors (2020).
Table 4: Expected demand by answer
time (X).
|
Expected demand by answer
time (X) |
||
|
Description |
Lead Time |
Demand |
|
Response time (medium) with demand
projection |
0,94 days |
678,24 / day |
|
Response time (minimum) with
demand projection |
0,58 days |
856,00 / day |
|
Response time (maximum) with
demand projection |
1,30 days |
1356,49 / day |
Source: Authors (2020).
5.
CONCLUSIONS
The achieved
results, although preliminary, revealed the importance of simulation as a
support tool for decision making in the process of lot dimensioning with
emphasis on inventory cost reduction for industry 4.0.
For complex
environments, it is not a simple task to plan the production flow points which
require buffer stages in order to maintain synchronism as well as dimensioning,
, especially in environments with an elevated number of items being moved and
processed, even if in a manufacturing cell.
The present
work achieved its objective regarding the development of a model and its
gauging, presenting preliminary results through the application of the Wanke and Saliby (2005) model
with positive results regarding the raising of production volume whilst
reduction of flow time.
REFERENCES
Biswas, S., & Narahari, Y. (2004). Object oriented modeling and decision
support for supply chains. European
Journal of Operational Research, Amsterdam, 153, 704-726.
Carvalho, P. S., Schneider, V. A.,
Parreira, L., & Chapoval Neto, A. (2019). Proposta de implantação de uma
célula de manufatura: um estudo de caso em uma metalúrgica. Gepros: Gestão da Produção, Operações e Sistemas, Bauru, 14(4), 114.
Catelan, M. C. F., Araujo, S. D., Fiorotto, D. J., & Carvalho, D. M. (2020). Heurísticas
para o problema de dimensionamento de lotes com máquinas paralelas flexíveis. TEMA, São Carlos, 21(2), 313-337.
Chwif, L. (1999). Redução de modelos de simulação de eventos discretos na sua concepção:
uma abordagem causal. 151 f. Tese (Doutorado em Engenharia Mecânica) -
Escola Politécnica, Universidade de São Paulo, São Paulo.
De Paula Ferreira, W., Armellini,
F., & De Santa-Eulalia, L. A. (2020). Simulation in industry 4.0: A
state-of-the-art review. Computers &
Industrial Engineering, 149.
Decker Junior, C. D., Henning, E.,
Ferreira, J. C. E., & Zappelino, B. F. (2020). Comparação dos projetos fatoriais
completo e fracionado em um modelo de simulação de eventos discretos em um
sistema de manufatura para os leiautes celular e celular virtual. Gepros: Gestão da Produção, Operações e Sistemas,
Bauru, 15(2), 23-57.
Ekren, B. Y., & Ornek, A. M. (2008). A simulation based
experimental design to analyse factors affecting
production flow time. Simulation
Modeling Practice and Theory, Amsterdam, 16, 278- 293.
Gaziero, C., & Cecconello, I.
(2019). Simulação
Computacional do Fluxo de Valor: uma proposta de Integração da Indústria 4.0 e
Lean Production. Scientia Cum Industria, 7(2), 52-67, 2019.
Goienetxea Uriarte
(2018). Ainhoa; NG, Amos HC; URENDA MORIS, Matías. Supporting the lean journey with simulation and
optimization in the context of Industry 4.0. In: Procedia Manufacturing. 586-59.
Hernandez, M. A. G., &
Librantz, A. F. H. (2013). Improvement of the supply chain for the sugar cane
exportation process employing discrete events simulation techniques. Acta Scientiarum,
Maringá, 35(4), 637-643.
Hirsch-Kreinsen
(2016). Hartmut. "Industry
4.0" as Promising Technology: Emergence, Semantics and Ambivalent
Character.
Junqueira, R. D. Á. R., & Morabito,
R. (2018). Programação e sequenciamento
das frentes de colheita de cana-de-açúcar: modelo e métodos de solução para
problemas de grande porte. Gestão & Produção, São Carlos,
25(1), 132-147. DOI: https://doi.org/10.1590/0104-530x2647-16.
Kagermann, Henning et al. (2013). Recommendations for implementing the strategic initiative INDUSTRIE
4.0: Securing the future of German manufacturing industry; final report of the Industrie 4.0 Working Group. Forschungsunion.
Liao, Yongxin
et al. (2017). Past, present and future of Industry 4.0-a systematic literature
review and research agenda proposal. International
journal of production research, 55(12), 3609-3629.
Lugert, Andreas, Batz, Aglaya, & Winkler, Herwig (2018). Empirical assessment
of the future adequacy of value stream mapping in manufacturing industries. Journal of Manufacturing Technology
Management.
Mancio, V. G., & Sellitto, M.
A. (2017). Sistemas
flexíveis de manufatura: definições e quadro de trabalho para futura pesquisa. Revista GEINTEC-Gestão, Inovação e
Tecnologias, São Cristóvão, 7(2), 3760-3773.
Morabito, R., & Pureza, V. (2010). Modelagem e simulação. In: Miguel, P.
A. C. (Org.). Metodologia de pesquisa em
engenharia de produção e gestão de operações. Rio de Janeiro: Elsevier, 165-194.
Negahban, Ashkan, & Smith,
Jeffrey S. (2014). Simulation for manufacturing system design and operation:
Literature review and analysis. Journal
of Manufacturing Systems, 33(2), 241-261.
Oliveira, W. A., & Santos, M.
O. (2017). Uma nova regra de ramificação para resolver o
problema de dimensionamento de lote capacitado e programação com configurações
dependentes de sequência. TEMA, São
Carlos, 18(3), 515-529, DOI: http://dx.doi.org/10.5540/tema.2017.018.
Peixoto, E. C., & Pinto, L. R. (2006).
Gerenciamento de estoques via previsão de vendas agregadas utilizando
simulação. Production (Online), 16(3), 569-581.
Sarkar, B., Guchhait,
R., Sarkar, M., Pareek, S., & Kim, N. (2019). Impact of safety factors and
setup time reduction in a two-echelon supply chain management. Robotics and Computer-Integrated
Manufacturing, New York, 55, 250-258.
Scheidegger, Anna Paula Galvão et al. (2018). An introductory guide for hybrid
simulation modelers on the primary simulation methods in industrial engineering
identified through a systematic review of the literature. Computers & Industrial Engineering, 124, 474-492.
Silva, W. A. (2005). Otimização de parâmetros da Gestão Baseada em Atividades (ABM) aplicada
em uma célula de manufatura. 2005. Dissertação (Mestrado em Engenharia de
Produção) - Universidade Federal de Itajubá, Itajubá.
Soares, J. P. M., Lemos, F. De O., Araújo, C. L. K.,
& Hansen, P. B. (2011). A contribuição da simulação computacional para a
análise sistêmica da reestruturação de layout e otimização de recursos na
manufatura celular: estudo de caso em uma célula de uma empresa do ramo
automotivo. Produto & Produção
(Online), [S. l.], 12(3), 49-68.
Soares, João Pedro M. et al. (2011). A contribuição da
simulação computacional para a análise sistêmica da reestruturação de layout e
otimização de recursos na manufatura celular: estudo de caso em uma célula de
uma empresa do ramo automotivo. Produto
& Produção (Online).
Ventura, M. M. (2007). O estudo de caso como
modalidade de pesquisa. Revista SoCERJ, 20(5), 383-386.
Wanke, P., & Saliby,
E. (2005). Proposta para a gestão de estoques de novos produtos: solução do
modelo (q, r) para a distribuição uniforme da demanda e do lead-time de
suprimento. Gestão & Produção,
São Carlos, 12(1), 1-20.
Yin. R. (2001). Estudo
de caso: planejamento e métodos. 2. ed. Porto Alegre, RS: Bookman.