Optimal Computing Budget Allocation Technique Research Articles

Differential Evolution Algorithm Combined with Uncertainty Handling Techniques for Stochastic Reentrant Job Shop Scheduling Problem

This paper considers two kinds of stochastic reentrant job shop scheduling problems (SRJSSP), i.e., the SRJSSP with the maximum tardiness criterion and the SRJSSP with the makespan criterion. Owing to the NP-complete complexity of the considered RJSSPs, an effective differential evolutionary algorithm (DEA) combined with two uncertainty handling techniques, namely, DEA_UHT, is proposed to address these problems. Firstly, to reasonably control the computation cost, the optimal computing budget allocation technique (OCBAT) is applied for allocating limited computation budgets to assure reliable evaluation and identification for excellent solutions or individuals, and the hypothesis test technique (HTT) is added to execute a statistical comparison to reduce some unnecessary repeated evaluation. Secondly, a reentrant-largest-order-value rule is designed to convert the DEA’s individual (i.e., a continuous vector) to the SRJSSP’s solution (i.e., an operation permutation). Thirdly, a conventional active decoding scheme for the job shop scheduling problem is extended to decode the solution for obtaining the criterion value. Fourthly, an Insert-based exploitation strategy and an Interchange-based exploration strategy are devised to enhance DEA’s exploitation ability and exploration ability, respectively. Finally, the test results and comparisons manifest the effectiveness and robustness of the proposed DEA_UHT.

Simulation Optimization on Complex Job Shop Scheduling with Non-Identical Job Sizes

This paper addresses the complex job shop scheduling problem with the consideration of non-identical job sizes. By simultaneously considering practical constraints of sequence dependent setup times, incompatible job families and job dependent batch processing time, we formulate this problem into a simulation optimization problem based on the disjunctive graph representation. In order to find scheduling policies that minimise the expectation of mean weighted tardiness, we propose a genetic programming based hyper heuristic to generate efficient dispatching rules. And then, based on the nested partition framework together with the optimal computing budget allocation technique, a hybrid rule selection algorithm is proposed for searching machine group specified rule combinations. Numerical results show that the proposed algorithms outperform benchmark algorithms in both solution quality and robustness.

Simulation optimization with PSO and OCBA for semiconductor back-end assembly

This study examines a dynamic parallel machine scheduling problem in a hybrid flow shop for semiconductor back-end assembly. The subject is a multi-line, multi-stage facility with multi-type parallel machine groups, and orders are scheduled with different start times. As a typical make-to-order and contract manufacturing business model, the main objective is to achieve the minimum manufacturing lead time, and the main decisions are to find an optimal assignment of production line and machine type by stage for each order. Nevertheless, production behavior and a number of conditions increase complexity. Such conditions include order split and merge of jobs for parallel processing with stochastic processing time and compliance with quality and traceability requirements. We proposed a simulation optimization approach, using particle swarm optimization algorithm combined with optimal computing budget allocation technique as solution. The proposed approach provides a novel application for a complex production system and directions for further study.

Memetic Algorithm for Real-Time Combinatorial Stochastic Simulation Optimization Problems With Performance Analysis

A three-phase memetic algorithm (MA) is proposed to find a suboptimal solution for real-time combinatorial stochastic simulation optimization (CSSO) problems with large discrete solution space. In phase 1, a genetic algorithm assisted by an offline global surrogate model is applied to find N good diversified solutions. In phase 2, a probabilistic local search method integrated with an online surrogate model is used to search for the approximate corresponding local optimum of each of the N solutions resulted from phase 1. In phase 3, the optimal computing budget allocation technique is employed to simulate and identify the best solution among the N local optima from phase 2. The proposed MA is applied to an assemble-to-order problem, which is a real-world CSSO problem. Extensive simulations were performed to demonstrate its superior performance, and results showed that the obtained solution is within 1% of the true optimum with a probability of 99%. We also provide a rigorous analysis to evaluate the performance of the proposed MA.

Investigating the impact of operational variables on manufacturing cost by simulation optimization

In this paper, we focus on the relationship between operations-based variables (specifically, production speed, scrap rate and maintenance speed) and the manufacturing cost. These variables usually produce opposite influences on the variable cost and the fixed cost. For example, setting the production speed at a high level is beneficial for reducing the variable cost. However, maintaining the high speed incurs considerable fixed costs at the same time. Therefore, an optimization approach is necessary to determine the optimal values of the operational variables for minimizing the average cost. First, a discrete-event simulation procedure is designed for describing the stochastic production environment and for evaluating the settings. Then, an optimization approach based on the ordinal optimization (OO) philosophy and particle swarm optimization (PSO) is used to search in the continuous space of the operational variables. In this process, the optimal computing budget allocation technique is applied so as to fully utilize the computational resource and potentially save the computational time. Finally, numeric computations are conducted for verifying the effectiveness of the proposed algorithm. Sensitivity analysis and discussions are also presented.

Evolutionary algorithm assisted by surrogate model in the framework of ordinal optimization and optimal computing budget allocation

This work proposes an evolutionary algorithm (EA) that is assisted by a surrogate model in the framework of ordinal optimization (OO) and optimal computing budget allocation (OCBA) for use in solving the real-time combinatorial stochastic simulation optimization problem with a huge discrete solution space. For real-time applications, an off-line trained artificial neural network (ANN) is utilized as the surrogate model. EA, assisted by the trained ANN, is applied to the problem of interest to obtain a subset of good enough solutions, S. Also for real-time application, the OCBA technique is used to find the best solution in S, and this is the obtained good enough solution. Most importantly, a systematic procedure is provided for evaluating the performance of the proposed algorithm by estimating the distance of the obtained good enough solution from the optimal solution. The proposed algorithm is applied to a hotel booking limit (HBL) problem, which is a combinatorial stochastic simulation optimization problem. Extensive simulations are performed to demonstrate the computational efficiency of the proposed algorithm and the systematic performance evaluation procedure is applied to the HBL problem to quantify the goodness of the obtained good enough solution.

Optimal base-stock policy of the assemble-to-order systems

In this work, an ordinal optimization-based evolution algorithm (OOEA) is proposed to solve a problem for a good enough target inventory level of the assemble-to-order (ATO) system. First, the ATO system is formulated as a combinatorial optimization problem with integer variables that possesses a huge solution space. Next, the genetic algorithm is used to select N excellent solutions from the solution space, where the fitness is evaluated with the radial basis function network. Finally, we proceed with the optimal computing budget allocation technique to search for a good enough solution. The proposed OOEA is applied to an ATO system comprising 10 items on 6 products. The solution quality is demonstrated by comparing with those obtained by two competing methods. The good enough target inventory level obtained by the OOEA is promising in the aspects of solution quality and computational efficiency.