Showing papers in "International Journal of Production Research in 1988"
TL;DR: In this article, the authors systematically study the performance behavior of beam search with other heuristic methods for scheduling, and the effects of using different evaluation functions to guide the search, and develop a new variation of beam searching, called filtered beam search, which is computationally simple yet produces high quality solutions.
Abstract: Beam search is a technique for searching decision trees, particularly where the solution space is vast. The technique involves systematically developing a small number of solutions in parallel so as to attempt to maximize the probability of finding a good solution with minimal search effort. In this paper, we systematically study the performance behaviour of beam search with other heuristic methods for scheduling, and the effects of using different evaluation functions to guide the search. We also develop a new variation of beam search, called filtered beam search which is computationally simple yet produces high quality solutions.
354 citations
TL;DR: In this article, a two-stage procedure for the design of a cellular manufacturing system is proposed, where the first stage forms the part families and the second stage form the machine cells.
Abstract: A two-stage procedure for the design of a cellular manufacturing system is proposed. The first stage forms the part families. The use of clustering techniques with a new proximity measure is advocated for this stage. The proximity measure uses the manufacturing operations and the operations' sequences. The second stage forms the machine cells. An integer programming model is proposed for this stage. The solution of this model will specify the type and the number of machines in each cell and the assignment of the part families to the cells. The relevance of this approach in the design of flexible manufacturing systems is discussed.
273 citations
TL;DR: This article describes the evolutionary process from basically a computerized Kanban to an attempted computerization of the Drum-Buffer-Rope technique, and the reasoning that brought this evolution so quickly to its ultimate conclusion.
Abstract: Computerized shop floor scheduling is probably one of the most glaring examples of an area in which the tremendous efforts and investments of a large number of companies have produced at best only a partial solution. One of the more successful attempts is known as OPTR (optimized production technology) on which numerous papers have been published highlighting the various aspects of this package. What was realized by only a small group of people, namely the user base, was that this package was evolving at an incredible rate from one year to the next. This article describes that evolutionary process from basically a computerized Kanban to an attempted computerization of the Drum-Buffer-Rope technique. The major emphasis will not be so much on the technical developments, but more on the reasoning—gained from the experience of implementing this package—that brought this evolution so quickly to its ultimate conclusion. Namely, the real key lies mainly in the conceptual framework under which we run our organiza...
262 citations
TL;DR: Load-oriented manufacturing control is a new solution for job shops and its successful implementation in a plastic leaves factory as mentioned in this paper, where the idea is to limit and balance work-in-process inventory on a level as low as possible in order to accomplish a high workcentre utilization as well as a rapid and in-time flow of orders.
Abstract: The paper describes the principles of load-oriented manufacturing control as a new solution for job shops and its successful implementation in a plastic leaves factory. The idea of load-oriented manufacturing control is to limit and balance work-in-process inventory on a level as low as possible in order to accomplish a high workcentre utilization as well as a rapid and in-time flow of orders. As the new system keeps actual lead times on a planned level in a self-regulating way it allows reliable due date scheduling. Furthermore, it points out the bottlenecks and performs applicable mid- and short-term capacity planning.
231 citations
TL;DR: Petri nets are used as a design representation of an FMS for the purpose of comparing different systems on a flexibility basis thus unveiling their desirable flexibility properties.
Abstract: A discussion on the multi-dimensional character of flexibility is presented and an approach is suggested to define and measure flexibility. Petri nets are used as a design representation of an FMS for the purpose of comparing different systems on a flexibility basis thus unveiling their desirable flexibility properties.
151 citations
TL;DR: A survey of early implementators of the just-in-time (JIT) philosophy in the United States with particular emphasis on the identification of implementation problems and operating problems was conducted by as discussed by the authors.
Abstract: This paper reports the results of a survey of early implementators of the just-in-time (JIT) philosophy in the United States with particular emphasis on the identification of implementation problems and operating problems. There were 39 respondents to the survey who were in the process of implementing JIT. Five classes of implementation problems and eight classes of operating problems were identified.
139 citations
TL;DR: In this article, a study of the collection and use of quality-related costs in manufacturing industry was carried out and the authors found in the literature many notional models purporting to indicate the relationships between the major categories of quality costs and a few sets of real data.
Abstract: While carrying out a study of the collection and use of quality-related costs in manufacturing industry the authors found in the literature many notional models purporting to indicate the relationships between the major categories of quality costs and a few sets of real data. Despite being based on common principles, there are wide differences between some of the models and between the models and real data. The paper categorizes and discusses the models in the light of the research experience. It is concluded that many of the models are inaccurate and misleading and serious doubts are cast on the concept of an optimum quality level corresponding to a minimum point on the total quality-cost curve.
139 citations
TL;DR: In this article, a composite flow control system is used to achieve just-in-time operation relevant to the overall business time-constants, supported by matching supply processes and levelled output scheduling.
Abstract: Competitive modern manufacturing businesses must operate total materials flow control systems to ensure low stock levels and short lead times. When a variety of products is produced, to achieve just-in-time operation relevant to the overall business time-constants, composite control systems are needed within the factory, supported by matching supply processes and levelled output scheduling.
119 citations
TL;DR: In this paper, surface finish data were generated for aluminium alloy 390, ductile cast iron, medium carbon leaded steel 10L45, and medium carbon alloy steel 4130 for a wide range of machining conditions defined by cutting speed, feed and tool nose radius.
Abstract: Surface finish data were generated for aluminium alloy 390, ductile cast iron, medium carbon leaded steel 10L45, medium carbon alloy steel 4130, and inconel 718 for a wide range of machining conditions defined by cutting speed, feed and tool nose radius. These data were used to develop surface finish prediction models, as a function of cutting speed, feed, and tool nose radius, for each individual metal. A general purpose surface finish prediction model is also proposed for ductile cast iron, medium carbon leaded steel, and alloy steel. Statistical analysis of experimental data indicated that surface finish is strongly influenced by the type of metal, speed and feed of cut, and tool nose radius. While the effects of feed and tool nose radius on surface finish were generally consistent for all materials, the effect of cutting speed was not. The surface finish improved with speed for ductile cast iron, medium carbon leaded steel, medium carbon alloy steel, and aluminium alloy, but it deteriorated with speed...
109 citations
TL;DR: A job-shop scheduling software currently under development is described, based on Artificial Intelligence programming techniques, which achieves the management of time, empirical knowledge about priority rules and their influence on production objectives and practical knowledge about technological constraints to be satisfied in a given application.
Abstract: A job-shop scheduling software currently under development is described, based on Artificial Intelligence programming techniques. The idea is to be able to make three kinds of knowledge cooperate in the derivation of a feasible schedule : theoretical knowledge (issued form scheduling theory) which achieves the management of time ; empirical knowledge about priority rules and their influence on production objectives; and practical knowledge (provided by shop-floor managers) about technological constraints to be satisfied in a given application. The latter is usually not considered in pure Operations Research algorithms. The system is actually implemented in COMMON LISP and runs on a Texas Explorer LISP Machine and a SUN workstation. Computational results are reported.
108 citations
TL;DR: In this paper, the problem of determining the economic levels for the base stock and lead times for production and transportation in integrated production, inventory and distribution systems (IPIDS) is addressed.
Abstract: This paper deals with the problem of how to determine, with high reliability, economic levels for the base stock and lead times for production and transportation in integrated production, inventory and distribution systems (IPIDS). This system model has a ‘pull type’ ordering system and is of the three stage tandem type, consisting of manufacturer, wholesaler and retailer. A method is described for fixing ‘base stock’ levels and lead times, when product production and transportation stop, which prevents out of stock and minimizes dead stock at each stock point in the case where product models are changing in the market.
TL;DR: In this article, the authors present the proposition that US firms must recognize the potential of purchasing activities as "quality and productivity centres" through successful implementation of just-in-time (JIT) purchasing, and identify the major activities of JIT purchasing that have favorable effects on product quality and productivity.
Abstract: Recently, Just-In-Time (JIT) purchasing practices have been recognized as a cornerstone of Japan's success in improving product quality and productivity. Traditionally, however, US firms have treated purchasing functions as a ‘profit centre’. This paper presents the proposition that US firms must recognize the potential of purchasing activities as ‘quality and productivity centres’ through successful implementation of JIT purchasing. Personal interviews have been conducted at four US manufacturing firms in an effort, first, to identify the major activities of JIT purchasing that have favourable effects on product quality and productivity and, second, to describe how these activities can improve product quality and productivity in US firms.
TL;DR: In this article, the authors investigated the effect of several key factors related to the automated guided vehicles on the overall performance of a flexible manufacturing system, such as the number of pallets allowed in the system, number of vehicles used and the carrying capacity of each and the input and output queue capacities of the machining stations; finally bidirectional traffic is allowed in some routes.
Abstract: Although the technology exists for more advanced applications of automated guided vehicles in flexible manufacturing systems, the current employment of these vehicles in material handling generally subscribes to a simple mode of operation : single-load-carrying capacity for each vehicle and unidirectional traffic on each route of the system. Through a simulation programme, this study investigates the effect of several key factors related to the automated guided vehicles on the overall performance of a flexible manufacturing system. These are the number of pallets allowed in the system, the number of vehicles used and the carrying capacity of each and the input and output queue capacities of the machining stations; finally bidirectional traffic is allowed in some routes. The results show that there is a strong interaction among these factors and reveal their combined effects on the throughput from a small flexible manufacturing system. Upon the user's request, the simulation program also provides ...
TL;DR: In this article, a three-stage procedure for configuring machines into manufacturing cells and assigning the cells to process specific sets of jobs is presented, with the objective of minimizing the deviation between available capacity and the workload assigned to each machine.
Abstract: We present a three-stage procedure for configuring machines into manufacturing cells, and assigning the cells to process specific sets of jobs. First, operations are assigned, with the objective of minimizing the deviation between available capacity and the workload assigned to each machine. We then extended King's algorithm to group machines based on similarities of operations. Finally a direct-search algorithm for defining the composition of manufacturing ceils is offered. A comprehensive example is also included.
TL;DR: A simple modification of Palmer's heuristic for scheduling jobs in a flow shop compares very well with that of the more sophisticated and better algorithm of Campbell, et al. (1970) at a fraction of the effort required by the latter.
Abstract: We describe a simple modification of Palmer's heuristic for scheduling jobs in a flow shop. While the additional computation required is relatively small, the performance of the algorithm compares very well with that of the more sophisticated and better algorithm of Campbell, et al. (1970) at a fraction of the effort required by the latter.
TL;DR: A mathematical model, incorporating a matrix-tree-chain method for generating a tolerance chart and the method of tracing has been established based on the tolerance chart technique and an interactive computer aided dimensional planning system CADP is presented.
Abstract: A mathematical model, incorporating a matrix-tree-chain method for generating a tolerance chart and the method of tracing has been established based on the tolerance chart technique. A mathematical method for performing calculations of working dimensions and specification of tolerances in manufacturing processes is proposed. An interactive computer aided dimensional planning system CADP is presented.
TL;DR: Heuristic algorithms for batching a set of orders such that the total distance travelled by the order picking machine is minimized are presented and their efficiency and validity are illustrated through computer simulation.
Abstract: This paper deals with an order picking problem in an automated storage and retrieval system (AS/RS). We present heuristic algorithms for batching a set of orders such that the total distance travelled by the order picking machine is minimized. These algorithms are based on cluster analysis and their efficiency and validity are illustrated through computer simulation. The results show that the algorithms developed perform substantially better than those from previous studies.
TL;DR: In this paper, the authors developed approximations for the distribution of the total time spent in a dynamic job shop using an exponential limit theorem and an heuristic decomposition of open queueing networks.
Abstract: In this paper we develop approximations for the distribution of the total time spent in a dynamic job shop. In particular, using an exponential limit theorem and an heuristic decomposition of open queueing networks, we show that for a large class of dynamic job shops, the total time spent in a dynamic job shop can be approximated by an exponential random variable with an appropriate mean. Approximations for job shops that do not belong to this class are also developed. Numerical results show that the proposed approximations are in general very good. Application of these approximations in the assignment of due-dates is also illustrated.
TL;DR: In this paper, the authors show how the Kanban method may be adapted to the job shop and show that the adaptation is extensively tested through simulation and then implemented, and shows marked improvement over previous practice.
Abstract: Japanese industrial management techniques have been applied in a large number of large Western enterprises. In particular, the Kanban method has been used to control materials, production rate and volume, and to adjust production to requirements. The authors show how the Kanban method may be adapted to the job shop. This adaptation was extensively tested through simulation and then implemented. Actual performance is consistent with the simulation results, and shows marked improvement over previous practice.
TL;DR: In this paper, a method of estimating the amount of safety stock needed in each station of a production line due to variation in processing times, machine breakdowns and demand fluctuation in order to meet a predetermined desired level of performance is proposed.
Abstract: This report proposes a method of estimating the amount of safety stock needed in each station of a production line due to variation in processing times, machine breakdowns and demand fluctuation in order to meet a predetermined desired level of performance. The production line is assumed to operate as a pull system and the measure of performance is the average percentage of demand backlogged. Multiple machines and different batch sizes in the stations are included in the model. Dynamic production control is used and is based on the current inventory level in every station of the system. Simulation results are used to test the performance of the system in which the maximum inventory level allowed in each station is based on the estimation given by our method.
TL;DR: In this paper, the authors describe the application of the artificial intelligence techniques of blackboard and actor based systems for intelligent cell control in a framework termed Production Logistics and Timings Organizer (PLATO-Z).
Abstract: Cell control forms one level of a hierarchical approach to the control of automated manufacturing systems. This paper describes the application of the artificial intelligence techniques of blackboard and actor based systems for intelligent cell control in a framework termed Production Logistics and Timings Organizer (PLATO-Z). The blackboards required are described and the implementation is detailed. The implications of some practical considerations are also described.
TL;DR: In this paper, the authors describe the development of an intelligent controller for a fleet of free-ranging automated guided vehicles, which is capable of detecting and preventing collisions before they occur.
Abstract: This paper describes the development of an intelligent controller for a fleet of free-ranging automated guided vehicles. The vehicle controller will operate in real-time. Once a transportation request is received, the location where the demand is generated and the nature of the demand is sent to the vehicle controller. Based on this information and the current status of the system, the vehicle controller will perform dispatching, routing, and scheduling tasks for the free-ranging vehicles. For the selected vehicle, the vehicle controller finds a route and then it will schedule the vehicle along that route in a non-conflicting manner. The vehicle controller is capable of detecting and preventing collisions before they occur. Once a collision is detected, the vehicle controller will explore different solutions to prevent the collision. The alternative which will result in the least delay time is the solution which will be selected by the vehicle controller. The inputs to the system are the layout t...
TL;DR: Two basic components of the knowledge based system, namely the expert system and heuristic clustering algorithm are discussed, which considers alternative process plans and multiple machines for solving the generalized group technology problem.
Abstract: In this paper a knowledge based system (EXGT-S) for solving the generalized group technology problem is presented. The formulation of the group technology problem involves constraints related to machine capacity, material handling system capabilities, machine cell dimensions and technological requirements. Il has been developed for an automated manufacturing system. EXGT-S is based on the tandem system architecture presented in Kusiak (1987). It considers alternative process plans and multiple machines. EXGT-S takes advantage of the developments in expert systems and optimization. Two basic components of the knowledge based system, namely the expert system and heuristic clustering algorithm are discussed. Each partial solution generated by the clustering algorithm is evaluated by the expert system which modifies search directions of the algorithm.
TL;DR: An integrated hierarchical framework of a process planning system with a CAD interface is presented to integrate design with process planning using AI techniques and sample results from the CAD interface are presented.
Abstract: Artificial intelligence (AI) techniques provide various tools for use in the development of automated process planning systems. AI can be utilized for automated reasoning about the shape, features and relationship between features and for development of expert systems for creating the process plan itself. Most of the previous work on AI in process planning deals with one specific application. This paper presents an integrated hierarchical framework of a process planning system with a CAD interface. The objective of the project discussed in the paper is to integrate design with process planning using AI techniques. The development of a CAD interface is discussed with respect to automated feature recognition, determination of tool approach direction and deciding the precedence relationship between the features. Sample results from the CAD interface are presented. The expert system for the process planning module is discussed with the part representation and knowledge base and the plan generation procedure. ...
TL;DR: In this article, an economic model to determine the most profitable specifications for the complete inspection plan with the considerations of economic loss caused by quality deviations, rework cost, and inspection cost is presented.
Abstract: In a complete inspection plan, all outgoing items are subject to inspection, and the items failing to conform to the predetermined specifications are reworked so that their quality characteristics are exactly equal to the expected value (target value). This paper presents an economic model to determine the most profitable specifications for the complete inspection plan with the considerations of (a) economic loss caused by quality deviations, (b) rework cost, and (c) inspection cost. The solution to the cases in which the inspection cost is a constant provides important insights of the model. A detailed solution procedure is also given to a model in which the inspection cost is a linear function of the width of the specifications, the quality characteristic is normally distributed, and other loss and cost functions are assumed of a quadratic or linear form.
TL;DR: In this paper, the authors discuss how and why cyclic schedules exist and why they can be used to synchronize manufacturing tasks sequentially within a short lead time and by creating cycle times by which many different kinds of work can be fitted together like clockwork.
Abstract: Enormous improvements are possible by synchronizing manufacturing tasks sequentially within a short lead time and by creating cycle times by which many different kinds of work can be fitted together like clockwork. Actually achieving close synchronization requires utmost effort to improve every aspect of a manufacturing organization. A concept central to the improvement is the creation and use of cyclic schedules. The paper discusses how and why these relationships exist.
TL;DR: The relative advantages of top-down versus bottom-up forecasting strategies are outlined and the situations in which each should be preferred.
Abstract: Some controversy exists about the advocacy of top-down versus bottom-up forecasting strategies. Top-down forecasting refers to the process of forecasting the demand for the aggregate of items in a class and then inferring individual demands according to a percentage of the total; bottom-up refers to separately forecasting the requirements for each individual item. This paper outlines the relative advantages of each strategy and indicates the situations in which each should be preferred.
TL;DR: In this paper, the authors tried to see the effects of this imbalance of stage operation times in a just-in-time (JIT) production system, where the stage operation limes were random variables and the mean of each operation time was assumed deterministic in sampling from a group of different mean operation times.
Abstract: Ideally a production line is perfectly balanced and most efficient when all the stages have been allocated an equal amount of task times, but realistically this situation is almost impossible for a practical problem. For such an ideal situation, the just-in-time (pull system) production system holds no inventory in between the stages, such that the production process is just-in-time. The stage operation limes are random variables and the mean of each operation time has been assumed deterministic in sampling from a group of different mean operation times. Thus, because of the differences in stage operation times due to the variability of the operator's performance and/or unequal distribution of task times, the production line experiences various types of problems. In this paper we tried to see the effects of this imbalance of stage operation times in a just-in-time (JIT) production system. In addition to the implications of the simulation results, an indication for further research has been given in the co...
TL;DR: In this paper, the authors derived two formulae to calculate the average inventory yielded by fixed interval withdrawal Kanban and supplier Kanban, respectively, in a just-in-time production system.
Abstract: This paper derives two formulae to calculate the average inventory yielded by fixed interval withdrawal Kanban and supplier Kanban, respectively, in a just-in-time production system. By using the formulae, another two formulae to determine the minimum number of Kanbans to be required for the fixed interval withdrawal Kanban and the supplier Kanban are also shown. An algorithm to obtain the optimal order interval is proposed to minimize the total operation cost composed of the inventory cost and the withdrawal cost for the Kanban system
TL;DR: In this paper, the authors proposed rack-class-based storage assignment procedure and class selection procedure to find the minimum number of S/R machines required and identify the number of aisles each R machine serves.
Abstract: Multi-aisle S/R machine system (MASS) can substantially reduce high initial investment cost which is a major reason for the low popularity of AS/RS in manufacturing companies. The objective of this study is mainly related to the design aspects of MASS. With a travel time model developed, average travel time of S/R machine is determined. We propose rack-class-based storage assignment procedure and class selection procedure to find the minimum number of S/R machines required and identify the number of aisles each S/R machine serves. Example problems are solved to illustrate the procedures. The results show that MASS is effective in reducing initial installation cost, provided that the pallet demands are relatively low.