Search results

Locating hub facilities is important in different types of transportation and communication networks. The p‐Hub Median Problem (p‐HMP) addresses a class of hub location problems in which all hubs are interconnected and each non‐hub node is assigned to a single hub. The hubs are uncapacitated, and their number p is initially determined. Introduces an Artificial Intelligence (AI) heuristic called simulated annealing to solve the p‐HMP. The results are compared against another AI heuristic, namely Tabu Search, and against two other non‐AI heuristics. A real world data set of airline passenger flow in the USA, and randomly generated data sets are used for computational testing. The results confirm that AI heuristic approaches to the p‐HMP outperform non‐AI heuristic approaches on solution quality.

The purpose of this paper is to present an application of the simulated annealing algorithm to the redundant system reliability optimization. Its main aim is to analyze and compare this optimization method performance with those of similar application.

The methods that were used to compare results are the genetic algorithm, the Lagrange Multipliers, and the evolution strategy. A hybrid algorithm composed by simulated annealing and genetic algorithm was developed in order to achieve the general applicability of the methods. The hybrid algorithm also tries to exploit the positive aspects of each method.

The results presented by the simulated annealing and the hybrid algorithm are significant, and validate the methods as a robust tool for parameter optimization in mechanical projects development.

The main objective is to propose a method for redundancy optimization in mechanical systems, which are not as large as electric and electronic systems, but involves high costs associated to redundancy and requires a high level of safety standards like: automotive and aerospace systems.

This paper aims to present a new model for solving the integrated production planning and scheduling. Usually, the two decision levels are treated sequentially because of their complexity. Scheduling depends on the lot sizes calculated at the tactical level and ignoring scheduling constraints generates unrealistic and inconsistent decisions. Therefore, integrating more detail scheduling constraint in production planning is important for managing efficiently operations. Therefore, an integrated model was developed, and two evolutionary optimization approaches were suggested for solving it, namely, genetic algorithm (GA) and the hybridization of simulated annealing (SA) with GA HSAGA. The proposed algorithms have some parameters that must be adjusted using Taguchi method. Therefore, to evaluate the proposed algorithm, the authors compared the results given by GA and the hybridization. The SA-based local search is embedded into a GA search mechanism to move the GA away from being closed within local optima. The analysis shows that the combination of simulated annealing with GA gives better solutions and minimizes the total production costs.

The paper opted for an approached resolution method particularly GA and simulated annealing. The study represents a comparison between the results found using GA and the hybridization of simulated annealing and GA. A total of 45 instances were studied to evaluate job-shop problems of different sizes.

The results illustrate that for 36 instances of 45, the hybridization of simulated annealing and GA HSAGA has provided best production costs. The efficiency demonstrated by HSAGA approach is related to the combination between the exploration ability of GA and the capacity to escape local optimum of simulated annealing.

This study provides a new resolution approach to the integration of planning and scheduling while considering a new operational constrain. The model suggested aims to control the available capacity of the resources and guaranties that the resources to be consumed do not exceed the real availability to avoid the blocking that results from the unavailability of resources. Furthermore, to solve the MILP model, a GA is proposed and then it is combined to simulated annealing.

The performance of a welding process determines not only the cost, but also the quality of the product. How to control the welding process in order to ensure good welding performance with less cost and higher Productivity has become critical. The objective of this study is twofold: (1) developing artificial neural networks to predict welding performance using different learning algorithms: back propagation, simulated annealing and tabu search; (2) comparing and discussing the performance of neural networks trained using those algorithms. Statistical analysis shows that back propagation is able to make more accurate prediction than the other algorithms for this particular application. However, all three algorithms demonstrate impressive flexibility and robustness.

Robots are used in assembly lines because of their higher flexibility and lower costs. The purpose of this paper is to develop mathematical models and simulated annealing algorithms to solve the robotic assembly line balancing (RALB-II) to minimize the cycle time.

Four mixed-integer linear programming models are developed and encoded in CPLEX solver to find optimal solutions for small-sized problem instances. Two simulated annealing algorithms, original simulated annealing algorithm and restarted simulated annealing (RSA) algorithm, are proposed to tackle large-sized problems. The restart mechanism in the RSA methodology replaces the incumbent temperature with a new temperature. In addition, the proposed methods use iterative mechanisms for updating cycle time and a new objective to select the solution with fewer critical workstations.

The comparative study among the tested algorithms and other methods adapted verifies the effectiveness of the proposed methods. The results obtained by these algorithms on the benchmark instances show that 23 new upper bounds out of 32 tested cases are achieved. The RSA algorithm ranks first among the algorithms in the number of updated upper bounds.

Four models are developed for RALBP-II and their performance is evaluated for the first time. An RSA algorithm is developed to solve RALBP-II, where the restart mechanism is developed to replace the incumbent temperature with a new temperature. The proposed methods also use iterative mechanisms and a new objective to select the solution with fewer critical workstations.

The purpose of this paper is to present a study of the effect of different types of annealing schedules for a ribonucleic acid (RNA) secondary structure prediction algorithm based on simulated annealing (SA).

An RNA folding algorithm was implemented that assembles the final structure from potential substructures (helixes). Structures are encoded as a permutation of helixes. An SA searches this space of permutations. Parameters and annealing schedules were studied and fine-tuned to optimize algorithm performance.

In comparing with mfold, the SA algorithm shows comparable results (in terms of F-measure) even with a less sophisticated thermodynamic model. In terms of average specificity, the SA algorithm has provided surpassing results.

Most of the underlying thermodynamic models are too simplistic and incomplete to accurately model the free energy for larger structures. This is the largest limitation of free energy-based RNA folding algorithms in general.

The algorithm offers a different approach that can be used in practice to fold RNA sequences quickly.

The algorithm is one of only two SA-based RNA folding algorithms. The authors use a very different encoding, based on permutation of candidate helixes. The in depth study of annealing schedules and other parameters makes the algorithm a strong contender. Another benefit is that new thermodynamic models can be incorporated with relative ease (which is not the case for algorithms based on dynamic programming).

Disturbing increase in the use of virgin resources to produce new products has threatened the environment. Many countries have reacted to this situation through regulations which aim to eliminate negative impact of products on the environment shaping the concept of environmentally conscious manufacturing and product recovery (ECMPRO). The first crucial and the most time-consuming step of product recovery is disassembly. The best productivity rate is achieved via a disassembly line in an automated disassembly process. In this chapter, we consider a sequence-dependent disassembly line balancing problem (SDDLBP) with multiple objectives that is concerned with the assignment of disassembly tasks to a set of ordered disassembly workstations while satisfying the disassembly precedence constraints and optimizing the effectiveness of several measures considering sequence-dependent time increments among disassembly tasks. Due to the high complexity of the SDDLBP, there is currently no known way to optimally solve even moderately sized instances of the problem. Therefore, an efficient methodology based on the simulated annealing (SA) is proposed to solve the SDDLBP. Case scenarios are considered and comparisons with ant colony optimization (ACO), particle swarm optimization (PSO), river formation dynamics (RFD), and tabu search (TS) approaches are provided to demonstrate the superior functionality of the proposed algorithm.

In order to achieve excellence in manufacturing, goals like lean, economic and quality production with enhanced productivity play a crucial role in this competitive environment. It also necessitates major improvements in generally three primary technical areas: variation reduction, equipment reliability, and production scheduling. Complexity of the real world scheduling problems also increases with interactive multiple decision‐making criteria. This paper aims to deal with multi‐objective flow shop scheduling problems, including sequence dependent set up time (SDST). The paper also aims to consider the objective of minimizing the weighted sum of total weighted tardiness, total weighted earliness and makespan simultaneously. It proposes a new heuristic‐based hybrid simulated annealing (HSA) for near optimal solutions in a reasonable time.

Six modified NEH's based HSA algorithms are proposed for efficient scheduling of jobs in a multi‐objective SDST flow shop. Problems of up to 200 jobs and 20 machines are tested by the proposed HSA and a defined relative percentage improvement index is used for analysis and comparison of different MNEH's based hybrid simulated annealing algorithms.

From the results, it has been derived that performance of SA_EWDD (NEH) up to ten machines' problems, and SA_EPWDD (NEH) up to 20 machines' problems, were better over others especially for large sized SDST flow shop scheduling problems for the considered multi‐objective fitness function.

HSA and multi‐objective decision making proposed in the present work is a modified approach in the area of SDST flow shop scheduling.

The purpose of this paper is to propose a novel method under the name of genetic simulated annealing algorithm (GSAA) and ant colony optimization (ACO) algorithm for assembly sequence planning (ASP) which is possessed of the competence for assisting the planner in generating a satisfied and effective assembly sequence with respect to large constraint assembly perplexity.

Based on the genetic algorithm (GA), simulated annealing, and ACO algorithm, the GSAA are put forward. A case study is presented to validate the proposed method.

This GSAA has better optimization performance and robustness. The degree of dependence on the initial assembly sequence about GSAA is decreased. The optimization assembly sequence still can be obtained even if the assembly sequences of initial population are infeasible. By combining GA and simulated annealing (SA), the efficiency of searching and the quality of solution of GSAA is improved. As for the presented ACO algorithm, the searching speed is further increased.

Traditionally, GA heavily depends on the choosing original sequence, which can result in early convergence in iterative operation, lower searching efficiency in evolutionary process, and non‐optimization of final result for global variable. Similarly, SA algorithms may generate a great deal of infeasible solutions in the evolution process by generating new sequences through exchanging position of the randomly selected two parts, which results in inefficiency of the solution‐searching process. In this paper, the proposed GSAA and ACO algorithm for ASP are possessed of the competence for assisting the planner in generating a satisfied and effective assembly sequence with respect to large constraint assembly perplexity.

The purpose of this paper is to: develop an effective cellular manufacturing system (CMS) design methodology by simultaneously considering system costs and individual machine reliabilities; and propose a combinatorial search‐based solution procedure to solve large‐sized problems.

This paper presents a multi‐objective mixed integer‐programming model for the design of CMS with the objective of minimizing costs and maximizing system reliability. The approach optimizes inter‐cell material handling costs, the variable cost of machining operations, and the machine under‐utilization costs. It also maximizes the system reliability by selecting process routes for the part types with the highest system reliability for the machines along the routes. To solve the multi‐objective, multiple process plan model, a simulated annealing (SA)‐based algorithm is developed. The algorithm follows the basic steps of SA, but also incorporates the genetic algorithm (GA) operations of crossover and mutations to generate better neighboring solutions from the current good solutions.

The algorithm in the paper solves the multi‐objective CMS design model and generates near optimal solutions for medium to large‐sized problems within reasonable limits of CPU time.

In the paper the CMS design approach can be implemented to improve reliability performance of the CMS.

A new CMS design methodology in this paper, which minimizes system costs and maximizes machine‐related system reliability, is developed. The proposed algorithm, which combines the basic steps of SA and crossover and mutation operations of GA, will enable CMS designers and users to obtain near optimal solutions for practical‐sized problems within reasonable time limits.