Tabu search-based algorithm for the TOC product mix decision

The theory of constraints (TOC) is a management philosophy that maximizes profits in a manufacturing plant with a demonstrated bottleneck. The product mix decision is one application of TOC that involves determination of the quantity and the identification of each product to produce. However, the original TOC heuristic is considered to produce unrealizable solution when a manufacturing plant has multiple resource constraints. This paper presents a tabu search-based TOC product mix heuristic to identify optimal or near optimal product mix for small problem instances under conditions where the original TOC heuristic failed. The tabu search-based TOC product mix heuristic is further used to solve large problem instances typical of practical manufacturing scenario. The experimental results for small to medium size problem show that the tabu search-based TOC heuristic compares favourably with those of optimal methods. Large size problems for which optimal methods have not been established in terms of feasibility in computation times were also solved in reasonable times with good quality solutions, thus confirming that the proposed approach is appropriate for adoption by production planners for the product mix problem in the manufacturing industry.     

When less is better: Insights from the product mix dilemma from the Theory of Constraints perspective

Perhaps due to its origins in a production scheduling software called Optimised Production Technology (OPT), plus the idea of focusing on system constraints, many believe that the Theory of Constraints (TOC) has a vocation for optimal solutions. Those who assess TOC according to this perspective indicate that it guarantees an optimal solution only in certain circumstances. In opposition to this view and founded on a numeric example of a production mix problem, this paper shows, by means of TOC assumptions, why the TOC should not be compared to methods intended to seek optimal or the best solutions, but rather sufficiently good solutions, possible in non-deterministic environments. Moreover, we extend the range of relevant literature on product mix decision by introducing a heuristic based on the uniquely identified work that aims at achieving feasible solutions according to the TOC point of view. The heuristic proposed is tested on 100 production mix problems and the results are compared with the responses obtained with the use of Integer Linear Programming. The results show that the heuristic gives good results on average, but performance falls sharply in some situations.     

TOC会计

主要介绍TOC的基本原理、TOC会计以及产品组合。     

TOC-based performance measures and five focusing steps in a job-shop manufacturing environment

Theory of constraints (TOC) views a company as a set of interdependent processes working in harmony to achieve the profit goal of the company as a whole, and thus it emphasizes total system performance over localized measures to guide operational decisions. This paper demonstrates the usefulness of employing TOCbased global performance measures to make operational-decisions (e.g. product mix, continuous improvement, inventory management, production planning and scheduling) to strengthen the internal supply chain in a relatively complex manufacturing environment, i.e. a job shop. An ARENA-based simulation model is presented and a number of scenarios are discussed that provide insights regarding the characteristic features of TOC, such as goals and necessary conditions, performance measures, five-focusing steps for continuous improvement, and drumbuffer-rope scheduling. These insights will assist managers in making important decisions regarding approaches to successful TOC implementations, and will provide academics with a broad range of future research opportunities.     

Cyclic optimisation for localisation in freeform surface inspection

Over the last decade, several papers have been published on the contribution of the Theory of Constraints (TOC) and Throughput Accounting to product mix decisions. In these papers, the quality of the TOC-based approach to generate good or even optimal solutions is assessed with very different results, particularly when compared with other product mix decision tools. It is the objective of this article to expose the often, not explicitly revealed, premises that are mandatory for generating optimal product mix decisions using a TOC-based approach. By analysing several examples, which are modifications of one and the same basic example, premises on the solution space and the objective function are examined and conclusions made, which are shown in a checklist at the end of the paper. If more than one binding multi-product constraint occurs, optimal product mix decisions can only be derived in special cases. Also, the assumption of integer solutions may lead to non-optimal TOC-based solutions. Furthermore, a correct cost allocation is mandatory, even if sometimes not easily conductible. In particular, taking into account only (all) material costs may lead to wrong decisions. Finally, it is shown that discontinuities in the objective functions deriving, for example, from price discounts, may affect the product mix decision and require special modifications of the TOC-based approach.     

基于瓶颈的产品组合决策方法

产品组合决策关系到企业的效益。本文基于约束理论中确定瓶颈的方法,针对约束理论关于产品组合决策方法在多约束条件下的不可行性,提出了一种同时考虑主约束和最大次约束的启发式方法,分为五个步骤对其进行了详细的说明。文中还通过实例仿真来说明了新方法的可行性及简单、易用性。     

Selecting quality improvement projects and product mix together in manufacturing: an improvement of a theory of constraints-based approach by incorporating quality loss

The quality of products and processes needs to be improved continuously in today's competitive environments. Unless these improvement efforts are focused properly, companies might not achieve desirable results in terms of sales, quality and productivity. Many quality improvement (QI) approaches have a limited evaluation of the factors involved in the selection of QI projects. Theory of constraints (TOC) has been proposed by some researchers as a remedy for the better selection of QI projects. However, these TOC-based approaches do not accurately capture ongoing product quality and its long-term effects on sales. Quality loss, on the other hand, can be used as a measure of customer dissatisfaction, which in turn determines the sales. The paper proposes an improvement of a TOC-based algorithm by incorporating quality loss with it. Using two simple manufacturing examples, it demonstrates that quality losses affect the product mix and QI project selection. It provides suggestions for further research directions for wider applicability of the proposed approach.     

Optimizing the multiple constrained resources product mix problem using genetic algorithms

The theory of constraints (TOC) is s a management philosophy for maximizing throughput. Since its introduction, many have criticized it as being inefficient when multiple constrained resources exist. The application of the five steps contained in TOC have been criticized by some researchers on the grounds that the application of five steps of TOC to the product mix decision leads to implicit or unrealizable solutions when multiple resource constraints in a plant exist. This paper views TOC as a management philosophy and a genetic algorithm-based TOC procedure is presented for solving combinatorial problems encountered in practice which cannot be solved using linear-integer programming or similar techniques. For smaller size problems, the results of the proposed procedure are compared with results of optimal methods published in the literature. The results are encouraging and therefore support the use of the proposed approach in an industrial setting.     

Framework for understanding the complementary nature of TOC frames: Insights from the product mix dilemma

The product mix problem has been discussed in the management science literature for decades, and in the Theory of Constraints literature since Goldratt and Fox first presented the simple P's and Q's problem. This paper will investigate the product mix dilemma using a variety of Theory of Constraints approaches that complement and extend traditional treatments such as Linear Programming, spreadsheet and graphical approaches. The Theory of Constraints approaches include the product mix algorithm, Five Focusing Steps, Evaporating Cloud and reality trees. The paper will surface insights and draw out implications arising from the use of these different approaches, and show how the choice of Theory of Constraints tool reflects, in essence, a 'framing' of the problem in different ways. We will show that while the various Theory of Constraints frames are complementary in nature, they may also substitute for each other to some degree. As a result, we suggest there is in most cases no one best frame, and the often implicit search for a 'best-fit' frame should be recognized and abandoned. It is argued that there should be greater acceptance of the use of a variety of frames that help us deal with the inevitable ambiguities and complexities that arise in the real world. The application of the framing approach is demonstrated using a real case. Finally, the authors' experience with the use of this and other examples of framing in teaching will be briefly discussed.     

Unidirectional AGV guidepath network design: A heuristic algorithm

In an AGV system, the design of the guidepath network is one of the most important factors that determine the system effectiveness. Many alternative guidepath design schema have been introduced in the literature. However, the methods generally address the static production environment where the product mix or machine routings are assumed to be stable over time. In today's dynamic production systems with small lot sizes and short product cycles, the assumption of an unchanging product mix over an extended period is unrealistic in some situations. Considerable system inefficiency can be introduced when a network designed under the assumption of a stable product mix is used when indeed the product mix has changed. To avoid this kind of hidden source of inefficiency, what is needed is a tool that can recognize when a previously designed network is inappropriate for a new product mix and then uses the new product routing information to reconfigure a new network appropriate for the new production condition. A heuristic algorithm for the design of AGV guidepath in an environment with changing product mix is proposed here. The algorithm not only makes it possible to adapt the system network as the product mix changes, but also produces new designs at reasonably short computational time.     

Theory of constraints and linear programming: a comparison

Since initial work done by Goldratt in the mid-1980s and as the concepts related to the theory of constraints (TOC) have become more developed, many have concluded that the TOC approach offers nothing in addition to what can be accomplished through linear programming (LP). Through the use of an example, this paper compares TOC to LP and clarifies the differences surrounding the TOC philosophy and the LP technique     

Assembly planning with an ordering genetic algorithm

The goal of assembly planning consists in generating feasible sequences to assemble a product and selecting an efficient assembly sequence from amongst these. This paper describes an original ordering genetic algorithm (OGA) developed to solve this problem. The developed approach permits the generation of assembly trees for a mechanical product. The algorithm is based on three main ideas. First, a mapping transforms any studied assembly plan into a valid one using 'precedence values' changing through the sequence, so that an invalid sequence will never be proposed. Secondly, to identify subsets, trace is kept all along the sequence of the components membership to a set of parts. Finally, the individuals of the OGA are compared with each other using a multi-criteria decision aided method called PROMETHEE II. The use of this method avoids aggregating several technical criteria into a unique fitness value. The proposed algorithm, illustrated through the simple example of a mouse device, has been applied on an industrial signalling relay made of 34 parts.     

基于协同过滤的大规模定制个性化推荐方法

为了对用户的产品定制决策进行动态辅助和引导,让用户在大规模定制系统中的产品配置过程变得更简单、方便,在产品定制环节引入了协同过滤推荐算法的思想,并结合大规模定制的特点对推荐算法进行适当改进后提出了一种新的面向大规模定制的个性化推荐算法。给出一个手机定制的实例,对算法产生推荐结果的具体过程进行了模拟与分析。仿真实验产生的可行推荐方案表明,该算法对于解决大规模定制模式下的个性化推荐问题是可行有效的。     

面向客户的产品规模定制设计及其应用

讨论产品设计过程中的基本模块，并提出规模定制设计的概念。面向客户的产品设计的核心是开发面向规模定制的产品族结构，实现与统一的产品开发和送货机制的后期集成。介绍了产品族结构的特点，以及明确产品族结构的步骤。同时给出了一个规模定制设计的研究实例。     

An evolutionary approach for product line adaptation

A product line is rarely designed from scratch but usually evolves with products phased in and phased out. The adaptations of a product line and the associated changes in product mix and prices have complicated implications upon both market demand and manufacturing cost. This paper proposes an evolutionary approach to support coordinated decision-making among marketing, engineering and manufacturing regarding product line adaptation. A mixed logit discrete choice model and an activity based costing model are developed to quantify the demand and cost implications of product line adaptation, respectively. Product line adaptation is then formulated as a mixed integer non-linear programming problem. Given the coexistence of both discrete and continuous variables, a bi-level optimisation procedure combining genetic algorithm and differential evolution is developed for problem-solving. The proposed methodology is illustrated with an example of netbook product line design.     

基于TOC和TRIZ理论的儿童下肢康复训练仪改良设计研究

目的 解决现有儿童下肢康复训练仪难以适配儿童患者、功能性单一及缺乏互动的问题,使儿童下肢康复训练达到更好的训练效果。方法 提出一种基于约束理论（TOC）和发明问题解决理论（TRIZ）的儿童下肢康复训练仪改良设计方法,将TOC理论中的当前现实树和冲突解决图表与TRIZ理论中的冲突解决工具相结合,提出一种针对交互体验问题的基于TOC和TRIZ理论的改良设计方法与流程,并将其应用于儿童下肢康复训练仪的改良设计中。结论 基于TOC和TRIZ理论的儿童下肢康复训练仪改良设计能够显著提高患儿的训练效果,其改良点包括造型和交互体验两个方面。利用VR技术与故事（游戏）情节结合的方式,提高了患儿的训练兴趣和参与意愿;将患儿信息上传至云端数据库,可生成不同的故事（游戏）情节,匹配不同的训练阶段,提高了训练的针对性;智能控件降低了训练过程中对康复师的依懒性,从产品角度缓解了康复师供不应求的现状。     

An improved algorithm for solving equations for intra-band tunneling current in heterojunction solar cells

Numerical simulation of heterojunction solar cells is a useful approach to investigate the inner physics of the devices, and can be implemented by simulation tools today. An intra-band tunneling model has been described previously and applied to some heterojunction devices. However, the commonly used algorithm for solving the associated equations, the Gummel method, fails in the case of high recombination rates such as in amorphous Si solar cells. In this work, we present an improved algorithm that enhances convermicmgence when solving for intra-band tunneling in heterojunction solar cells with high defect state densities. The algorithm uses a sequential application of the Gummel and Newton methods tailored for optimal stability and convergence. As an example, simulation results for a heterojunction with intrinsic thin layer solar cell are discussed. The improved algorithm and related simulations in this work are implemented through a program we have developed, wxAMPS.     

用户集群驱动的适老化产品优化策略研究

目的 为解决用户诉求表述模糊所导致的需求获取不足、设计决策复杂难定等问题对适老化产品及应用服务推广的阻碍,最大程度提升用户体验满意度,提出一种用户集群评价驱动的短周期、高效率的感性设计方法。方法 引入评价信息对目标用户集群的需求进行精准而真实地深层挖掘,结合TextRank算法量化感性描述文本,构建评价指标体系;充分借助犹豫模糊决策算法构建方案评价模型(I-HFCJM),计算各产品评价指标得分函数值对样本产品优劣予以可视化分析,以现有产品的可借鉴点与注意问题为依据指导产品改进设计。结论 以老年家用制氧机改进设计为例,计算改进方案同样本产品的综合权重并进行优劣排序,验证了该方法的有效性,为老年家用制氧机及其他适老化产品设计开发提供了改进策略。     

Discussion: Theory of constraints and linear programming: A re-examination

In a previous paper in this journal, Luebbe and Finch (1992) compared the theory of constraints (TOC) and linear programming (LP) using the five-step improvement process in TOC. This paper clarifies some of their results. Specifically, by modifying their example we show that some of their conclusions are not generalizable. Further it is shown that using LP is preferable to the $ return/constraint unit method in helping increase throughput and the reasons for this are discussed. Thus LP is a useful tool in the TOC analysis.     

A block iterative algorithm for tomographic reconstruction of ionospheric electron density

Ionospheric tomography is a technique whereby a vertical cross section through ionospheric electron density can be imaged. The vertical resolution of ionospheric tomography systems is inherently poor, but can be improved by using a priori information in the tomographic reconstruction algorithm. Care must be exercised in using a priori information, since if too much of it is used, the reconstruction algorithm may discard some of the information contained in the tomographic data in favor of satisfying some of the a priori assumptions. Orthogonal decomposition (OD) is an existing technique that uses a priori information to constrain the reconstruction to lie in a space of reasonable images without weighting the reconstruction toward any particular a priori image. In this way a priori information can be used in a manner that does not overwhelm information contained in the data. Gauss-Seidel (GS) is an iterative algorithm that is used to calculate solutions for large systems of linear equations. In this article, a block version of the GS algorithm will be used to calculate the solution of the least-squares problem that is created using OD. The complete algorithm presented here will be called the residual correction method (RCM), since it involves calculation of successively better approximations based on the residual error. RCM is a fast and numerically stable algorithm that extracts as much information from the data as possible. A numerical example demonstrating the properties of RCM will also be presented. © 1996 John Wiley & Sons, Inc.