International Journal of Research and Scientific Innovation (IJRSI) | Volume V, Issue IV, April 2018 | ISSN 2321–2705
A Review of Kanban-based Manufacturing Systems
Shobha N S*, Dr. Subramanya K N**
*
Assistant Professor, ** Principal and Professor,
Department of Industrial Engineering and Management, R V College of Engineering, Bengaluru – 560 059, Karnataka, India
Corresponding Author: Shobha N S
Abstract: - The purpose of this paper is to review the earlier
research works carried out on Kanban based manufacturing and
the analytical tools used to determine number of Kanbans. The
review is made by considering various factors including
variations of Kanban, production systems adopted in various
types of organizations, and analytical tools and techniques
adopted for modeling the system. One of the important finding is
the behavior of modified Kanban and its effects on production
systems. The results of various studies would also help in
designing a novel production system which might be better than
the existing systems. This review serves as guidelines for
implementing Kanban system in various production systems and
also to identify the critical parameters.
Keywords: Kanban, Simulation, Manufacturing, Optimization
I. INTRODUCTION
K anban is a framework used as a tool in lean assembly
lines in various organizations. Kanban method gained
popularity in automobile sector and served as an effective tool
to support production system as a single unit. It helps to
achieve improvement in a better way. Kanban recognizes the
upper limit in the work in process inventory, thus, reducing
overload in a manufacturing system environment. The Kanban
framework utilizes cards for withdrawal, store, and different
signs which creates and hones a smoother stream of processes
in the production line. The utilization of signals and images
will send a message to the assembling group. The Kanban is
used to for speedy reaction between suppliers, manufacturer
and warehouse and reduce the overall time of the system.
There is many ways to use a Kanban system to create a
uniform flow. For example, in an automated or a partially
automated manufacturing unit, communication with various
departments are done using electronic signals such as visual
display systems known as Electronic Kanban.
It is important to remember that the materials, spare parts, and
different components that are expected to create particular
items are constantly required in stock with the goal to meet
the customer requirements. Kanban system expects perfect
networking and co-ordination among various stages in the
shop floor. It is good for organizations to investigate and
watch these connections to see more about each group's
shortcoming and qualities with the goal to meet the customer
demand. The Kanban arrangement brings some measure of
positive change in the workplace. The Kanban system has
positive reactions in assembly lines, while it is still to be
tested in administration and IT / ITES organizations.
www.rsisinternational.org
Subsequently, this paper investigates different examinations
which endeavored to "enhance" Kanban framework to
accomplish wanted outcomes.
II. APPLICATION OF OPTIMIZATION TECHNIQUES TO
DESIGN KANBAN SYSTEM
The main purpose is to analyze various optimization
techniques used by researchers. The earlier works are
presented in a chronological order of the usage.
Kimura and Terada have proposed an arrangement of balance
conditions for a Kanban-based serial creation framework.
They depict the framework by an arrangement of three
relations including production order quantities, time lag
between withdrawal and production of Kanban, amplification
at a certain workstation to decide the variances at the last
workstation. The authors have used simulation tool to model
Kanban system [1]. The authors have also improved on serial
production lines, bottleneck workstations and assembly job
shop by using Kanban system [2]. Decision support system
models are used for production control and visually represent
production capacity, production runs, order changes, machine
stacking, work in progress and completion time and also
supervisor and work station productivity. Kanban system is
also implemented in a flexible manufacturing system using
different allocation rules for Kanban [3].
The simulation using ARENA, mathematical programming
and stochastic modelling tools are used for operational
flexibility of kanbans. Also, Markov process models are used
to model the system behavior and determine the optimal
number of Kanbans. The authors have developed Incremental
Optimization algorithm to maximize the efficiency of Kanban
by creating auto-adjusting kanbans at every stage of
production control. Stochastic algorithm is also proposed to
tackle variants of Kanban scenarios [4]. An Adaptive Kanban
pull type mechanism is developed by Tardif and Maaseidvaag
to determine release re-order raw material parts based on
demand and inventory. A single-stage, single- product Kanban
system following Poisson process and exponential processing
times are considered. Simulation is used to achieve the desired
results and also highlight drawbacks on multi-stage,
multiproduct Kanbans [5].
Kochel and Nielander provide an optimal design of Kanban
system to a multistage production system using simulation
and evolution approach. The simulation approach involves
Page 244
International Journal of Research and Scientific Innovation (IJRSI) | Volume V, Issue IV, April 2018 | ISSN 2321–2705
simulation and optimization tools together to optimize the size
of the Kanban system [6]. Dub et al optimized the stock
reserves at an existing assembly line using Stochastic time
Petri Nets. They use Petri Nets because the whole assembly
line cannot be modelled as a single unit. The results depicted
improvement in storage space and risk factor. Subsequent to
use of Stochastic Petri Nets, simulation was used as tool to
optimize number of Kanbans at every stage [7].
Krishnamurthy and Suri have proposed an algorithm
depending on the parametric portrayal of the movement forms
in the system and utilizations two-moment approximations to
evaluate execution measures at each individual
station/arrange. A set of non-linear equations are obtained to
estimate performance of Kanban System [8]. Sendilkumar &
Panneerselvam have made a literature survey of Kanban and
JIT based system and also on optimization of Kanban systems
[9]. Ramanan and Rajendran have proposed a heuristic based
simulated annealing to reduce the makespan of containers in a
lean environment [10]. The simulated annealing technique is
again used by Liang et.al. to design a multiproduct Kanban
system based on bottleneck analysis. The Simulated
Annealing Algorithm took less computational efforts and also
run with less time [11].
Al-Tahat and Mukattash designed a production control
scheme for a traditional Kanban-based JIT environment. A
synchronization mechanism for a single-stage single-product
Kanban-controlled production line is developed resulting in
achieving feasible operating costs as well as average WIP
[12]. Lage and Filho have done a survey to address the
modified Kanban system by considering the quantity
preserved in the variety, the operational contrasts between
every variety and the customary Kanban, the focal points in
connection to the first Kanban and the inconveniences in
connection to the conventional Kanban [13].
III. DESIGNING MULTI-STAGE KANBAN SYSTEM
In the previous section, most of the studies were done on
single-stage, product Kanban systems, but, there were cases of
modified Kanban that were focused on research. In this
section, the use of Multi-stage Kanban system is explored.
Shivakumar and Shahabudeen have designed the multi-stage
Kanban system by disintegrating into stages. Each stage is
controlled by a Kanban mechanism. The parameters of the
control are the number of kanbans for each stage. These
parameters assume an imperative part of the effectiveness of
the Kanban control system. The assumptions include an
infinite supply of raw parts at starting stage and ceaseless
demand. The administration time is exponentially distributed
and demand takes after Poisson. The average arrival rate of
demand less that the throughput [14].
Genetic algorithm is used to calculate the Kanban solutions by
Kochel and Nielander . Now, a detailed description about all
the stages is made in the study using crossover and mutations.
A Kanban system is designed in a multi-stage, mixed-model
www.rsisinternational.org
assembly line by L.Yang et.al. The model considers
production processes, estimation of setup time, the WIP in the
buffer area, the Kanban limit, and the takt-time proportion to
enhance the activity execution of the Kanban system. The
outcomes demonstrate the Kanban framework, smaller the
Kanban limit leading to lower stock-out and diminishing the
Kanban limit implies expanding the WIP [15]. The study was
conducted by Liang on Kanban controlled pull systems with
exponential machines and infinite buffer. The measure of
performance including customer service level, finished goods
inventory, and the release time in the line were considered.
The outcome reveals that longer the up and downtime levels,
the lower the customer service and insists on maintaining a
lean number of Kanban [11].
IV. SUMMARY
In light of the above studies, Kanban method is widely
beneficial regardless of its originality or modification. This is
mainly due to the fact that, it focusses mainly on the method
rather than the process. The primary target of this paper was
to conduct a literature survey to distinguish, group, and break
down the Kanban framework varieties proposed by different
authors. A large portion of the changes proposed to the first
Kanban framework is in a general sense worried about the
utilization of signals, which build up methods for controlling
the signals number or quantity. Deciding the correct number
of signals is one of the important issues in the hypothetical
investigation of Kanban based manufacturing system. There
may be limitations in some cases nevertheless advanced
systems may prove to be fruitful in the research study of
Kanban systems.
REFERENCES
[1]. Kimura, O., & Terada, H. (1981), “Design and analysis of Pull
System, a method of multi-stage production control”, The
International Journal of Production Research, 19(3), 241-253.
[2]. Price, W., Gravel, M., & Nsakanda, A. L. (1994), “A review of
optimization models of Kanban-based production systems”,
European Journal of Operational Research, 75(1), 1-12.
[3]. Sengupta, S., Sharief, F., & Dutta, S. P. (1999), “Determination of
the optimal number of kanbans and kanban allocation in a FMS: A
simulation based study”, Production Planning & Control, 10(5),
439-447.
[4]. Panayiotou, C. G., & Cassandras, C. G. (1999), “Optimization of
kanban-based manufacturing systems”, Automatica, 35(9), 1521-
1533.
[5]. Tardif, V., & Maaseidvaag, L. (2001). “An adaptive approach to
controlling kanban systems”, European Journal of Operational
Research, 132(2), 411-424.
[6]. Kochel, P., & Nieländer, U. (2002), “Kanban optimization by
simulation and evolution”, Production Planning & Control, 13(8),
725-734.
[7]. Dub, M., Pipan, G., & Hanzálek, Z. (2002), “Stock optimization of
a kanban-based assembly line”, In Proc. of the 12th int. conf. on
flexible automation and intelligent manufacturing, Dresden,
Germany, 1-10.
[8]. Krishnamurthy, A., & Suri, R. (2006), “Performance analysis of
single stage kanban controlled production systems using
parametric decomposition”, Queuing Systems, 54(2), 141-162.
[9]. Kumar, C. S., & Panneerselvam, R. (2007), “Literature review of
JIT-KANBAN system”, The International Journal of Advanced
Manufacturing Technology, 32(3-4), 393-408.
Page 245
International Journal of Research and Scientific Innovation (IJRSI) | Volume V, Issue IV, April 2018 | ISSN 2321–2705
[10]. Ramanan, G. V., & Rajendran, C. (2003), “Scheduling in kanban-
controlled flow shops to minimise the makespan of containers”,
The International Journal of Advanced Manufacturing
Technology, 21(5), 348-354.
[11]. Liang, H., Yuan, G., Feng, Q., Suzhou, T., & Delun, W. (IEEE,
February 2010), “Design of a multi-product kanban system based
on bottleneck analysis”, In Computer and Automation Engineering
(ICCAE), Vol. 4, 80-83.
[12]. Al-Tahat, M. D., & Mukattash, A. M. (2006), “Design and
analysis of production control scheme for Kanban-based JIT
environment”, Journal of the Franklin Institute, 343(4-5), 521-
531.
[13]. Junior, M. L., & Godinho Filho, M. (2010), “Variations of the
kanban system: Literature review and classification”, International
Journal of Production Economics, 125(1), 13-21.
[14]. Sivakumar, G. D., & Shahabudeen, P. (2008), “Design of multi-
stage adaptive kanban system”, The International Journal of
Advanced Manufacturing Technology, 38(3-4), 321.
www.rsisinternational.org
[15]. Yang, L., Zhang, X., & Jiang, M. (2010), “An optimal Kanban
system in a multi-stage, mixed-model assembly line”, Journal of
Systems Science and Systems Engineering, 19(1), 36-49.
[16]. Ananth Krishnamurthy & Rajan Suri. (2006), “Performance
analysis of single stage kanban controlled production systems
using parametric decomposition”. Queuing Syst Vol.54, 141–162.
[17]. Zhang, L. (2013), “Kanban-controlled exponential production
lines: analysis and design”, Journal of Manufacturing Technology
Management, 24(3), 358-383.
[18]. Ghelichi, A., & Abdelgawad, A. (IEEE, December 2014), “A
study on RFID-based Kanban system in inventory management”,
Industrial Engineering and Engineering Management (IEEM),
1357-1361.
[19]. Tregubov, A., & Lane, J. A. (2015), “Simulation of Kanban-based
scheduling for systems of systems: initial results”, Procedia
Computer Science, 44, 224-233.
Page 246