The scheduling problem is often treated as a combinatorial optimization problem based on the definite processing time. In real production systems, however, the processing time is not always definite. In this paper, assuming that a shop is operated on a schedule sequence predetermined based on estimated processing times, the actual required-time performance of a schedule which is optimal with respect to the estimated processing times, is analyzed under the condition that the actual processing times fluctuate randomly around the estimated values. Extensive experimental results show that 1) finding and following the optimal or near-optimal schedule with respect to estimated times is meaningful enough, to justify the schedule optimization, 2) even though the variability of processing times increases, the relative precision performance of schedules can deteriorate only a littel linearly, whereas the associated precision worsens very quickly.
This paper shows the two-stage lot scheduling algorithm for multi-lot production process. The algorithm consists of the first stage selecting between three scheduling methods and the second stage executing the method selected in the first stage. Then, this algorithm proposes two new control means for decreasing the gap between the evaluation function values of the first 'plan' stage and the second 'execution' stage. One is a feedback control, which uses the information obtained by executing the already planned scheduling method for the lower production process in the plan for the upper production process that follows. The other is a feed-forward control, which uses the estimated information about all upper production processes obtained by applying each of the three scheduling methods to every production process.
In practical scheduling situations, processing times are often unknown exactly. Representing these imprecise processing times by fuzzy numbers can help achieve more realistic and more natural solutions, but the fuzzy arithmetic causes some new calculative difficulties. In this paper, an effective method for solving the scheduling problems with triangular or trapezoidal fuzzy processing times is proposed. It solves a scheduling problem with fuzzy processing times through solving an ordinary scheduling problem with content processing times, which are the generalized mean values of the fuzzy processing times. Tested using numerous examples of parallel machine scheduling problems generated randomly, the proposed method has been proved effective.
This paper presents the method for designing mixed product lines as an expert system. This method consists of the production rules for taking the following design procedures. (1) To select the position of each work out of twelve positions. (2) To determine the kind of robot to be set in each work station and the position to set it. (3) To assign elemental works to the robots. The position of works is changed by spinning two axes. The design goals are to minimize the number of work stations and the spinning frequency to maximize the efficiency of performing desired the production plans.
When different models of trucks or prefabricated houses are manufactured by a mixed-model assembly line, production efficiency usually decreases since the assembly times differ significantly among such different models. The variation of the assembly times can be leveled by installing a bypass line which processes a part of assembly operations of models with relatively longer assembly times. In this paper, a sequencing problem in the mixed-model assembly line with a bypass line is formulated and a heuristic algorithm is developed to obtain a near-optimal solution.
The purpose of this study is to clarify the interrelations among four work commitment forms (i.e., work involvement, career commitment, job involvement, and organizational commitment). The survey sample consisted of 314 lower male managers in a large-scale manufacturing car plant. The factor analysis showed that these four work commitment forms were conceptually distinguished from one another. The correlations among them have provided little evidence for any concept redundancy. These four concepts were differently related to the personal variables, such as age and personality, and other job-related variables, such as role status and job satisfaction. The implications of the analysis results and future issues on the work commitment studies in Japan were discussed.
A dispatching priority problem is proposed which seeks an optimal or near-optimal dispatching priority in an FMS environment where several distinct jobs are produced concurrently and repetitively. Characterization of the problem as a special type of scheduling problems is clarified together with its practicality and importance. Simulation would be required to evaluate the effects of dispatching priorities on scheduling performance, and thus a combined simulation/approximate optimization approach is presented to seek a near-optimal dispatching priority that tries to minimize the total tardiness within a given limitation on computational requirements. Extensive experiments with local search and simulated annealing as means for approximate optimization, reveal appropriate parameter strategies and demonstrate the power of exploring better dispatching priority. The applicability of the proposed problem and the solution strategy is verified by showing the results obtained for a real commercial FMS.