Da-Zhong Zheng

An effective hybrid optimization strategy for job-shop scheduling problems

Abstract Simulated annealing is a naturally serial algorithm, but its behavior can be controlled by the cooling schedule. Genetic algorithm exhibits implicit parallelism and can retain useful redundant information about what is learned from previous searches by its representation in individuals in the population, but GA may lose solutions and substructures due to the disruptive effects of genetic operators and is not easy to regulate GAs convergence. By reasonably combining these two global probabilistic search algorithms, we develop a general, parallel and easily implemented hybrid optimization framework, and apply it to job-shop scheduling problems. Based on effective encoding scheme and some specific optimization operators, some benchmark job-shop scheduling problems are well solved by the hybrid optimization strategy, and the results are competitive with the best literature results. Besides the effectiveness and robustness of the hybrid strategy, the combination of different search mechanisms and structures can relax the parameter-dependence of GA and SA. Scope and purpose Job-shop scheduling problem (JSP) is one of the most well-known machine scheduling problems and one of the strongly NP-hard combinatorial optimization problems. Developing effective search methods is always an important and valuable work. The scope and purpose of this paper is to present a parallel and easily implemented hybrid optimization framework, which reasonably combines genetic algorithm with simulated annealing. Based on effective encoding scheme and some specific optimization operators, the job-shop scheduling problems are well solved by the hybrid optimization strategy.

An improved evolutionary programming for optimization

To avoid premature convergence and balance the exploration and exploitation abilities of classic evolutionary programming, this paper proposes an improved evolutionary programming for optimization. Firstly, multiple populations are designed to perform parallel search with random initialization in divided solution spaces. Secondly, multiple mutation operators are designed to enhance the search templates. Thirdly, selection with probabilistic updating strategy based on annealing schedule like simulated annealing is applied to avoid the dependence on fitness function and to avoid being trapped in local optimum. Lastly, re-assignment strategy for individuals is designed for every sub-population to fuse information and enhance population diversity. Furthermore, the implementations of the proposed algorithm for function and combinatorial optimization problems are discussed and its effectiveness is demonstrated by numerical simulation based on some benchmarks.

A quantum-inspired genetic algorithm for scheduling problems

This paper is the first to propose a quantum-inspired genetic algorithm (QGA) for permutation flow shop scheduling problem to minimize the maximum completion time (makespan). In the QGA, Q-bit based representation is employed for exploration in discrete 0-1 hyperspace by using updating operator of quantum gate as well as genetic operators of Q-bit. Meanwhile, the Q-bit representation is converted to random key representation, which is then transferred to job permutation for objective evaluation. Simulation results and comparisons based on benchmarks demonstrate the effectiveness of the QGA, whose searching quality is much better than that of the famous NEH heuristic.

Ordinal optimisation of genetic control parameters for flow shop scheduling

Genetic algorithm (GA) has been widely applied to many non-polynomial hard optimisation problems, such as flow shop and job shop scheduling. It is well known that the efficiency and effectiveness of GA highly depend on its control parameters, but even setting suitable parameters often suffers from tedious trial and error. Currently, setting optimal parameters is still an open problem and one of the most important and promising areas for GA. In this paper, the determination of optimal GA control parameters with limited computational effort and total simulation replication constraint, namely, population size, crossover and mutation probabilities, is firstly formulated as a stochastic optimisation problem. Ordinal optimisation and optimal computing budget allocation are then applied to select the optimal GA control parameters while providing reasonable performance evaluation for hard flow shop scheduling problems. Lastly the effectiveness of the methodology is demonstrated by simulation results based on benchmarks.

Clusters partition and sensors configuration for target tracking in wireless sensor networks

Decisions on the number of clusters and the sensing radius will effectively affect the quality and energy metrics of WSN (wireless sensor networks) tracking systems. By presenting the mean number of the detectable sensors as a trajectory independent quality metric, an energy-quality optimization model is derived and Pareto based optimization strategy is proposed. The obtained non-bad solutions (Pareto Fronts) can be used to direct the clusters partition and sensors configuration. Comparing with simulation, more than 80% of these Pareto Fronts are coincident with those in experiment results.

Load balancing control of Web-server clusters: N-tanks model and a CTCT method

With an increasing of the traffic on Internet, a Web-server usually overloads during a heavy-request period. Web-server cluster architectures have been widely employed to provide high-efficiency Internet services. This paper is concerned with load balancing control of Web-server clusters. A Web-server is modelled as a tank with an inflow and an outflow, and hence a Web-server cluster is modelled as a multiple-tanks system with certain inflow constraint. A control technique, called continuous token control technique (CTCT), which adopts conventional PID approach is proposed to address the issue of the traffic balancing on the Web-server clusters. The principle of the proposed approach is explained and demonstrated. Simulation results show that the proposed CTCT technique is effective and scalable.

Fault tolerant supervisory for discrete event systems based on event observer

The fault tolerant supervisory problem for discrete event systems is addressed in this paper. The proposed approach is based on the state avoidance control theory and observer-based control for Petri net. The key idea of the authors is to use a simple linear algebraic formalism to estimate system states and generate diagnostic information. Hence, the state explosion problem is avoided and the observer-based fault diagnosis algorithm can be made on-line.

Switching Difference Control of Parallel Streams Temperatures

Abstract An industrial furnace with multiple parallel passes and multiple burners is commonly used in petroleum refineries to heat the preprocessed crude oil to a specific temperature. Due to that maintaining multiple outlet temperatures of such parallel passes equal is significant for improving product quality, plant safety, and economic efficiency, etc., great efforts have been taken to control such temperatures. In this paper, a control technique based on switching control schemes, called switching difference control technique (SDCT), is proposed to distribute the inlet oil flowrates such that the outlet temperatures are as identical as possible. The principle of the proposed technique is explained, and several switching policies are introduced. Simulation examples are provided to demonstrate the effectiveness of the proposed strategy. The SDCT technique has the following advantages: it avoids the flow valves too frequently being regulated; it solves the problem of the flow coupling among multiple passes conveniently, etc.

The Balanceabililty of the FMM Switching Control Policy

Abstract This paper mainly studies the balanceabililty of the FMM switching control policy for the generalized switched server system. The balanceability investigates the question that, under certain algebra constraint, whether the GSS system could be driven to reach the task-balancing status by the given switching control policy. Several switching control policies are first briefly reviewed, then the balanceability of the related switching control policies is elaborated, and finally a sufficient condition for the balanceability of the FMM policy is obtained.

System Analysis and Control Design for Generalized Switched Server Systems

Abstract The generalized switched server (GSS) systems model and some switching control strategies have been presented to address a class of load-balancing problems with control input algebra constraint. Based on the proposed framework of the GSS systems and the related switching control strategies, this paper mainly studies the problem of the system analysis and control design for the generalized switched server systems. Due to that the GSS system is a discretely controlled continuous time system, the procedure of the control design is divided into two steps: one is to design the continuous time controller, and the other is to design and select the discrete-event driven switching control strategy. The two steps are independent from each other, and can be finished separately.