A performance comparison between Kanban and CONWIP controlled assembly systems

There exists controversy on the superiority of logistics control systems. Kanban and CONWIP systems are focused on and analyzed in this paper. CONWIP is a well-known production control system, and some papers have shown it has better performance than the Kanban system. Our research shows that the Kanban is more flexible for the assembly system under concern with respect to a given objective than the CONWIP. In some cases, if the number of kanbans at each manufacturing/assembling station is optimally set, Kanban system outperforms CONWIP with a lower average WIP and the same level of throughput. That is, the distribution of kanbans can be an important design parameter of the system. We also propose two different policies to release cards in a CONWIP controlled assembly system, followed by their comparison results.     

Development and evaluation of a Basestock-CONWIP pull production control strategy in balanced assembly systems

In this study, a new pull production control strategy called Basestock-Constant Work-in-Process (B-CONWIP) is proposed. It is used to control the flow of materials and information in balanced assembly production systems. This proposed control strategy uses one type of authorization cards called CONWIP card that limits the work-in-process (WIP) in the whole system. It has been applied in a single-product and a mixed-product assembly system balanced by two efficient Genetic algorithms introduced in literature. The performance of this control strategy is compared with another pull production control strategy called Basestock Kanban CONWIP (BK-CONWIP), which is a very promising production control strategy found in literature. The proposed strategy has two control parameters, CONWIP authorization cards and basestock levels while BK-CONWIP has three control parameters Kanban authorization cards, CONWIP authorization cards and basestock levels. The comparison is based on three performance measures average system WIP, percentage of satisfied customer demand (service level) and WIP variation between workstations. The performance of the proposed strategy B-CONWIP and BK-CONWIP is mainly similar in both types of assembly systems when mean demand rates are low with respect to mean service rates with the proposed strategy being easier to control and optimize. On the other hand, when mean demand rates are high with respect to mean service rates; B-CONWIP is preferable if service level is more important, while BK-CONWIP is preferable if WIP level is more important. Regarding WIP variation, it mainly depends on the efficiency of the balancing approach. The more efficient the balancing approach, the less WIP variation. Treating demand as lost instead of backordered results in decreased average system WIP and does not affect service levels in both PCSs. It is also shown that S-KDP is more flexible in dealing with situations of variable product mixes than d-KDP because control parameters can be used by any product which minimizes the effect of the unbalanced systems.     

An integrated framework for card-based production control systems

Although card-based control systems, such as Kanban and CONWIP, for production processes have been successfully employed, the design discipline does not seem to be clear yet. Therefore, the superiority of one control over the other is controversial. This paper proposes a novel design discipline for card-based control of production processes, by developing the theory of token transaction systems. The theory shows how the three indices represented in Little??s law are decided by the structure of a production process with control cards and deployment of work-in-process (WIP). That is, the relation of WIP, cycle time and throughput on specific sub-network of a production process is clarified. We show how Little??s law should be used in the design of card-based production control systems. As an application of the theory, we resolve complicated result of comparison between Kanban and CONWIP. In doing so, this theory does not restrict the target of analysis to serial production lines, but any shaped processes can be analyzed.     

Selection of a pull production control policy under different demand situations for a manufacturing system by AHP-algorithm

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.     

CONWIP based control of a lamp assembly production line

Efficient and effective production control systems are very important for manufacturing plants. CONWIP, one of these production control systems, has a high potential of becoming the best one available because it suits a variety of production environments and is easy to implement. In the following paper, we compare the single-loop and multi-loop CONWIP production control systems for an actual lamp assembly production line producing different kinds of products with discrete distribution processing time and demand. A model is formulated with respect to total cost and service level. A novel rule-based genetic algorithm (GA) approach is proposed for the multi-loop CONWIP system to find the optimum parameter setting. The results have shown that the single-loop CONWIP production control system is more efficient than the multi-loop system. It can greatly decrease the total cost and the WIP (Work-In-Process) with zero shortage probability.     

Adaptive smart card-based pull control systems in context-aware manufacturing systems: Training a neural network through multi-objective simulation optimization

Nowadays, context aware manufacturing systems offer interesting capabilities to improve the performance of pull controlled production systems. Smart Kanbans can be used instead of physical cards, and the information become available about the production context, collected for example through sensors and RFID. Such information can be exploited by intelligent pull control strategies so as to dynamically adapt the number of cards. This is particularly useful for production systems that are subjected to unpredictable changes in the customers’ demand, and need to react quickly to preserve a high level of performance. For this reason, we aim, in this article, at proposing an intelligent system, which can communicate with the information system, whose purpose is to autonomously decide or to help managers in adding or removing cards. In this respect, we propose an approach that uses a neural network which is trained offline, directly from simulation, to decide when it is relevant to change the number of cards, and at what production stage. The learning process, based on multi-objective simulation optimization, aims at reducing the production costs as well as the number of changes to avoid nervousness. The use of stochastic simulation, allows various types of complex problems, related to manufacturing systems, to be addressed and fluctuating demand phenomena to be taken into account. The relevance of our approach is illustrated using six published adaptive ConWIP and Kanban systems. Comparisons with adaptive Kanban and ConWIP systems show that the neural network can automatically learn very relevant knowledge. Good results are obtained in terms of performance, with fewer changes in the number of cards. Several possible future research directions are pointed out.     

Stability of pull production control methods for systems with significant setups

In manufacturing, a pull production control method is a method of authorizing production based on replenishing current and past consumption, as opposed to forecasts or orders for future consumption. We consider special classes of pull systems for operations that involve significant setup times. In particular, we present a model for variations of the signal Kanban method and the pattern production method, each of which is used in industry when conventional Kanban methods are inappropriate. The paper examines such systems under demands with unpredictable load content but an upper bound on the total load. It is shown that, under appropriate conditions, such systems are stable in the sense that cumulative production at any time trails cumulative demand by no more than a constant. We determine buffer parameters under each protocol (including reorder points in the signal Kanban case) such that the backorder queue will clear to zero and remain empty thereafter. The results are then extended to consider multiple machines fulfilling production authorizations in parallel.     

Improvement of the evaluation of closed-loop production systems with unreliable machines and finite buffers

A closed-loop production system, or loop, is a system in which a constant amount of material flows through a single fixed cycle of workstations and storage buffers. Manufacturing processes that utilize pallets or fixtures can be viewed as loops. Control policies such as CONWIP and Kanban create conceptual loops. Gershwin and Werner (2007) developed a decomposition algorithm that accurately evaluates Buzacott type closed-loop systems of any size. However there are cases where the evaluated production rate, as a function of some system parameter, is discontinuous. Such a discontinuity may give misleading results for loop system design and optimization method. We present two modifications that improve the algorithm of Gershwin and Werner (2007). Two new special types of two-machine one-buffer building blocks are developed for the decomposition, and analytical solutions for them are found. Numerical experiments are provided to show the improvement of the evaluation accuracy as compared with the existing algorithm. The discontinuity in production rate is greatly diminished with these modifications.     

基于模糊控制的拉式策略在装配生产控制中的应用

利用模糊控制鲁棒性强的特点, 将其引入单产品装配线的生产控制系统中, 构造新的拉式控制策略. 在建立控制系统周期审查模型和完成控制器设计的基础上, 通过缩小变异范围改进遗传算法求解以极小化在制品量和投放波动为目标, 顾客满意率为约束条件的多目标规划, 从而达到优化控制系统参数的目的. 为考察模糊控制系统性能, 通过实例与多阶段定量在制品法(Constant work-in-process, CONWIP)、Kanban和一般拉式策略 (Generic pull, GP) 系统进行比较, 结果表明模糊生产控制系统不仅能维持较低的在制品水平, 更重要的是能维持较低的订单投放波动水平.     

Effect of product mix on multi-product pull control

Product mix influences the performance of pull production control strategy in multi-product manufacturing systems. The complexity of product mix on the performance of a manufacturing system is primarily based on the characteristics of the demand and production control strategies. Demands are mainly characterised by volume and product-type while production control strategy is characterised by material release time, part flow, inventory control and throughput times. In multi-product systems, pull production control strategy operates dedicated or shared Kanban allocation policy. This paper examines the performance of the Generalised Kanban Control Strategy (GKCS), Extended Kanban Control Strategy (EKCS) and Basestock Kanban-CONWIP (BK-CONWIP) control strategy operating Shared Kanban Allocation Policies (S-KAP) or Dedicated Kanban Allocation Policies (D-KAP) for a healthcare parallel/serial assembly line with setup times. A simulation based multi-objective optimisation technique was adopted to examine the effect of different product mixes on the strategies and policies. A ranking and selection technique for multiple systems was used to screen the performance of the strategies. It was shown that product mix variability in a system influence the inventory levels of the pull control strategies examined. However, the performances of the strategies vary with strategies operating S-KAP having better inventory control than strategies operating D-KAP. Similarly, BK-CONWIP outperformed its alternatives.     

Token-based pull production control systems: an introductory overview

Since the advent of Kanban, pull systems have been widely used in practice and have been widely studied by researchers. During the last decades, several types of pull systems have emerged. Numerous articles have been published to introduce new paradigms or new principles for just-in-time systems as well as new approaches to evaluate them or to optimize their performance. An important feature of these systems, which is common to many other production control systems, is the use of tokens, which usually consist of cards that authorize certain production tasks to be performed. Tokens can be used in various manners to control production and can be combined with several other mechanisms with the objective of reducing the work in progress and the lead times, while meeting customers’ demand. This article proposes an introductory overview of existing research works in this area. In this respect, we suggest a classification of pull-inspired production control systems, which allows us to distinguish up to 18 different systems. For each type of system, we study its basic principles, the flow control strategy, and the parameters affecting its performance. This survey aims at facilitating the understanding of the different proposals made by researchers and highlighting their common points and differences.     

供应链存储系统无线射频识别使能的定量在制品法控制策略仿真

无线射频识别(radio frequency identification,RFID)技术的出现为库存控制策略的创新提供了一条有效的途径.本文利用RFID技术,将原有单个制造系统的流通卡系统推广到供应链的多级库存网络中,设计了RFID使能的电子流通卡系统.基于电子流通卡,实现了RFID使能的定量在制品法(constant work-in-process,CONWIP)控制策略;并建立了CONWIP控制策略的分销网络模型.为比较和验证CONWIP控制策略性能,开发了三级分销网络仿真系统,使用该系统,将其与RFID使能的(r,Q)策略进行比较,验证了CONWIP策略在一定条件下,能够有效降低存储成本、提高服务水平.     

Sustainable MSD prevention: Management for continuous improvement between prevention and production. Ergonomic intervention in two assembly line companies

To increase output and meet customers' needs, companies have turned to the development of production management systems: Kaizen, one piece flow, Kanban, etc. The aim of such systems is to accelerate decisions, react to environmental issues and manage various productions. In the main, this type of management system has led to the continuous improvement of production performance. Consequently, such production management systems can have unexpected negative effects on operators' health and safety. Conversely, regulation and control systems focusing on work-related risks have obliged firms to implement health and safety management systems such as OHSAS 18001. The purpose of this type of system, also based on continuous improvement, is to reduce risks, facilitate work-related activities and identify solutions in terms of equipment and tools. However, the prevention actions introduced through health and safety systems often result in other unexpected and unwanted effects on production. This paper shows how companies can improve the way they are run by taking into account both types of management system.     

A new fuzzy methodology-based structured framework for RAM and risk analysis

The aim of this paper is to propose a new hybridized framework for analyzing the performance issues of a chemical process plant by utilizing uncertain, imprecise and vague information. In the proposed framework, Fuzzy Lambda–Tau (FLT) approach has been used for computing reliability, availability and maintainability (RAM) parameters of the considered system. Further, for enhancing the RAM characteristics of the system, improved Fuzzy Failure Mode Effect Analysis (FMEA) approach is adopted. Under improved Fuzzy FMEA approach, defined Fuzzy linguistic rating values in the form of triangular and trapezoidal Fuzzy numbers have been assigned by the experts to each risk factor of the listed failure causes. The proposed framework is demonstrated with an industrial application in a chlorine production plant of a chemical process industry. The results show decreasing trend for system availability and deposition of solid Nacl, mechanical failure, corrosion due to wet chlorine, scanty lubrication, improper adsorption and valve malfunctioning are identified as the most critical failure causes for the considered system. A comparative performance analysis between the proposed framework, Fuzzy technique for order of preference by similarity to ideal solution (Fuzzy TOPSIS), Fuzzy evaluation based on distance from average solution (Fuzzy EDAS) and Fuzzy Vlse Kriterijumska Optimizacija I Kompromisno Resenje (Fuzzy VIKOR) are then carried out to show the competence of the proposed framework. It is expected that the analytical results would be highly useful in formulating an optimal maintenance policy for such complex systems and may also be used for improving performance of similar plants.     

Dynamic programming model for multi-stage single-product Kanban-controlled serial production line

The executive concern of this paper is how to control and synchronize the flow of materials in Kanban controlled serial production line so as to build a dynamic material-flow system that successfully meets customer demand Just-In-Time. The proposed approach should yield a consistent integrated control policy with a feasible level of Work-In-Process and a feasible corresponding operational cost. The production line is described as queuing network, and then a Dynamic Programming (DP) algorithm is used to solve the network by decomposing it into several numbers of single-stage sub-production lines. Backward computations of DP are done recursively with synchronization mechanism, in the since that the solution of one sub-production line is used as an input to the previous one. A performance measure is then developed to determine and to compare the values of production parameters. Numerical examples are used to demonstrate the computations of different system parameters, the results are validated by discrete events simulation using ProModel software version 6.0, the performance measure coincided with the results of the model with very small error (0.044). As a result the number of Kanbans that are needed to deliver the batches from upstream stage to the downstream stage is determined in such a way that keeps the stages synchronized with the external customer demand.     

A case study in the design of BTO/CTO shop floor control system

Although computerized information systems are conventionally used to control production, applying a manual approach and improper logistical assignments typically reduces operating efficiency and increases customer service response time. Satisfying the requirements of remaining competitiveness in the current marketplace, such as offering build-to-order (BTO) or configuration-to-order (CTO) services is difficult. This study presents an integrated information system for use in shop floor controlling systems (SFCS) for kitting parts, producing goods, packing finished goods and distributing them, to enhance the performance of the BTO/CTO production system. Several information technologies and devices, such as barcode system, the electronic pick-to-light picking system, electronic Kanban (eKanban), and others are adopted to support the relevant logistics. A personal computer (PC) manufacturer is used to demonstrate the proposed system. The findings of the study reveal that the proposed information system can respond to the status of the production line in real time. This responsiveness can help the production controller to make a suitable decision and quickly meet the requirements of the BTO/CTO production system.     

多重入生产线的CONWIP投放方法研究

CONWIP（Constant Work in Process）是一种基于拉式（pull）控制物料投放的方法。最初应用于Flow Shop式生产系统,通过限制半成品数量以控制生产线运作提供可靠的制造周期。然而,原始的CONWIP方法不能适应半导体这类同一加工站多重人的生产系统。该文根据半导体生产线多重人的特性,提出与多重人瓶颈相结合的CONWIP投放方法,不仅控制生产线中的半成品数量,并且结合了TOC（Theory of Constraints）思想,从而提高生产线的性能。最后,通过一些仿真证明了方法的有效性。     

复杂需求环境下中小型制造企业作业系统建模与分析

针对中小型制造企业（small medium production enterprises,SMPEs）对复杂需求的响应速度普遍比较慢的问题,从调整系统参数合理性的角度提出了一种提高系统响应速度的方法.首先,构建了融合员工关键技能的SMPEs生产作业系统;其次,基于连续时间马尔可夫链建立系统的稳态平衡方程模型,在采用两种不同控制策略的情况下,深入研究了生产线的切换时间等参数对系统性能的影响;最后,在给定的生产实例下,通过理论推导和仿真实验分析了系统性能指标随着关键参数的变化趋势.实验结果表明,理论推导结果与实际生产的运行情况吻合良好,为SMPEs生产企业进行合理的系统参数设计提供了重要的基础.     

A flexible and adaptable planning and control system for an MTO supply chain system

Evolving consumer demand has led to the emergence of a new business era where change plays a dominant role. This new era has in turn posed a number of competences that supply chain systems need to posses so as to survive and prosper under change and uncertainty. The main aim of this work is to model the performance of an make-to-order (MTO) production environment that operates as part of a wider supply chain network under various control policies and identify the best performing ones. A simulation model has been created and used as a tool to help quantify the effects of three control policies on this MTO system. A number of performance metrics such as Work-in-process (WIP), tardiness, responsiveness and mean flow time have been considered to provide further insight into the impact of these three localised control policies. The three control policies implemented were push, and two modes of pull control, which are, CONWIP and tight pull.     

A performance evaluation framework for association mining in spatial data

The evaluation of the process of mining associations is an important and challenging problem in database systems and especially those that store critical data and are used for making critical decisions. Within the context of spatial databases we present an evaluation framework in which we use probability distributions to model spatial regions, and Bayesian networks to model the joint probability distribution and the structural relationships among spatial and non-spatial predicates. We demonstrate the applicability of the proposed framework by evaluating representatives from two well-known approaches that are used for learning associations, i.e., dependency analysis (using statistical tests of independence) and Bayesian methods. By controlling the parameters of the framework we provide extensive comparative results of the performance of the two approaches. We obtain measures of recovery of known associations as a function of the number of samples used, the strength, number and type of associations in the model, the number of spatial predicates associated with a particular non-spatial predicate, the prior probabilities of spatial predicates, the conditional probabilities of the non-spatial predicates, the image registration error, and the parameters that control the sensitivity of the methods. In addition to performance we investigate the processing efficiency of the two approaches.