Showing papers on "Job shop published in 1981"

Due Date Assignment for Production Systems

Abstract: This paper is concerned with the study of the constant due-date assignment policy in a dynamic job shop. Assuming that production times are randomly distributed, each job has a penalty cost that is some non-linear function of its due-date and its actual completion time. The due date is found by adding a constant to the time the job arrives to the shop. This constant time allowed in the shop is the lead time that a customer might expect between time of placing the order and time of delivery. The objective is to minimize the expected aggregate cost per job subject to restrictive assumptions on the priority discipline and the penalty functions. This aggregate cost includes 1 a cost that increases with increasing lead times, 2 a cost for jobs that are delivered after the due dates: the cost is proportional to tardiness and 3 a cost proportional to earliness for jobs that are completed prior to the due dates. We present an algorithm for solving this problem and show that the optimal lead time is a unique minimum point of strictly convex functions. The algorithm utilizes analytical procedures; computations can be made manually. No specific distributions are assumed; the distribution of total time a job is in the shop is utilized by the algorithm. This distribution can be theoretical or empirical. An example of a production system is presented.

Analytical Evaluation of Hierarchical Planning Systems

Abstract: Hierarchical planning systems have become popular for multilevel decision problems. After reviewing the concept of hierarchical planning and citing some examples, we describe a method for analytic evaluation of a hierarchical planning system. We show that multilevel decision problems can be nicely modeled as multistage stochastic programs. Then any hierarchical planning system can be measured against the yardstick of optimality in this stochastic program. We demonstrate this approach on a hierarchical system that can be shown to be asymptotically optimal for a job shop design/scheduling problem.

Combined scheduling system for reducing job tardiness in a job shop

Abstract: Two formulae to give the mean and the standard deviation of job flow-times are derived from the machine number, the load ratio, the required number of operations, and the mean and the standard deviation of processing times, which specify the characteristics of a job shop. On the basis of the two formulae, a method of due-date assignment which contains a due-date adjustment factor is proposed. The assignment method is combined with the sequencing procedure to construct the total scheduling system for reducing job tardiness. Two new dispatching rules are presented as the second phase of the system. The experimental comparison shows that the efficiency of the proposed systems can be better than the conventional scheduling systems

Job shop control

Abstract: The paper defines batch or job shop production and reviews the functions of production planning and production control in a job shop manufacturing situation. It argues that the separation of planning and control has resulted in the artificial isolation of the sequencing problem in job shop research. It attempts to redefine the production control function for a job shop, now called job shop control, and discuss the activities it involves. The major decision-making problems associated with job shop control are highlighted and an objective function of costs to aid in management decision making is evolved.

Minimizing maximum lateness in a two-machine unit-time job shop

Abstract: AnO (nlogn)-algorithm is given for the two-machine, job-shop scheduling problem withn unit-time tasks in which maximum lateness is to be minimized. This algorithm generalizes recent results by Hefetz and Adiri for the corresponding makespan problem.

The Simulation Data Language (Sdl™): Applications and Examples

Abstract: The Simulation Data Language (SDL) is a database management system specifically designed for use in simulation. A typical example for the use of SDL is the evaluation of the operations of a job shop. In this example, SDL helps the analyst organize and use data gathered from the shop itself, analyze results independently of model runs, and compare alternative policies. Current applications of SDL include studies of health-care delivery and availability, scheduling of aircraft maintenance, manufacturing of discrete parts, steel production, large-scale socio- economic systems, and river systems. In these applications, SDL helped analysts produce statistical analyses, summaries, and reports of data independent ly of model runs, use data gathered from the real system in their models, decompose large, complex models into component models, provide data collection, post-run data analysis, and report generation capabilities that are transparent to the ultimate user of a special-purpose simulation language.

A note on the two machine job shop with exponential processing times

Abstract: Consider two machines, labeled 1 and 2. A set of tasks has to be processed first on machine 1 and after that on machine 2. A second set of tasks has to be processed first on machine 2 and after that on machine 1. All the processing times are exponentially distributed. We present a policy which minimizes the expected completion time of all tasks.

Job-shop layout analysis via computer simulation

Abstract: In this paper, the applications of computer simulation and statistically designed experiments are presented as a means to diagnostically evaluate and design more efficient, layouts for manufacturing job-shops. The simulation model is defined and driven based entirely on actual operating data, shop procedures and the physical configuration of the shop. Utilizing the flexibility and realism of a simulation model, layout analysis is conducted to evaluate the effects of changes in material handling methods, in-process storage locations and various operating procedures on shop floor congestion and job throughput. An experimental design is used to allow statistical comparison of the results in the light of the variability of the entire shop system and a predictive model for mean job flow time is developed using multiple regression analysis. The results show than an improved layout and better operating procedures can eliminate the need for additional capital investment and work force expenses, yet still...

SPT, Data Analysis, and a Bit of Common Sense in Bus Maintenance Operations: A Case Study

Abstract: A fundamental result in the theory of scheduling is the optimality of the shortest processing time (SPT) rule for minimizing mean flow time in the basic single-machine scheduling problem. The relationship between flow time and inventory also makes this rule an appropriate one to use if the objective is one of rapid turnaround, even when the scheduling environment is dynamic (i.e., where jobs arrive over time). This paper presents a case study where the principle behind SPT sequencing was used for a job shop bus maintenance operation. The results obtained by this implementation are further evidence that the application of SPT sequencing can yield significant improvements in scenarios considerably more complex than the basic single-machine scheduling problem.

A model to examine the effect of loading conditions on scheduling rules in a job shop environment

Abstract: Although a great deal of research has been carried out in the field of job scheduling this has generally been directed towards examining the benefits of particular rules and presenting improved algorithms. This paper examines how real job shop problems can be modelled and available scheduling rules examined for particular capacity loading conditions. A model of a medium-size production job shop is developed and it is shown that, for their particular shop layout and job mix, the performance and ranking of particular rules with respect to certain criteria, change with shop conditions. The model developed can easily be applied to a wide range of job shop situations and once performance charts have been produced for those scheduling rules available, they can be used to aid the existing scheduling system whether manual or computer based.

The Use of a Digital Simulation Model as a Management Tool for Job Shop Control

Abstract: The design and test of a detailed simulation model of the job shop of a machine tool company for use as a management tool is described. The model is written in GASP IV and runs on a CDC7600 computer. The model was produced from an in-depth study of the operation of the job shop during which time a detailed and validated descriptive model was prepared. The model uses actual company data (approximately 1300 components), and individual machines and operators and incorporates double shift working and overtime, machine breakdown and absenteeism, limited batch splitting, alternate job routings, and stores. The procedure for verification and validation of the model is described.

Primal search tree algorithms for the general job shop problem

Abstract: The scheduling of job-shops has been the subject of considerable study, partly because of the many possible applications, but also because of the somewhat tantalising nature of the problems involved. Much of the analytic effort has been devoted to the construction of the shortest possible schedule. In a notation used by Conway, Maxwell and Miller (1967) and others, this problem is referred to as n/m/G/F(,max) with the following interpretation. There are n jobs to be processed on m different machines. Each job consists of an ordered list of operations with the machine required and the time involved already determined for each operation. Each operation, once started, must run to completion. Each job may require the machines in a different order. The objective is to minimize the total schedule time. Of the many approaches to solving this problem, algorithms using branch-and-bound methods have proved to be the most successful. One of the very first applications of this method associated a complete schedule with every node in the search tree. The results of this approach were very poor and were easily surpassed by applications which scheduled only one extra operation at each node in the search tree. All investigations involving branch-and-bound algorithms since have used this latter application. This thesis proposes a different direction of investigation, one that is related to the schedule at every node approach. A partial, possibly complete, schedule is associated with every node in the search tree. In each schedule, a critical operation is identified and from it, an associated critical set of operations. This set generates a number of sub-problems. The optimum schedule will be the existing schedule or the optimum schedule for one of the sub-problems. A detailed study is then made on a number of algorithms using this approach as a basis. The results of the algorithms proved to be very promising and were better than the results of any previous algorithm. The conclusion is reached that the proposed approach to solving the general job-shop problem should be included in future investigations, not at the expense of the approach of scheduling only one operation at each node in the search tree but along with it. Further, some suggestions for future work are given.

Experimental optimization by simulation technique

Abstract: The purpose of this paper is two fold, (i) to propose an integrated framework for studying management systems by simulation and (ii) to evaluate the possibility of using simulation as an experimental optimization technique. A job shop model, which can be used to test both materials handling and dispatching rules, was developed in order to demonstrate the applicability of the proposed procedures.