GA-based learning bias selection mechanism for real-time scheduling systems

The use of machine learning technologies in order to develop knowledge bases (KBs) for real-time scheduling (RTS) problems has produced encouraging results in recent researches. However, few researches focus on the manner of selecting proper learning biases in the early developing stage of the RTS system to enhance the generalization ability of the resulting KBs. The selected learning bias usually assumes a set of proper system features that are known in advance. Moreover, the machine learning algorithm for developing scheduling KBs is predetermined. The purpose of this study is to develop a genetic algorithm (GA)-based learning bias selection mechanism to determine an appropriate learning bias that includes the machine learning algorithm, feature subset, and learning parameters. Three machine learning algorithms are considered: the back propagation neural network (BPNN), C4.5 decision tree (DT) learning, and support vector machines (SVMs). The proposed GA-based learning bias selection mechanism can search the best machine learning algorithm and simultaneously determine the optimal subset of features and the learning parameters used to build the RTS system KBs. In terms of the accuracy of prediction of unseen data under various performance criteria, it also offers better generalization ability as compared to the case where the learning bias selection mechanism is not used. Furthermore, the proposed approach to build RTS system KBs can improve the system performance as compared to other classifier KBs under various performance criteria over a long period.     

Fuzzy rule generation for adaptive scheduling in a dynamic manufacturing environment

This paper proposes a fuzzy rule-based system for an adaptive scheduling, which dynamically selects and applies the most suitable strategy according to the current state of the scheduling environment. The adaptive scheduling problem is generally considered as a classification task since the performance of the adaptive scheduling system depends on the effectiveness of the mapping knowledge between system states and the best rules for the states. A rule base for this mapping is built and evolved by the proposed fuzzy dynamic learning classifier based on the training data cumulated by a simulation method. Distributed fuzzy sets approach, which uses multiple fuzzy numbers simultaneously, is adopted to recognize the system states. The developed fuzzy rules may readily be interpreted, adopted and, when necessary, modified by human experts. An application of the proposed method to a job-dispatching problem in a hypothetical flexible manufacturing system (FMS) shows that the method can develop more effective and robust rules than the traditional job-dispatching rules and a neural network approach.     

Online preference learning for adaptive dispatching of AGVs in an automated container terminal

This paper proposes an online preference learning algorithm named OnPL that can dynamically adapt the policy for dispatching AGVs to changing situations in an automated container terminal. The policy is based on a pairwise preference function that can be repeatedly applied to multiple candidate jobs to sort out the best one. An adaptation of the policy is therefore made by updating this preference function. After every dispatching decision, each of all the candidate jobs considered for the decision is evaluated by running a simulation of a short look-ahead horizon. The best job is then paired with each of the remaining jobs to make training examples of positive preferences, and the inversions of these pairs are each used to generate examples of negative preferences. These new training examples, together with some additional recent examples in the reserve pool, are used to relearn the preference function implemented by an artificial neural network. The experimental results show that OnPL can relearn its policy in real time, and can thus adapt to changing situations seamlessly. In comparison to OnPL, other methods cannot adapt well enough or are not applicable in real time owing to the very long computation time required.     

Maintenance scheduling in manufacturing systems based on predicted machine degradation

In this paper, we propose a new method for scheduling of maintenance operations in a manufacturing system using the continuous assessment and prediction of the level of performance degradation of manufacturing equipment, as well as the complex interaction between the production process and maintenance operations. Effects of any maintenance schedule are evaluated through a discrete-event simulation that utilizes predicted probabilities of machine failures in the manufacturing system, where predicted probabilities of failure are assumed to be available either from historical equipment reliability information or based on the newly available predictive algorithms. A Genetic Algorithm based optimization procedure is used to search for the most cost-effective maintenance schedule, considering both production gains and maintenance expenses. The algorithm is implemented in a simulated environment and benchmarked against several traditional maintenance strategies, such as corrective maintenance, scheduled maintenance and condition-based maintenance. In all cases that were studied, application of the newly proposed maintenance scheduling tool resulted in a noticeable increase in the cost-benefits, which indicates that the use of predictive information about equipment performance through the newly proposed maintenance scheduling method could result in significant gains obtained by optimal maintenance scheduling.     

A neural network model and algorithm for the hybrid flow shop scheduling problem in a dynamic environment

A hybrid flow shop (HFS) is a generalized flow shop with multiple machines in some stages. HFS is fairly common in flexible manufacturing and in process industry. Because manufacturing systems often operate in a stochastic and dynamic environment, dynamic hybrid flow shop scheduling is frequently encountered in practice. This paper proposes a neural network model and algorithm to solve the dynamic hybrid flow shop scheduling problem. In order to obtain training examples for the neural network, we first study, through simulation, the performance of some dispatching rules that have demonstrated effectiveness in the previous related research. The results are then transformed into training examples. The training process is optimized by the delta-bar-delta (DBD) method that can speed up training convergence. The most commonly used dispatching rules are used as benchmarks. Simulation results show that the performance of the neural network approach is much better than that of the traditional dispatching rules.This revised version was published in June 2005 with corrected page numbers.     

Material supply scheduling in a ubiquitous manufacturing system

Recently, ubiquitous manufacturing has attracted wide attention in both academia and industry. To create a successful ubiquitous manufacturing system, an efficient material handling system is essential. In accordance with this reason, mobile robots have been used for transporting materials. This paper aims at developing a methodology for scheduling the material supply for a single mobile robot in a ubiquitous manufacturing environment. In this type of environment, the processing rate of the materials along with supply quantity corresponds to the cycle of material supply. The carrying capacity of the robots are limited and thus the problem of determining the material supply quantity and material supply schedule without lack of materials for production or service processes becomes complicated. In this work, a nonlinear program is formulated to schedule the supply of material and determine the required material quantity. A heuristic algorithm based on genetic algorithm is developed to solve the problem. From the numerical experiments conducted in this study, it is observed that the proposed algorithm shows good performance and can also be implemented to solve large scale problems.     

Intelligent acoustic rotor speed estimation for an autonomous helicopter

Acoustic sensing to gather information about a machine can be highly beneficial, but processing the data can be difficult. In this work, a variety of methodologies have been studied to extract rotor speed information from the sound signature of an autonomous helicopter, with no a-priori knowledge of its underlying acoustic properties.     

Learning enabled cooperative agent behavior in an evolutionary and competitive environment

The proposed method is implemented in three steps: first, when a variation in environment is perceived, agents take appropriate actions. Second, the behaviors are stimulated and controlled through communication with other agents. Finally, the most frequently stimulated behavior is adopted as a group behavior strategy. In this paper, two different reward models, reward model 1 and reward model 2, are applied. Each reward model is designed to consider the reinforcement or constraint of behaviors. In competitive agent environments, the behavior considered to be advantageous is reinforced as adding reward values. On the contrary, the behavior considered to be disadvantageous is constrained by reducing the reward values. The validity of this strategy is verified through simulation.     

Self-teaching adaptive dynamic programming for Gomoku

In this paper adaptive dynamic programming (ADP) is applied to learn to play Gomoku. The critic network is used to evaluate board situations. The basic idea is to penalize the last move taken by the loser and reward the last move selected by the winner at the end of a game. The results show that the presented program is able to improve its performance by playing against itself and has approached the candidate level of a commercial Gomoku program called 5-star Gomoku. We also examined the influence of two methods for generating games: self-teaching and learning through watching two experts playing against each other and presented the comparison results and reasons.     

Developing a dynamic rolling-horizon decision strategy for yard crane scheduling

As a hub for land and marine transportation, container terminals play an important role in global trade. In today’s competitive environment, container terminals should improve their service quality, i.e., effective space resource handling and equipment resource scheduling, for their prosperity or even survival. Although intensive researches were attempted on yard crane scheduling, the solutions from these approaches likely reached a local optimum, and thereafter a rational strategy towards global optimum was still lacking. Accordingly, it became an imperative to explore a rational strategy for this purpose. To resolve this problem, a novel dynamic rolling-horizon decision strategy was proposed for yard crane scheduling in this study. Initially, an integer programming model was established to minimize the total task delaying at blocks. Due to the computational scale with regard to the yard crane scheduling problem, a heuristic algorithm, along with a simulation model, was then applied. In this fashion, the simulation model was next investigated to alternate the periods and evaluate the task delaying. Subsequently, a genetic algorithm was employed to optimize the initial solutions generated. Consequently, computational experiments were used to illustrate the proposed strategy for yard crane scheduling and verify the effectiveness and efficiency of the proposed approach.