Design of bi-criteria kanban system using simulated annealing technique

doi:10.1016/S0360-8352(01)00060-2

Computers & Industrial Engineering

Volume 41, Issue 4, February 2002, Pages 355-370

https://doi.org/10.1016/S0360-8352(01)00060-2 Get rights and content

Abstract

In the kanban system, the main decision parameters are the number of kanbans and lot size. In this paper, an attempt has been made to set the number of kanbans at each station and the lot size required to achieve the best performance using simulated annealing technique. A simulation model with a single-card system has been designed and used for analysis. A bi-criterion objective function comprising of mean throughput rate and aggregate average kanban queue has been used for evaluation. Different perturbation schemes have been experimented and compared.

Introduction

Just in time (JIT) production system is the manufacturing philosophy of producing what is needed at the right time and in right quantity (Hutchins, 1993). Kanban coupled with pull system of production is used as means of implementing JIT. Kanban means a ‘visible card’, which serves as a planning and information tool to smoothen the flow of material through the manufacturing and assembly process. The workstations located along the production lines only produce or deliver desired components when they receive a card and empty container, indicating that more parts will be needed in production. Each workstation will only produce enough components to fill containers and then stop. In addition, kanban limits the amount of inventory in the process by acting as an authorisation to produce.

The essential elements in the design of kanban production system are the number of kanbans needed to link processes together and the appropriate unit of lot size (Berkley, 1992a).

Section snippets

Literature review

Kanban based operational planning and control issues have been tackled in a number of studies by means of analytical or simulation modelling. Berkley (1992a) has reviewed 50 papers in the area of kanban production control and organised them based on the type of system. He has also listed 24 vital operation design factors of kanban system. Price, Gravel and Nsakanda (1994) have reviewed optimisation models of kanban based systems. They have concluded that interesting direction for future

Bi-criteria objective

It is found that, most of the researchers have used only single objective function as the performance measure. In this study, a bi-criteria objective kanban problem has been considered. The first criterion is the maximisation of mean cumulative throughput rate, which is defined as the ratio of total satisfied demand to the total generated demand. The other criterion is the minimisation of aggregate average kanban queue, which is the sum of average number of kanbans waiting in the queue at all

Simulated annealing

Simulated annealing is a search heuristic intensively used to solve optimisation problems and especially combinatorial ones. It was derived from an analogy to the physical annealing process and first introduced by Kirkpatrick et al., 1983, Cerny, 1985. Pirlot, 1996, Eglese, 1989, Sridhar and Rajendran, 1993 described simulated annealing and its implementation with some choice parameters. A generic simulated annealing algorithm for a maximization problem is given below:

Step 1: Select an initial

The kanban system model

The system under study is dynamic kanban system with stochastic demand and processing times with the following assumptions.

Experimental design and results

Experiments to decide the run length, to set a limit for number of kanbans and lot size are performed. The details are as follows.

Conclusion

A simulation model of the single-card kanban system has been developed to determine the number of kanbans and lot size. A bi-criteria objective function consisting of throughput and aggregate kanban queue has been employed to obtain a solution that maximises the objective function value. Simulated annealing algorithm has been employed to search the solution space. Two types of perturbation schemes have been tried out. The results showed that there is no significant difference between the two

References (28)

G. Abdou et al.
A systematic simulation approach for the design of JIT manufacturing systems
Journal of Operations Management
(1993)
T. Fukukawa et al.
The determination of optimum number of kanbans in a just in time production system
Computers and Industrial Engineering
(1993)
M. Pirlot
General local search methods
European Journal of Operational Research
(1996)
W. Price et al.
Review of optimisation models of kanban based production systems
European Journal of Operations Research
(1994)
H. Wang et al.
Optimum number of kanbans between two adjacent workstations in a JIT system
International Journal of Production Economics
(1991)
A. Andijani
Trade off between maximizing throughput rate and minimizing system time in kanban systems
International Journal of Operations and Productions Management
(1997)
J.F. Bard et al.
Determining the number of kanbans in a multiproduct multistage production system
International Journal of Production Research
(1991)
B.J. Berkley
A review of the kanban production control research literature
Production and Operations Management
(1992)
B.J. Berkley
A decomposition approximation for periodic kanban controlled flow shops
Decision Sciences
(1992)
B.J. Berkley
Simulation tests of FCFS and SPT sequencing in kanban systems
Decision Sciences
(1993)

B.J. Berkley

Testing minimum performance levels for kanban-controlled lines

International Journal of Production Research

(1994)

B.J. Berkley

A simulation study of container size in two-card kanban system

International Journal of Production Research

(1996)

B.J. Berkley et al.

A simulation study of sequencing rules in a kanban — Controlled flow shop

Decision Sciences

(1991)

G.R. Bitran et al.

A mathematical programming approach to a deterministic kanban system

Management Sciences

(1987)

Cited by (26)

An optimal maintenance strategy for multi-state deterioration systems based on a semi-Markov decision process coupled with simulation technique
2020, Mechanical Systems and Signal Processing
Citation Excerpt :
However, when the size of a system becomes large, such strategies are likely to suffer from the curse of dimensionality. To address this issue, several researchers [38–41] proposed simulation-based policy iteration methods to reduce the size of the search space for SMDP. For example, Belomestny et al. [38] presented a multilevel simulation-based policy iteration algorithm to reduce the complexities of optimal stopping problems.
Maintenance optimization of multi-state systems is a research topic of practical significance for various manufacturing systems. Nevertheless, design of satisfactory maintenance strategies remains a challenging problem due to the complex industrial field and large strategy space for maintenance optimization of such systems. In this study, a model of the deterioration process of multi-state systems with the multi-phase property was constructed by utilizing semi-Markov decision theory and linear systems of integral equation. An efficient maintenance strategy was developed with the purpose of maximizing the net revenue per unit period based on the simulation technique. In the constructed model, a system of linear integral equations with a recursive structure was constructed to obtain all possible transition probabilities between the states. Due to this recursive structure, conventional solutions of linear integral equations are not capable of solving this equation. For this reason, an algorithm was developed that was instrumental in resolving this issue. In the designed maintenance strategy, to avoid the potential curse of dimensionality resulting from the scales of the systems and the strategy space, a semi-Markov decision process coupled with the simulation technique was adopted in each decision period to reduce the search space by eliminating unreachable states. Finally, numerical examples of two systems with different scales based on real data of the inspection record database at Dongbei Special Steel Group Co., Ltd. are given to illustrate the effectiveness of the proposed model and the performance of the designed strategy. The numerical results demonstrate the high efficiency of the optimal strategy, especially for large-scale systems.
The optimal control of just-in-time-based production and distribution systems and performance comparisons with optimized pull systems
2011, European Journal of Operational Research
Citation Excerpt :
Their conclusion was that tabu search is the best among the three. Shahabudeen et al. (2002, 2003) also proposed a simulation-based optimization algorithm using simulated annealing to find an optimal setting for kanban systems. Recently, Koulouriotis et al. (2010) proposed a simulation-based optimization algorithm using genetic algorithms to find optimal settings for pull systems such as base stock, kanban, CONWIP, hybrid and extended kanban.
In just-in-time (JIT) production systems, there is both input stock in the form of parts and output stock in the form of product at each stage. These activities are controlled by production-ordering and withdrawal kanbans. This paper discusses a discrete-time optimal control problem in a multistage JIT-based production and distribution system with stochastic demand and capacity, developed to minimize the expected total cost per unit of time. The problem can be formulated as an undiscounted Markov decision process (UMDP); however, the curse of dimensionality makes it very difficult to find an exact solution. The author proposes a new neuro-dynamic programming (NDP) algorithm, the simulation-based modified policy iteration method (SBMPIM), to solve the optimal control problem. The existing NDP algorithms and SBMPIM are numerically compared with a traditional UMDP algorithm for a single-stage JIT production system. It is shown that all NDP algorithms except the SBMPIM fail to converge to an optimal control.
Additionally, a new algorithm for finding the optimal parameters of pull systems is proposed. Numerical comparisons between near-optimal controls computed using the SBMPIM and optimized pull systems are conducted for three-stage JIT-based production and distribution systems. UMDPs with 42 million states are solved using the SBMPIM. The pull systems discussed are the kanban, base stock, CONWIP, hybrid and extended kanban.
Determination of number of kanbans in a supply chain system via Memetic algorithm
2009, Advances in Engineering Software
Citation Excerpt :
Moattar Husseini et al. [5] present a method to enhance the flexibility of the system that is controlled by. Shahabudeen et al. [11] calculate optimum number of kanbans and batch sizes via simulated annealing. Moreover, numerous models are developed to describe the supply chain system.
In this research multi-stage supply chain system which is controlled by kanban system, is evaluated. In kanban system, decision making is based on determination of the number of kanbans as well as batch sizes. This paper attempts to model supply chain system with regard to costs under just-in-time (JIT) production philosophy. Since adopted model is of mixed integer non-linear programming (MINLP) type and solving it by exact algorithm such as branch and bound (B&B) takes a lot of time, a heuristic method via Memetic algorithm (MA) is presented. Some problems are solved by our proposed MA to illustrate its performance.
Selection of a pull production control policy under different demand situations for a manufacturing system by AHP-algorithm
2009, Computers and Operations Research
Citation Excerpt :
Petroni and Rizzi [21] used average WIP, average flow (production lead) time, mean tardiness as performance measures for predicting the performance of a manufacturing system. Shahabudeen et al. [22,23] design single and two cards dynamic kanban systems using a simulated annealing algorithm. They proposed a universal test that may be suited for the MOP (measure of performance) in any JIT system, which are percentage zero demand (PZD), mean lead-time (MLT) and mean total WIP (MTW).
A multistage serial production system is considered in the present paper. A generalized model has been developed with the use of probabilistic demand situations for the end product. The demand situations considered are binomial, exponential, lognormal and Poisson. These demand patterns are used as input parameter for various production control policies. The output values for performance parameters are obtained by simulation. The production policies analyzed are Kanban, CONWIP and Hybrid as alternatives for controlling the engineering manufacture. In the presence of several performance measures, it often becomes difficult for the management to select the most appropriate policy. Analytic hierarchy process (AHP) has been implemented for the purpose of selection. Computational results have been reported along with the sensitivity analysis after designing and conducting various experiments.
Algorithm for the design of single-stage adaptive kanban system
2008, Computers and Industrial Engineering
The traditional kanban system with fixed number of cards does not work satisfactorily in unstable environment. In the adaptive kanban-type pull control mechanism the number of kanban is allowed to change with respect to the inventory and backorder level. It is required to set the threshold values at which cards are added or deleted which is a part of the design. Previous studies used the local search method to design the adaptive kanban system. In this paper Genetic Algorithm- and Simulated annealing-based heuristics are developed and used to set the design parameters of adaptive kanban system. The numerical results indicate that simulated annealing based heuristics produces better solution with improved computational efficiency.
Improvement of Manufacturing Process Based on Value Stream Mapping: A Case Study
2023, EMJ - Engineering Management Journal

View all citing articles on Scopus

View full text

Design of bi-criteria kanban system using simulated annealing technique

Abstract

Introduction

Section snippets

Literature review

Bi-criteria objective

Simulated annealing

The kanban system model

Experimental design and results

Conclusion

Journal of Operations Management

Computers and Industrial Engineering

European Journal of Operational Research

European Journal of Operations Research

International Journal of Production Economics

Trade off between maximizing throughput rate and minimizing system time in kanban systems

International Journal of Operations and Productions Management

Determining the number of kanbans in a multiproduct multistage production system

International Journal of Production Research

A review of the kanban production control research literature

Production and Operations Management

A decomposition approximation for periodic kanban controlled flow shops

Decision Sciences

Simulation tests of FCFS and SPT sequencing in kanban systems

Decision Sciences

Testing minimum performance levels for kanban-controlled lines

International Journal of Production Research

A simulation study of container size in two-card kanban system

International Journal of Production Research

A simulation study of sequencing rules in a kanban — Controlled flow shop

Decision Sciences

A mathematical programming approach to a deterministic kanban system

Management Sciences