Search results

The purpose of this paper is to study the use of a heterogeneous and evolutionary team approach based on different sources of knowledge to address a real-world problem within the telecommunication domain: the frequency assignment problem (FAP). Evolutionary algorithms have been proved as very suitable strategies when they are used to solve NP-hard optimization problems. However, these algorithms can find difficulties when they fall into local minima and the generation of high-quality solutions when tacking real-world instances of the problem is computationally very expensive. In this scenario, the use of a heterogeneous parallel team represents a very interesting approach.

The results have been validated by using two real-world telecommunication instances which contain real information about two GSM networks. Contrary to most of related publications, this paper is focussed on aspects which are relevant for real communication networks. Moreover, due to the stochastic nature of metaheuristics, the results are validated through a formal statistical analysis. This analysis is divided in two stages: first, a complete statistical study, and after that, a full comparative study against results previously published.

Comparative study shows that a heterogeneous evolutionary proposal obtains better results than proposals which are based on a unique source of knowledge. In fact, final results provided in the work surpass the results of other relevant studies previously published in the literature.

The paper provides a complete study of the contribution provided by the different metaheuristics included in the team and the impact of using different sources of evolutionary knowledge when the system is applied to solve a real-world FAP problem. The conclusions obtained in this study represent an original contribution never reached before for FAP.

Marine transportation has been faced with an increasing demand for containerized cargo during the past decade. Marine container terminals (MCTs), as the facilities for connecting seaborne and inland transportation, are expected to handle the increasing amount of containers, delivered by vessels. Berth scheduling plays an important role for the total throughput of MCTs as well as the overall effectiveness of the MCT operations. This study aims to propose a novel island-based metaheuristic algorithm to solve the berth scheduling problem and minimize the total cost of serving the arriving vessels at the MCT.

A universal island-based metaheuristic algorithm (UIMA) was proposed in this study, aiming to solve the spatially constrained berth scheduling problem. The UIMA population was divided into four sub-populations (i.e. islands). Unlike the canonical island-based algorithms that execute the same metaheuristic on each island, four different population-based metaheuristics are adopted within the developed algorithm to search the islands, including the following: evolutionary algorithm (EA), particle swarm optimization (PSO), estimation of distribution algorithm (EDA) and differential evolution (DE). The adopted population-based metaheuristic algorithms rely on different operators, which facilitate the search process for superior solutions on the UIMA islands.

The conducted numerical experiments demonstrated that the developed UIMA algorithm returned near-optimal solutions for the small-size problem instances. As for the large-size problem instances, UIMA was found to be superior to the EA, PSO, EDA and DE algorithms, which were executed in isolation, in terms of the obtained objective function values at termination. Furthermore, the developed UIMA algorithm outperformed various single-solution-based metaheuristic algorithms (including variable neighborhood search, tabu search and simulated annealing) in terms of the solution quality. The maximum UIMA computational time did not exceed 306 s.

Some of the previous berth scheduling studies modeled uncertain vessel arrival times and/or handling times, while this study assumed the vessel arrival and handling times to be deterministic.

The developed UIMA algorithm can be used by the MCT operators as an efficient decision support tool and assist with a cost-effective design of berth schedules within an acceptable computational time.

A novel island-based metaheuristic algorithm is designed to solve the spatially constrained berth scheduling problem. The proposed island-based algorithm adopts several types of metaheuristic algorithms to cover different areas of the search space. The considered metaheuristic algorithms rely on different operators. Such feature is expected to facilitate the search process for superior solutions.

The purpose of this study is to develop a new parallel metaheuristic algorithm for solving unconstrained continuous optimization problems.

The proposed method brings several metaheuristic algorithms together to form a coalition under Weighted Superposition Attraction-Repulsion Algorithm (WSAR) in a parallel computing environment. The proposed approach runs different single solution based metaheuristic algorithms in parallel and employs WSAR (which is a recently developed and proposed swarm intelligence based optimizer) as controller.

The proposed approach is tested against the latest well-known unconstrained continuous optimization problems (CEC2020). The obtained results are compared with some other optimization algorithms. The results of the comparison prove the efficiency of the proposed method.

This study aims to combine different metaheuristic algorithms in order to provide a satisfactory performance on solving the optimization problems by benefiting their diverse characteristics. In addition, the run time is shortened by parallel execution. The proposed approach can be applied to any type of optimization problems by its problem-independent structure.

The purpose of this paper is to address a multiobjective FAP (frequency assignment problem) formulation. More precisely, two conflicting objectives – the interference cost and the separation cost – are considered to characterize FAP as an MO (multiobjective optimization) problem.

The contribution to this specific telecommunication problem in a real scenario follows a recent approach, for which the authors have already accomplished some preliminary results. In this paper, a much more complete analysis is performed, including two well‐known algorithms (such as the NSGA‐II and SPEA2), with new results, new comparisons and statistical studies. More concretely, in this paper five different algorithms are presented and compared. The popular multiobjective algorithms, NSGA‐II and SPEA2, are compared against the Differential Evolution with Pareto Tournaments (DEPT) algorithm, the Greedy Multiobjective Variable Neighborhood Search (GMO‐VNS) algorithm and its variant Greedy Multiobjective Skewed Variable Neighborhood Search (GMO‐SVNS). Furthermore, the authors also contribute with a new design of multiobjective metaheuristic named Multiobjective Artificial Bee Colony (MO‐ABC) that is included in the comparison; it represents a new metaheuristic that the authors have developed to address FAP. The results were analyzed using two complementary indicators: the hypervolume indicator and the coverage relation. Two large‐scale real‐world mobile networks were used to validate the performance comparison made among several multiobjective metaheuristics.

The final results show that the multiobjective proposal is very competitive, clearly surpassing the results obtained by the well‐known multiobjective algorithms (NSGA‐II and SPEA2).

The paper provides a comparison among several multiobjective metaheuristics to solve FAP as a real‐life telecommunication engineering problem. A new multiobjective metaheuristic is also presented. Preliminary results were enhanced with two well‐known multiobjective algorithms. To the authors' knowledge, they have never been investigated for FAP.

Metaheuristic algorithms have been commonly used as an optimisation tool in various fields. However, optimisation of real-world problems has become increasingly challenging with to increase in system complexity. This situation has become a pull factor to introduce an efficient metaheuristic. This study aims to propose a novel sport-inspired algorithm based on a football playing style called tiki-taka.

The tiki-taka football style is characterised by short passing, player positioning and maintaining possession. This style aims to dominate the ball possession and defeat opponents using its tactical superiority. The proposed tiki-taka algorithm (TTA) simulates the short passing and player positioning behaviour for optimisation. The algorithm was tested using 19 benchmark functions and five engineering design problems. The performance of the proposed algorithm was compared with 11 other metaheuristics from sport-based, highly cited and recent algorithms.

The results showed that the TTA is extremely competitive, ranking first and second on 84% of benchmark problems. The proposed algorithm performs best in two engineering design problems and ranks second in the three remaining problems.

The originality of the proposed algorithm is the short passing strategy that exploits a nearby player to move to a better position.

The purpose of this paper is to demonstrate the applicability of swarm and evolutionary techniques for regularized machine learning. Generally, by defining a proper penalty function, regularization laws are embedded into the structure of common least square solutions to increase the numerical stability, sparsity, accuracy and robustness of regression weights. Several regularization techniques have been proposed so far which have their own advantages and disadvantages. Several efforts have been made to find fast and accurate deterministic solvers to handle those regularization techniques. However, the proposed numerical and deterministic approaches need certain knowledge of mathematical programming, and also do not guarantee the global optimality of the obtained solution. In this research, the authors propose the use of constraint swarm and evolutionary techniques to cope with demanding requirements of regularized extreme learning machine (ELM).

To implement the required tools for comparative numerical study, three steps are taken. The considered algorithms contain both classical and swarm and evolutionary approaches. For the classical regularization techniques, Lasso regularization, Tikhonov regularization, cascade Lasso-Tikhonov regularization, and elastic net are considered. For swarm and evolutionary-based regularization, an efficient constraint handling technique known as self-adaptive penalty function constraint handling is considered, and its algorithmic structure is modified so that it can efficiently perform the regularized learning. Several well-known metaheuristics are considered to check the generalization capability of the proposed scheme. To test the efficacy of the proposed constraint evolutionary-based regularization technique, a wide range of regression problems are used. Besides, the proposed framework is applied to a real-life identification problem, i.e. identifying the dominant factors affecting the hydrocarbon emissions of an automotive engine, for further assurance on the performance of the proposed scheme.

Through extensive numerical study, it is observed that the proposed scheme can be easily used for regularized machine learning. It is indicated that by defining a proper objective function and considering an appropriate penalty function, near global optimum values of regressors can be easily obtained. The results attest the high potentials of swarm and evolutionary techniques for fast, accurate and robust regularized machine learning.

The originality of the research paper lies behind the use of a novel constraint metaheuristic computing scheme which can be used for effective regularized optimally pruned extreme learning machine (OP-ELM). The self-adaption of the proposed method alleviates the user from the knowledge of the underlying system, and also increases the degree of the automation of OP-ELM. Besides, by using different types of metaheuristics, it is demonstrated that the proposed methodology is a general flexible scheme, and can be combined with different types of swarm and evolutionary-based optimization techniques to form a regularized machine learning approach.

In recent times, fuzzy logic is gaining more and more attention, and this is because of the capability of understanding the functioning of the system as per human knowledge-based system. The main contribution of the work is dynamically adapting the important parameters throughout the execution of the flower pollination algorithm (FPA) using concepts of fuzzy logic. By adapting the main parameters of the metaheuristics, the performance and accuracy of the metaheuristic have been improving in a varied range of applications.

The fuzzy logic-based parameter adaptation in the FPA is proposed. In addition, type-2 fuzzy logic is used to design fuzzy inference system for dynamic parameter adaptation in metaheuristics, which can help in eliminating uncertainty and hence offers an attractive improvement in dynamic parameter adaption in metaheuristic method, and, in reality, the effectiveness of the interval type-2 fuzzy inference system (IT2 FIS) has shown to provide improved results as matched to type-1 fuzzy inference system (T1 FIS) in some latest work.

One case study is considered for testing the proposed approach in a fault tolerant control problem without faults and with partial loss of effectiveness of main actuator fault with abrupt and incipient nature. For comparison between the type-1 fuzzy FPA and interval type-2 fuzzy FPA is presented using statistical analysis which validates the advantages of the interval type-2 fuzzy FPA. The statistical Z-test is presented for comparison of efficiency between two fuzzy variants of the FPA optimization method.

The main contribution of the work is a dynamical adaptation of the important parameters throughout the execution of the flower pollination optimization algorithm using concepts of type-2 fuzzy logic. By adapting the main parameters of the metaheuristics, the performance and accuracy of the metaheuristic have been improving in a varied range of applications.

This work highlights the optimization of the consumed amount of sewing thread required to make up a pair of jeans using three different metaheuristic methods; particular swarm optimization (PSO), ant colony optimization (ACO) and genetic algorithm (GA) techniques. Indeed, using metaheuristic optimization techniques enable industrialists to reach the lowest sewing thread quantities in terms of bobbins per garments. Besides, the compared results of this research can obviously prove the impact of each input parameter on the optimization of the sewing thread consumption per pair of jeans.

To assess objectively the sewing thread consumption, the optimized sewing conditions such as thread composition, needle size and fabric composition are investigated and discussed. Hence, a Taguchi design was elaborated to evaluate and optimize objectively the linear model consumption. Thanks to its principal characteristics and popularity, denim fabric is selected to analyze objectively the effects of studied input parameters. In addition, having workers with same skills and qualifications to repeat each time the same sewing process will involve having the same sewing thread consumption values. This can occur in some levels such as end of sewing, the number of machine failures, the kind of failure and its complexity, the competency of the mechanic and his way to repair failure, the loss of thread caused by threading and its frequency. Seam repetition due to operator lack of skill will obviously affect clothing appearance and hence quality decision. Interesting findings and significant relationship between input parameters and the amount of sewing thread consumption are established.

According to the comparative results obtained using metaheuristic methods, the PSO and ACO technique gives the lowest values of the consumption within the best combination of input parameters. The results show the accuracy of the applied metaheuristic methods to optimize the consumed amount needed to sew a pair of jeans with a notable superiority of both PSO and ACO methods compared to experimental ones. However, compared to GA method, ACO and PSO algorithms remained the most accurate techniques allowing industrials to minimize the consumed thread used to sew jeans. They can also widely optimize and predict the consumed thread in the investigated experimental design of interest. Consequently, compared to experimental results and regarding the low error values obtained, it may be concluded that the metaheuristic methods can optimize and evaluate both studied input and output parameters accurately.

This study is most useful for denim industrial applications, which makes it possible to anticipate, calculate and minimize the high consumption of sewing threads. This paper has not only practical implications for clothing appearance and quality but also for reduction in thread wastage occurring during shop floor conditions like machine running, thread breakage, repairs, etc. (Kawabata and Niwa, 1991). Unless the used sewing machine is equipped within a thread trimmer improvement in garment seam appearance cannot be achieved. By comparing and analyzing the operating activities of the regular lock stitch 301 machine with and without a thread trimmer, a difference in time processing can be grasped (Magazine JUKI Corporation, 2008). Time consumed in trimming by a lockstitch machine without a thread trimmer equals 3.1 s compared to 2.6 s by a thread trimming one. Hence, the reduction rate in the time processing equals 16.30%. This paper aimed to implement the optimal consumption (thread waste outstanding number of trials). Unless highly skilled workers are selected and well-motivated, the previous recommended changes will not be applied. The saved cost of the sewing thread reduction can be used to buy a better quality of fabric and/or thread. However, these factors are not always the same as they can vary according to customer's requirements because thread consumption is never a standard for sewn product categories such as trousers, shirts and footwear (Khedher and Jaouachi, 2015).

Until now, there is no work dealing with the investigation of the metaheuristic optimization of the consumed thread per pair of jeans to minimize accurately the amount of sewing thread as well as the sewing thread wastage. Even though these techniques of optimization are currently in full development due to some advantages such as generality and possible application to a large class of combinatorial and constrained assignment problems, efficiency for many problems in providing good quality approximate solutions for a large number of classical optimization problems and large-scale real applications, etc., are not applied yet to decrease sewing thread consumption. Some recent published works used statistical techniques (Taguchi, factorial, etc.), to evaluate approximate consumptions; conversely, other geometrical and mathematical approaches, considering some assumptions, used stitch geometry and remained insufficient to give the industrialists an implemented application generating the exact value of the consumed amount of sewing thread. Generally, in the clothing field 10–15% of sewing thread wastage should be added to the experimental approximate consumption value. Moreover, all investigations are focused on the approximative evaluations and theoretical modeling of sewing thread consumption as function of some input parameters. Practically, the obtained results are successfully applied and the ACO method gives the most accurate results. On the other hand, in the point of view of industrialists the applied metaheuristic methods (based on algorithms) used to decrease the amount of consumed thread remained an easy and fruitful solution that can allow them to control the number of sewing thread bobbin per garments.

The purpose of this paper is to optimize the frictional input parameters related to the yarn and woven fabric samples. Indeed, using metaheuristic techniques for optimization, it helps to attempt the best quality appearance of garment, by analysing their effects and relationships with the bagging behaviour of tested fabrics before and after bagging test. Using metaheuristic techniques allows us to select widely the minimal residual bagging properties and the optimized inputs to adjust them for this goal.

The metaheuristic methods were applied and discussed. Hence, the genetic algorithms (GA) and ant colony optimization (ACO) technique results are compared to select the best residual bagging behaviour and their correspondent parameters. The statistical analysis steps were implemented using Taguchi experimental design thanks to Minitab 14 software. The modelling methodology analysed in this paper deals with the linear regression method application and analysis to prepare to the optimization steps.

The regression results are essential for evaluate the effectiveness of the relationships founded between inputs and outputs parameters and for their optimizations in the design of interest.

This study is interesting for denim consumers and industrial applications during long and repetitive uses. Undoubtedly, the denim garments remained the largely used and consumed, hence, this particularity proves the necessity to study it in order to optimize the bagging phenomenon which occurs as function of number of uses. Although it is fashionable to have bagging, the denim fabric remains, in contrast with the worsted ones, the most popular fabric to produce garments. Moreover, regarding this characteristic, the large uses and the acceptable value of denim fabrics, their aesthetic appearance behaviour due to bagging phenomenon can be analysed and optimized accurately because compared to worsted fabrics, they have a high value and the repetitive tests to investigate widely bagged zones can fall the industrial. The paper has practical implications in the clothing appearance and other textile industry, especially in the weaving process when friction forms (yarn-to-yarn, yarn-to-metal frictions) and stresses are drastic. This can help to understand why residual bagging behaviour remained after garment uses due to the internal stress and excessive extensions.

Until now, there is no work dealing with the optimization of bagging behaviour using metaheuristic techniques. Indeed, all investigations are focused on the evaluation and theoretical modelling based on the multi linear regression analysis. It is notable that the metaheuristic techniques such as ACO and GA are used to optimize some difficult problems but not yet in the textile field excepting some studies using the GA. Besides, there is no sufficiently information to evaluate, predict and optimize the effect of the yarn-to-yarn friction as well as metal-to-yarn one on the residual bagging behaviour. Several and different denim fabrics within their different characteristics are investigated to widen the experimental analysis and thus to generalize the results in the experimental design of interest.

This paper proposes an integrated approach that seeks to jointly optimize project scheduling and material lot sizing decisions for time-constrained project scheduling problems.

A mixed integer linear programming model is devised, which utilizes the splitting of noncritical activities as a mean toward leveling the renewable resources. The developed model minimizes renewable resources leveling costs along with consumable resources related costs, and it is solved using IBM ILOG CPLEX optimization package. A hybrid metaheuristic procedure is also proposed to efficiently solve the model for larger projects with complex networks structure.

The results confirmed the significance of the integrated approach as both the project schedule and the material ordering policy turned out to be different once compared to the sequential approach under same parameter settings. Furthermore, the integrated approach resulted in substantial total costs reduction for low values of the acquiring and releasing costs of the renewable resources. Computational experiments conducted over 240 test instances of various sizes, and complexities illustrate the efficiency of the proposed metaheuristic approach as it yields solutions that are on average 1.14% away from the optimal ones.

This work highlights the necessity of having project managers address project scheduling and materials lot sizing decisions concurrently, rather than sequentially, to better level resources and minimize materials related costs. Significant cost savings were generated through the developed model despite the use of a small-scale example which illustrates the great potential that the integrated approach has in real life projects. For real life projects with complex network topology, practitioners are advised to make use of the developed metaheuristic procedure due to its superior time efficiency as compared to exact solution methods.

The sequential approach, wherein a project schedule is established first followed by allocating the needed resources, is proven to yield a nonoptimized project schedule and materials ordering policy, leading to an increase in the project's total cost. The integrated approach proposed hereafter optimizes both decisions at once ensuring the timely completion of the project at the least possible cost. The proposed metaheuristic approach provides a viable alternative to exact solution methods especially for larger projects.