Average Setup Time Research Articles

2104 ニューラルネットワークを用いた納期遅れコストと段取り時間の二目的スケジューリング : SA による学習

This report proposes a scheduling method for a bi-criteria job shop scheduling problem. The criteria considered in this report are tardiness cost and sequence-dependent setup time. Two priority rules each of which is effective to each scheduling criterion are used to make a composite dispatching rule. One of the two rules is SIMSET rule that selects the next job with the shortest setup time SIMSET rule can make effective schedules with regard to average setup time. The other rule is constructed using a neural network. This neural network is trained by a simulated annealing method to make effective schedules with regard to tardiness cost. The proposed composite rule has the parameter α and can make effective bi-criteria scheduling by tuning it. Numerical experiments show that the effectiveness of the proposed rule.

An Analytic Study of Two Probabilistic Models for Establishing Ad Hoc WLANs

Two contention-based algorithms are proposed in this paper for establishing ad hoc Wireless Local Area Networks (WLANs). They are named uniform and geometric, respectively, based on the nature of the probabilistic model in the decision making process used in each algorithm. The main performance consideration in this paper is the network setup time each algorithm requires. It is shown that the network setup time is an explicit function of the total solicitation rounds required. The paper is, therefore, focused on deriving the distribution of the total solicitation rounds needed for setting up a network. It is shown analytically that the performance of uniform is better than that of geometric. It also derives the asymptotic lower and upper bounds for the average setup times for uniform.

The Impact of Setup Times on the Performance of Multiclass Service and Production Systems

We characterize the impact of the means and distributions of setup times in polling systems. Starting with cyclic systems with exhaustive service, we show that the vector of average setup times affects the complete distribution of waiting times and queue sizes only via its sum. In addition, all moments of all waiting times and queue sizes are reduced if the higher order moments of one or more of the setup times is reduced while maintaining their means. We also obtain a fundamental decomposition result of the queue sizes at polling instants which is reminiscent of those obtained in vacation models. We show that the above results continue to apply under more general and more efficient protocols, e.g., general polling tables, systems with gated service, and systems with mixed service.

The optimum average lot size in split automatic transfer lines.

In this paper, we consider the modelling of split automatic transfer lines in which two types of jobs are manufactured and the setup is required in a common operation stage, and analyze the effects of average lot sizes, average setup times and buffer storage capacities on the line production rate. From the numerical analysis, it is shown that the average lot size has the opposite effect on the decrease of the production rates caused by the split interference and by the setup, and that there exists an optimum average lot size yielding the maximum production rate. Furthermore, we analyze the upper limit of the rate yielded in the lines with no split interference, and using the limit and another limit, i. e. the production rate yielded in the line with no setup, we estimate the optimum average lot size.

A simulation comparison of group technology with traditional job shop manufacturing

A simulation model of an actual job shop was used to compare group technology with traditional job shop manufacturing. The experiment compared shops which had four different layouts, designed to emphasize different features of traditional job shops and group technology shops, and four distributions of demand for end items. The group technology shops exhibited superior performance in terms of average move time and average set-up time. The traditional job shops had superior performance in queue related variables (average queue length, average waiting time, work-in-process inventory, etc.). This was caused by group technology's dedication of machines. The effects of the queue related variables outweighed the effects of average move time and average set-up time: the average flow time was shorter in the traditional job shop than in the group technology shops.