Hybridization Of Ant Colony Optimization Research Articles

Optimizing Deep Learning Model for Software Cost Estimation Using Hybrid Meta-Heuristic Algorithmic Approach.

Effective software cost estimation significantly contributes to decision-making. The rising trend of using nature-inspired meta-heuristic algorithms has been seen in software cost estimation problems. The constructive cost model (COCOMO) method is a well-known regression-based algorithmic technique for estimating software costs. The limitation of the COCOMO models is that the values of these coefficients are constant for similar kinds of projects whereas, in reality, these parameters vary from one organization to another organization. Therefore, for accurate estimation, it is necessary to fine-tune the coefficients. The research community is now examining deep learning (DL) as a forward-looking solution to improve cost estimation. Although deep learning architectures provide some improvements over existing flat technologies, they also have some shortcomings, such as large training delays, over-fitting, and under-fitting. Deep learning models usually require fine-tuning to a large number of parameters. The meta-heuristic algorithm supports finding a good optimal solution at a reasonable computational cost. Additionally, heuristic approaches allow for the location of an optimum solution. So, it can be used with deep neural networks to minimize training delays. The hybrid of ant colony optimization with BAT (HACO-BA) algorithm is a hybrid optimization technique that combines the most common global optimum search technique for ant colonies (ACO) in association with one of the newest search techniques called the BAT algorithm (BA). This technology supports the solution of multivariable problems and has been applied to the optimization of a large number of engineering problems. This work will perform a two-fold assessment of algorithms: (i) comparing the efficacy of ACO, BA, and HACO-BA in optimizing COCOMO II coefficients; and (ii) using HACO-BA algorithms to optimize and improve the deep learning training process. The experimental results show that the hybrid HACO-BA performs better as compared to ACO and BA for tuning COCOMO II. HACO-BA also performs better in the optimization of DNN in terms of execution time and accuracy. The process is executed upto 100 epochs, and the accuracy achieved by the proposed DNN approach is almost 98% while NN achieved accuracy of up to 85% on the same datasets.

Solving Mixed-Model Assembly Lines Using a Hybrid of Ant Colony Optimization and Greedy Algorithm

The assembly line balancing problem deals with the assignment of tasks to work stations. Mixed-model assembly line problem is a type of assembly line balancing problem at which two or more models of the same product are assembled sequentially at the same line. To achieve optimality and efficiency of solving this problem, tasks at each work station have to be well balanced satisfying all constraints. This paper deals with the mixed-model assembly line balancing problem (MALBP) in which the objective is to minimize the cycle time for a given number of work stations. The problem is solved by using a hybrid of an ant colony optimization and a greedy algorithm (Ant-Greedy). MATLAB Software is used to perform the proposed method. Then, the proposed method is applied to a real case problem found in the literature for the assembly line of automatic changeover in the Electronic Industries Company in Iraq. The results of the proposed method are compared with the performance of the Merging Shortest and Longest Operation (MMSLO) method. The comparison shows that the Ant-Greedy optimization method is more efficient, where the efficiency increased from 93.53% for MMSLO method to 97.26% for the Ant-Greedy method.

Comparisons of VRP Optimization Algorithmic Methods for the Optimal Routing of Multiple Delivery Vehicles with Time Constraint

Transportation costs account for a large portion of business expense in any logistics firm; thus, achieving proper solutions that manage those transportation activities well and reduce such expense should be the number one priority for the business. Essentially, such logistics management involves the routing plans for company vehicles that perform delivery/pick up and also the number of vehicles utilized. This study investigated and compared the optimization performances of routing algorithms using simulated geographic data based in Chiang Rai, Thailand, emulating the post office operation which had 1 post office, 4 delivery vehicles and 2 delivery zones (2 vehicles per zone): 65 customer locations for zone A and 74 for zone B. The major objective of this particular routing problem, called Vehicle Routing Problem (VRP), was that the total delivery distance of those 4 delivery vehicles combined should be minimized; moreover, those vehicles mush finish their delivery operation within a time constraint, set at 2 hours. The optimization algorithms, employed for the routing procedures, were Large Neighborhood Search (LNS), Particle Swarm Optimization (PSO), Differential Evolution (DE) and Ant Colony Optimization (ACO), combined into 3 hybrid algorithms (LNS|PSO, LNS|DE and LNS|ACO). Those hybrid algorithms including pure ones (DE, PSO and ACO) were compared for their optimization performances; and the results showed that LNS|ACO hybrid algorithm was superior than the other two hybrid ones and also far better than pure DE, PSO and ACO algorithms at various parameter variants. Additionally, most algorithms (hybrid and pure ones) finished the delivery routing within the time constraint of 2 hours except only for the pure algorithms at minimum parameter variant.

A Hybrid Hierarchical Heuristic-ACO With Local Search Applied to Travelling Salesman Problem, AS-FA-Ls

The combinatorial optimization problem is attracting research because they have a wide variety of applications ranging from route planning and supply chain optimization to industrial scheduling and the IoT. Solving such problems using heuristics and bio-inspired techniques is an alternative to exact solutions offering acceptable solutions at fair computational costs. In this article, a new hierarchical hybrid method is proposed as a hybridization of Ant Colony Optimization (ACO), Firefly Algorithm (FA), and local search (AS-FA-Ls). The proposed methods are compared to similar techniques on the traveling salesman problem, (TSP). ACO is used in a hierarchical collaboration schema together with FA which is used to adapt ACO parameters. A local search strategy is used which is the 2 option method to avoid suboptimal solutions. A comparative review and experimental investigations are conducted using the TSP benchmarks. The results showed that AS-FA-Ls returned better results than the listed works in the following cases: berlin52, st70, eil76, rat99, kroA100, and kroA200. Computational investigations allowed determining a set of recommended parameters to be used with ACO for the TSP instances of the study.

Differential evolution and ACO based global optimal feature selection with fuzzy rough set for cancer data classification

In the biomedical research field, feature selection plays the predominant role in prediction of diseases. The main objective of this paper is to predict cancer from microarray gene expression data by proposing two feature selection algorithms, namely (1) differential evolution with fuzzy rough set feature selection and (2) ant colony optimization with fuzzy rough set feature selection algorithms, which solve the multi-objective optimization problems. The first algorithm represents the hybridization of differential evolution and fuzzy rough set and aims to select the global optimal features by applying the fuzzy rough evaluation function as the fitness function. The second algorithm, i.e., hybridization of ant colony optimization and fuzzy rough set, selects global optimal features by applying the fuzzy rough evaluation function as the fitness function. The performance of proposed two features selection algorithms is evaluated with various classification metrics, which are computed from decision tree classifier using tenfold cross-validation. Five datasets are applied to analyze the performance of the feature selection algorithms. The datasets used are diffuse large B cell lymphoma, breast cancer, Leukemia and small round blue-cell tumors which are cancer datasets. In addition, a non-medical dataset, namely Gisette, is also used to demonstrate the generalization capability of the proposed algorithms. The metrics used for comparison are, namely, accuracy, precision, recall, f-measure, specificity, processing time and receiver operating characteristics. The performance comparison evidenced improved performances for the proposed algorithms. Similar to the hybridization of differential evolution and ant colony optimization with fuzzy rough set, particle swarm optimization can be extended in the future.

Optimized Hybridization of Ant Colony Optimization and Genetic Algorithm (HACOGA) Based Interpretable Intuitive and Correlated-Contours Fuzzy Neural Network Classifier for Abalone

Classification generally assigns objects to enormous predefined categories and it is pervasive crisis that covers various application. Preparing the data for Classification and Prediction is the major problem in classification. In order to rectify this issue, an approximate function is proposed using Interpretable intuitive and Correlated-contours Fuzzy Neural Network (IC-FNN). For acquiring cor- related fuzzy rules and non-separable rules that comes under proper optimization problem. The extracted fuzzy rule’s parameter was fine-tuned sourced on hierarchical Levenberg Marquardt (LM) learning method for enhancing performance. But here parameters of fuzzy rules aren’t tuned per- fectly. Hybridization of Ant Colony Optimization Genetic Algorithm (HACOGA) is proposed here to rectify these issues. It tunes the parameters of the extracted fuzzy rules. Hybridization is enforced to certain factors and ACO and GA variables that share same characteristics in the computation. Experimental results shows that proposed HACOGA assist in enhancing the performance of FNN with recall, precision, accuracy and F -measure for the Abalone age prediction dataset.

Bug Report Summarization by Using Swarm Intelligence Approaches

Background: Bug reports are considered as a reference document, during the maintenance phase of the software development process. The developer's counsel them at whatever point they have to know about the reported bug and need to explore past bug solution. This process requires a sustainable amount of time due to a large number of comments. Therefore, the best solution to prevent the developers from reading the whole bug report is to summarize the entire discussion in a couple of sentences. Bug report summarization is the extraction of some important part of bug reports that are useful for investigation, resolving the bug with similar problems, reproducing the bug and checking the status of the bug. If the bug reports have huge volume, variety and velocity information as big data, extractive bug report summarization would be emerging as an issue. However, the examiners, find out the bug report summaries do not meet the expectation of the developer; there is a still need for reading the entire discussion. Objective: (1) To generate generalized unsupervised extractive bug report summarization system, which is easily applicable on any dataset without the need of effort and cost of manually creating summaries for training dataset (2) To handle the extensive number of comments and generate short summaries. (3) To reduce the data sparsity, reduction of information, redundancy and convergence issue for short and lengthy data set. (4) To achieve the semantic summarization solution and explore the large search space. (5) To provide the facility of adjusting the summary percentage. Methods: Particle swarm optimization and Hybridization of Ant Colony Optimization and Particle Swarm Optimization approaches are used with the advantage of feature weighting technique. Conclusion: The efficiency of the proposed approaches are compared with the best existing supervised and unsupervised approaches. The result shows that the Hybrid swarm intelligence approach provided better.

Improved priority-based data aggregation congestion control protocol

With advancement in multimedia applications, Wireless sensor networks (WSNs) are becoming popular due to their inherent characteristics and wide range of applications. However, WSNs contain very small sensor nodes, these nodes are battery constrained. Also, the batteries of these sensor nodes are not either replaceable or rechargeable. Therefore, many energy efficient protocols have been implemented so far to improve the network lifetime. However, the data aggregation at sink may suffer from data flooding issue, which reduces the network lifetime of WSNs. For handling this issue, in this paper, an effective data aggregation approach is designed. We have designed a priority-based data aggregation control protocol, which considers token bucket, Lempel–Ziv–Welch (LZW) compression and a hybrid of ant colony optimization and particle swarm optimization-based soft computing approach. Extensive experiments reveal that the presented protocol provides better network lifetime in contrast to the existing energy efficient protocols.

A comprehensive review on nature inspired computing algorithms for the diagnosis of chronic disorders in human beings

Diabetes and cancer are two major life-threatening human chronic disorders that have a high rate of disability and mortality. These diseases have been diagnosed using different deterministic and nature inspired computing algorithms. Here, an effort is made to represent the role of five different insect-based nature inspired computing algorithms [ant colony optimization (ACO), artificial bee colony (ABC), glow-worm swarm optimization (GSO), firefly algorithm (FA) and antlion optimization (ALO)] used for the diagnosis of these two chronic disorders. Initially, the basic statistics of diabetes and cancer patients have been presented. The main intention of this study lies in exploring the usage and performance of ACO, ABC, GSO, FA and ALO in diagnosing different stage and types of diabetes and cancer. It has been revealed that most of the diabetes diagnosis work has been carried out using ACO followed by ABC. As far as cancer is concerned, the three insect-based algorithms, i.e. ACO, ABC and FA, have been also effectively employed for detection of breast, lung, liver, prostate and ovarian cancer. In general, most of the disease diagnostic work has been carried out using ACO, whereas GSO found to be least explored. The rate of predictive accuracy achieved using the hybridization of ACO and neural network is found to be more promising as compared to other individual or hybrid approaches. Likewise, for breast cancer, the amalgamated use of ABC and neural network is more productive. Similarly, the hybrid approach of ACO and neural network is also found useful for early prognosis of lung and gastric cancer. In general, the diagnostic results obtained using hybrid approaches are more promising than their individual use. However, several hybrid combinations are still needed to be explored for the diagnosis of diabetes and different types of cancer, viz. liver, gastric, ovarian, leukaemia as well as a brain tumour. Finally, there is also a scope to use and explore the efficiency of binary and chaotic variants of ACO, ABC, GSO, FA and ALO for the diagnosis of these two critical human disorders.

A HYBRID ANT COLONY OPTIMIZATION ALGORITHM FOR SOLVING A HIGHLY CONSTRAINED NURSE ROSTERING PROBLEM

Distribution of work shifts and off days to nurses in a duty roster is a crucial task. In hospital wards, much effort is spent trying to produce workable and quality rosters for their nurses. However, there are cases, such as mandatory working days per week and balanced distribution of shift types that could not be achieved in the manually generated rosters, which are still being practiced. Hence, this study focused on solving those issues arising in nurse rostering problems (NRPs) strategizing on a hybrid of Ant Colony Optimization (ACO) algorithm with a hill climbing technique. The hybridization with the hill climbing is aiming at fine-tuning the initial solution or roster generated by the ACO algorithm to achieve better rosters. The hybrid model is developed with the goal of satisfying the hard constraints, while minimizing the violation of soft constraints in such a way that fulfill hospital’s rules and nurses’ preferences. The real data used for this highly constrained NRPs was obtained from a large Malaysian hospital. Specifically, three main phases were involved in developing the hybrid model, which are generating an initial roster, updating the roster through the ACO algorithm, and implementing the hill climbing to further search for a refined solution. The results show that at a larger value of pheromone, the chance of obtaining a good solution was found with only small penalty values. This study has proven that the hybrid ACO is able to solve NRPs with good potential solutions that fulfilled all the four important criteria, which are coverage, quality, flexibility, and cost. Subsequently, the hybridmodel is also beneficial to the hospital’s management whereby nurses can be scheduled with balanced distribution of shifts, which fulfil their preferences as well.

Two-echelon logistics delivery and pickup network optimization based on integrated cooperation and transportation fleet sharing

The optimization of the two-echelon logistics delivery and pickup network (2E-LDPN) is a strategical and tactical task which can efficiently be achieved by establishing cooperative alliances. Under the coordination of logistics services providers or logistics facilities of the existing network, high operating costs caused by cross and long distance transportation can be reduced via the inclusive reorganization of the entire network. In order to minimize the total cost, this study simultaneously considers semitrailer truck and vehicle sharing, and establishes a linear mathematical model capable of interpreting real world practices under single or multiple alliances scenarios. An Improved Particle Swarm Optimization (IPSO) algorithm and the Ant Colony Optimization (ACO) algorithm are reasonably combined into a hybrid meta-heuristics to solve the cooperative 2E-LDPN optimization problem. This algorithm combines the merits of IPSO and ACO with local and global search capabilities, and redistributes customer zones on the basis of region partitioning solutions in order to rationalize transportation activities. Finally, an Improved Shapley value model is applied to guarantee profits allocation's fairness and is proved reliable in term of alliance stability. Empirical results out of a case study in Chongqing city show that the IPSO–ACO hybrid algorithm is superior to three well-known algorithms on the cost solution and the number of iterations. Using the Improved Shapley value model and strictly monotonic path (SMP) selection principles, optimal adhesion sequences for two alliances and a grand alliance are yielded. The implemented transition from two sub-alliances based network to the grand alliance is in line with real-world's practices and provides decision makers with a useful tool for the design of cooperative alliances. In addition, the proposed cooperation strategy and profit allocation method enable companies to increase cost savings and the logistics network's efficiency. Besides, semitrailer truck and vehicle sharing as feature of collaboration conditional clauses reduces the size of transportation fleets, and promotes greener logistics operations.

Hybridization of Ant Colony Optimization and Decision Tree for Test Paths Generation and Optimization

Hybridization of Ant Colony Optimization and Decision Tree for Test Paths Generation and Optimization

A new hybrid ant colony algorithm for scheduling of no-wait flowshop

In this paper, the no-wait flow shop scheduling problem under makespan and flowtime criteria is addressed. The no-wait flowshop is a variant of the well-known flowshop scheduling problem where all processes follow the previous one without any interruption for operations of a job. Owing to the problem is known to be NP-hard for more than two machines, a hybrid meta-heuristic algorithm based on ant colony optimization (ACO) and simulated annealing (SA) algorithm is improved. First, at each step, due to the characteristic of ACO algorithm that include solution construction and pheromone trail updating, some different areas of search space are checked and best solution is selected. Then, to enhance the quality and diversity of the solution and finding best neighbor of this solution, a novel SA is presented. Moreover, a new principle is applied for global pheromone update based on the probability function like SA algorithm. The proposed approach solution is compared with several the state-of-the-art algorithms in the literature. The reported results show that the proposed algorithms are effective and the new approach for local search in ACO algorithm is efficient for solving the no-wait flow shop problem. Then, we employed another hybrid ACO algorithm based on hybridization of ACO with variable neighborhood search (VNS) and compare the results given by two proposed algorithms. These results show that our new hybrid provides better results than ACO-VNS algorithm.

A low-complexity hybrid algorithm based on particle swarm and ant colony optimization for large-MIMO detection

With rapid increase in demand for higher data rates, multiple-input multiple-output (MIMO) wireless communication systems are getting increased research attention because of their high capacity achieving capability. However, the practical implementation of MIMO systems rely on the computational complexity incurred in detection of the transmitted information symbols. The minimum bit error rate performance (BER) can be achieved by using maximum likelihood (ML) search based detection, but it is computationally impractical when number of transmit antennas increases. In this paper, we present a low-complexity hybrid algorithm (HA) to solve the symbol vector detection problem in large-MIMO systems. The proposed algorithm is inspired from the two well known bio-inspired optimization algorithms namely, particle swarm optimization (PSO) algorithm and ant colony optimization (ACO) algorithm. In the proposed algorithm, we devise a new probabilistic search approach which combines the distance based search of ants in ACO algorithm and the velocity based search of particles in PSO algorithm. The motivation behind using the hybrid of ACO and PSO is to avoid premature convergence to a local solution and to improve the convergence rate. Simulation results show that the proposed algorithm outperforms the popular minimum mean squared error (MMSE) algorithm and the existing ACO algorithms in terms of BER performance while achieve a near ML performance which makes the algorithm suitable for reliable detection in large-MIMO systems. Furthermore, a faster convergence to achieve a target BER is observed which results in reduction in computational efforts.

A hybrid of ant colony optimization and artificial bee colony algorithm for probabilistic optimal placement and sizing of distributed energy resources

In this paper, a hybrid configuration of ant colony optimization (ACO) with artificial bee colony (ABC) algorithm called hybrid ACO–ABC algorithm is presented for optimal location and sizing of distributed energy resources (DERs) (i.e., gas turbine, fuel cell, and wind energy) on distribution systems. The proposed algorithm is a combined strategy based on the discrete (location optimization) and continuous (size optimization) structures to achieve advantages of the global and local search ability of ABC and ACO algorithms, respectively. Also, in the proposed algorithm, a multi-objective ABC is used to produce a set of non-dominated solutions which store in the external archive. The objectives consist of minimizing power losses, total emissions produced by substation and resources, total electrical energy cost, and improving the voltage stability. In order to investigate the impact of the uncertainty in the output of the wind energy and load demands, a probabilistic load flow is necessary. In this study, an efficient point estimate method (PEM) is employed to solve the optimization problem in a stochastic environment. The proposed algorithm is tested on the IEEE 33- and 69-bus distribution systems. The results demonstrate the potential and effectiveness of the proposed algorithm in comparison with those of other evolutionary optimization methods.

A Hybrid of Ant Colony Optimization and Chaos Optimization Algorithms Approach for Software Cost Estimation

The main challenge in the production and development of large and complex software projects is the cost estimation with high precision. Thus it can be said that estimating the cost of software projects play an important role in the organization productivity. With the increasing size and complexity of software projects the demand to offer new techniques to accomplish this important task increases day by day. Therefore, researchers have long attempted to provide models to fulfill this important task. The most documented algorithmic model is the Constructive Cost Model (COCOMO), which was introduced in 1981 by Barry W. Boehm. But due to the lack of values for the constant parameters in this model, it cannot meet the high precision for all software projects. Nowadays, regarding the increasing researches on machine learning algorithms and the success of these studies, in this paper, we have tried to estimate the cost of software projects according to meta-heuristic algorithms. In this paper, Ant Colony Optimization (ACO) and Lorentz transformation have been used as Chaos Optimization Algorithm (COA) and NASA datasets as training and testing sets. To compare and evaluate the results of the proposed method with COCOMO model, MARE is used, and the results show a decline in MARE to 0.078%.

An Effective Hybrid Ant Colony Optimization for Permutation Flow-Shop Scheduling

This paper proposes a hybrid ant colony optimization algorithm to solve the permutation flow-shop scheduling (PFS). The hybridization of ant colony optimization (ACO) with path relinking (PR), which combines the advantages of two individual algorithms, is the key innovative aspect of the approach. Path relinking (PR) can be interpreted as an evo- lutionary method where the high quality solutions are generated by introducing features of the guiding solution gradually into the initial solution. Moreover, the effective hybrid algorithm is a method to integrate intensification and diversifica- tion in the search, and it adopts the criterion function restricting the frequencies of using the PR procedure to improve the convergence speed. Finally, the proposed algorithm is applied to PFS benchmark problems. The experimental results have shown that the hybrid method yields better results to solve the permutation flow-shop scheduling than well-known exist- ing methods in terms of solution quality.

A hybrid of ant colony optimization and minimization of metabolic adjustment to improve the production of succinic acid in Escherichia coli

This paper presents a study on gene knockout strategies to identify candidate genes to be knocked out for improving the production of succinic acid in Escherichia coli. Succinic acid is widely used as a precursor for many chemicals, for example production of antibiotics, therapeutic proteins and food. However, the chemical syntheses of succinic acid using the traditional methods usually result in the production that is far below their theoretical maximums. In silico gene knockout strategies are commonly implemented to delete the gene in E. coli to overcome this problem. In this paper, a hybrid of Ant Colony Optimization (ACO) and Minimization of Metabolic Adjustment (MoMA) is proposed to identify gene knockout strategies to improve the production of succinic acid in E. coli. As a result, the hybrid algorithm generated a list of knockout genes, succinic acid production rate and growth rate for E. coli after gene knockout. The results of the hybrid algorithm were compared with the previous methods, OptKnock and MOMAKnock. It was found that the hybrid algorithm performed better than OptKnock and MOMAKnock in terms of the production rate. The information from the results produced from the hybrid algorithm can be used in wet laboratory experiments to increase the production of succinic acid in E. coli.

Ant-cuckoo colony optimization for feature selection in digital mammogram.

Digital mammogram is the only effective screening method to detect the breast cancer. Gray Level Co-occurrence Matrix (GLCM) textural features are extracted from the mammogram. All the features are not essential to detect the mammogram. Therefore identifying the relevant feature is the aim of this work. Feature selection improves the classification rate and accuracy of any classifier. In this study, a new hybrid metaheuristic named Ant-Cuckoo Colony Optimization a hybrid of Ant Colony Optimization (ACO) and Cuckoo Search (CS) is proposed for feature selection in Digital Mammogram. ACO is a good metaheuristic optimization technique but the drawback of this algorithm is that the ant will walk through the path where the pheromone density is high which makes the whole process slow hence CS is employed to carry out the local search of ACO. Support Vector Machine (SVM) classifier with Radial Basis Kernal Function (RBF) is done along with the ACO to classify the normal mammogram from the abnormal mammogram. Experiments are conducted in miniMIAS database. The performance of the new hybrid algorithm is compared with the ACO and PSO algorithm. The results show that the hybrid Ant-Cuckoo Colony Optimization algorithm is more accurate than the other techniques.

A New Feature Selection Method Based on Ant Colony and Genetic Algorithm on Persian Font Recognition

Dimensionality reduction of a feature set is a usual pre-processing step used for classification applications to improve their accuracy with a small and appropriate feature subset. In this article, a new hybrid of ant colony optimization (ACO) and genetic algorithm (GAs) as a feature selector, and support vector machine as a classifier are integrated effectively. Based on the combination of the fast global search ability of GA and the positive feedback mechanism of ACO, a novel algorithm was proposed in the domain of feature selection. Experiments show that the proposed feature selection can achieve better performance than that the normal GA and ACO does. We tested this method on the extracted features of ten common Persian fonts. The result shows it has affected slightly better on the performance. Furthermore, number of features decreased to almost half of the original feature number after this pre-processing step.