Multi-objective Ant Colony Optimization Research Articles

A multi-objective ACO for operating room scheduling optimization

Operating room (OR) scheduling problem is commonly recognized as a multi-objective combinatorial optimization problem with several objectives from different perspectives, e.g. minimizing waiting time from patients’ perspective, reducing overtime from medical staffs’ perspective, increasing resource utilization from OR management’s perspective etc. Those objectives are often conflicting. A meta-heuristic approach integrating Pareto sets and Ant Colony Optimization (ACO) is proposed to solve such multi-objective OR scheduling optimization problem. The Pareto sets construction and the modified ant graph model is introduced and two types of pheromone setting and updating strategies are compared to determine a more efficient multi-objective OR scheduling algorithm. The scheduling results by four different approaches, i.e. the simulation, the ACO with single objective of makespan (ACO-SO), the ACO with multi-objective by weighted sum method (ACO-weighted-sum), and the hybrid Pareto set-ACO with multi-objectives (PSACO-MO) are compared. The test case in the literature, which is from MD Anderson Cancer Center, is also used to evaluate the performance of the proposed approach. The computational results show that the PSACO-MO achieves good results in shortening makespan, reducing nurses’ overtime and balancing resources’ utilization in general.

A hybrid MACO and BFOA algorithm for power loss minimization and total cost reduction in distribution systems

This paper presents a multiobjective optimization methodology to optimally place a STATCOM in electric power distribution networks. The combination of multiobjective ant colony optimization (MACO) and the bacterial foraging optimization algorithm (BFOA) is proposed to minimize the power loss and total cost. The main intention of this analysis is to optimally place the STATCOM at multiple locations such as the transmission side, middle, and load side. Identifying the type and location of the STATCOM is a combinatorial optimization problem in power systems. In order to overcome this problem, the combination of hybrid MACO and BFOA algorithms is applied in this analysis to minimize the total cost and power loss. The total cost of the overall network is calculated by using system average interruption duration index and[ system average interruption frequency index metrics. Moreover, the BFOA is used in this paper to minimize the power loss during distribution, which is adequate in searching for the optimal solution. The problem of reducing power losses in distribution systems through the BFOA approach is described here for a 5-bus system. Test results of a 5-bus network show that the proposed MACO with BFOA method can efficiently ensure the power loss and total cost minimization.

Dimension reduction for microarray data using multi-objective ant colony optimisation

DNA microarray technology is a nascent technology having vast applications in the molecular biology. Despite of its many useful applications in disease diagnosis and drug discovery, analysing DNA microarray data has become a challenge for bio-analysts. Microarrays present curse of dimensionality problem and call for the development of new techniques to handle it. The problem of dimensionality reduction or feature/gene selection has been posed as a multi-objective problem in the literature and can be better solved using multi-objective meta-heuristics. In this paper, we have proposed a multi-objective ant colony optimisation (MOACO) algorithm for gene selection. The contribution of this paper is to obtain multiple non-dominated solutions instead of a single best solution. These multiple solutions enable a user to select a solution according to his/her preference or application domain. The predicted genes also called as bio-markers, can help in disease diagnosis and may direct the progress of drug efficacy. A comparative analysis of the proposed method has been done using some conventional feature selection techniques and a method based on multi-objective particle swarm optimisation (MOPSO) proposed by Mukhopadhyay and Mandal (2014). The results obtained confirm the superiority of MOACO-based approach over the others in terms of various performance metrics.

Dimension reduction for microarray data using multi-objective ant colony optimisation

DNA microarray technology is a nascent technology having vast applications in the molecular biology. Despite of its many useful applications in disease diagnosis and drug discovery, analysing DNA microarray data has become a challenge for bio-analysts. Microarrays present curse of dimensionality problem and call for the development of new techniques to handle it. The problem of dimensionality reduction or feature/gene selection has been posed as a multi-objective problem in the literature and can be better solved using multi-objective meta-heuristics. In this paper, we have proposed a multi-objective ant colony optimisation (MOACO) algorithm for gene selection. The contribution of this paper is to obtain multiple non-dominated solutions instead of a single best solution. These multiple solutions enable a user to select a solution according to his/her preference or application domain. The predicted genes also called as bio-markers, can help in disease diagnosis and may direct the progress of drug efficacy. A comparative analysis of the proposed method has been done using some conventional feature selection techniques and a method based on multi-objective particle swarm optimisation (MOPSO) proposed by Mukhopadhyay and Mandal (2014). The results obtained confirm the superiority of MOACO-based approach over the others in terms of various performance metrics.

Multi-objective Virtual Machine Placement for Load Balancing

The virtual machine placement is closely related to the efficient and balanced utilization of physical resources. In this paper, the influence of two scenarios about resource utilization on load balancing is analyzed. A multi-objective ant colony optimization algorithm is proposed to solve the virtual machine placement problem, which balances the load among physical machines and the internal load of physical machine simultaneously. The proposed algorithm is compared with two single objective ant colony optimization algorithms, first fit algorithm and greedy algorithm under some instances. The results show that the proposed algorithm can search and find solutions that exhibit good balance among objectives while others cannot. This demonstrates the proposed algorithm can balance the load in the process of mapping virtual machines to physical machines.

Random weight-based ant colony optimisation algorithm for the multi-objective optimisation problems

Over the years, ant colony optimisation (ACO) algorithms have been proposed particularly for solving the hard combinatorial optimisation problems, such as the travelling salesman problem (TSP) and the job-shop scheduling problem (JSSP). Also, most real-world applications are concerned with the multi-objective optimisation problems. In this paper a new ant colony optimisation (ACO) algorithm is proposed for solving two or more objective functions, simultaneously. It is based on the ant colony system (ACS) algorithm and uses the random weight-based method. It is applied on several benchmark instances of the TSP and the JSSP from the literature and compared with more recent multi-objective ant colony optimisation algorithms (MOACO). The experimental results have shown that the proposed algorithm achieves better performance for solving the travelling salesman problem and the job-shop scheduling problem with multiple objectives. It also obtained well distribution all over the Pareto-optimal front.

A Method for Entity Resolution in High Dimensional Data Using Ensemble Classifiers

In order to improve utilization rate of high dimensional data features, an ensemble learning method based on feature selection for entity resolution is developed. Entity resolution is regarded as a binary classification problem, an optimization model is designed to maximize each classifier’s classification accuracy and dissimilarity between classifiers and minimize cardinality of features. A modified multiobjective ant colony optimization algorithm is employed to solve the model for each base classifier, two pheromone matrices are set up, weighted product method is applied to aggregate values of two pheromone matrices, and feature’s Fisher discriminant rate of records’ similarity vector is calculated as heuristic information. A solution which is called complementary subset is selected from Pareto archive according to the descending order of three objectives to train the given base classifier. After training all base classifiers, their classification outputs are aggregated by max-wins voting method to obtain the ensemble classifiers’ final result. A simulation experiment is carried out on three classical datasets. The results show the effectiveness of our method, as well as a better performance compared with the other two methods.

Multi Objective Optimization of Virtual Machine Migration Placement Based on Cloud Computing

Abstract How to improve the resource utilization of the cloud computing system has been one of the key content of the research of the cloud computing. The traditional multi-objective ant colony optimization was improved, studied the virtual machine live migration framework, combined with the elimination method to solve virtual machine migration and placement of multi-objective optimization problem, the load balanced specific strategies are integrated into the framework of a dynamic migration, simulation experiments are carried out and the conclusions are made for it. The algorithm can obtain the optimal solution through the continuous updating of pheromone. The main consideration is the Service level contract violation rate(S), Resource loss(W),Power consumption (P). Experimental results show that ,compared with the traditional heuristic method and genetic algorithm, the algorithm is advantageous to the parallel computation, and it’s able to achieve the optimal tradeoff and compromise between multiple conflicting objectives. In the case of service level contract violation rate is low, system resource waste and power consumption are at the least, so it has feasibility.

A realistic variant of bi-objective unrelated parallel machine scheduling problem: NSGA-II and MOACO approaches

The problem investigated in this study involves an unrelated parallel machine scheduling problem with sequence-dependent setup times, different release dates, machine eligibility and precedence constraints. This problem has been inspired from a realistic scheduling problem in the shipyard. The optimization criteria are to simultaneously minimize mean weighted flow time and mean weighted tardiness. To formulate this complicated problem, a new mixed-integer programming model is presented. Considering the NP-complete characteristic of this problem, two famous meta-heuristics including a non-dominated sorting genetic algorithm (NSGA-II) and a multi-objective ant colony optimization (MOACO) which is a modified and adaptive version of BicriterionAnt algorithm are developed. Obviously, the precedence constraints increase the complexity of the scheduling problem in strong sense in order to generate feasible solutions, especially in parallel machine environment. Therefore a new corrective algorithm is proposed to obtain the feasibility in all stages of the algorithms. Due to the fact that appropriate design of parameter has a significant effect on the performance of algorithms, we calibrate the parameters of these algorithms by using new approach of Taguchi method. The performances of the proposed meta-heuristics are evaluated by a number of numerical examples. The results indicated that the suggested MOACO statistically outperformed the proposed NSGA-II in solving the test problems. In addition, the application of the proposed algorithms is justified by a real block erection scheduling problem in the shipyard.

A Multi-Objective Fuzzy Ant Colony Optimization Algorithm for Virtual Machine Placement

In cloud computing, the most important challenge is to enforce proper utilization of physical resources. To accomplish the mentioned challenge, the cloud providers need to take care of optimal mapping of virtual machines to a set of physical machines. In this paper, the authors address the mapping problem as a multi-objective virtual machine placement problem (VMP) and propose to apply multi-objective fuzzy ant colony optimization (F-ACO) technique for optimal placing of virtual machines in the physical servers. VMP-F-ACO is a combination of fuzzy logic and ACO, where we use fuzzy transition probability rule to simulate the behaviour of the ants and the authors apply the same for virtual machine placement problem. The results of fuzzy ACO techniques are compared with five variants of classical ACO, three bin packing heuristics and two evolutionary algorithms. The results show that the fuzzy ACO techniques are better than the other optimization and heuristic techniques considered.

Multi-Objective Ant Colony Optimization Based on the Physarum-Inspired Mathematical Model for Bi-Objective Traveling Salesman Problems.

Bi-objective Traveling Salesman Problem (bTSP) is an important field in the operations research, its solutions can be widely applied in the real world. Many researches of Multi-objective Ant Colony Optimization (MOACOs) have been proposed to solve bTSPs. However, most of MOACOs suffer premature convergence. This paper proposes an optimization strategy for MOACOs by optimizing the initialization of pheromone matrix with the prior knowledge of Physarum-inspired Mathematical Model (PMM). PMM can find the shortest route between two nodes based on the positive feedback mechanism. The optimized algorithms, named as iPM-MOACOs, can enhance the pheromone in the short paths and promote the search ability of ants. A series of experiments are conducted and experimental results show that the proposed strategy can achieve a better compromise solution than the original MOACOs for solving bTSPs.

Three metaheuristics improved by a mapping method

A mapping method (MaM) is used to guide the metaheuristic for a better exploration. The MaM is adapted to the tree well known metaheuristics: Strength Pareto Evolutionary Algorithm 2 (SPEA2), Multi Objective Ant Colony Optimization (MOACO) and Multi Objective Particle Swarm Optimization (MOPSO). The new hybridized Metaheuristics (MaM-SPEA2, MaM-MOACO, MaM-MOPSO) are applied to solve the Flexible Job Shop Problem (FJSP) with the objectives of minimizing the makespan (Cmax) and the production just in time. Then, the multi-objective metric C-Metric’ show that MaM hybridization improves the performances of the three algorithms.

Distributed query plan generation using multi-objective ant colony optimisation

In distributed relational databases, relations are fragmented and replicated at multiple disparate sites. As a result, for a distributed relational query, the number of possible query plans increases exponentially with an increase in the number of sites containing these relations. This leads to a large search space from which effective and efficient query plans are to be computed. This problem has already been addressed as a single objective optimisation problem using ant colony optimisation. In this paper, this problem is addressed as a bi-objective optimisation problem and solved using multi-objective ant colony optimisation MOACO. Accordingly, a MOACO-based distributed query plan generation DQPG algorithm is proposed herein that generates Top-K query plans for a distributed query. Experimental comparisons of the proposed MOACO-based DQPG algorithm with the existing ACO-based DQPG algorithm show that for higher numbers of relations, the former is able to generate, comparatively, cost-effective Top-K query plans.

A Performance Study for the Multi-objective Ant Colony Optimization Algorithms on the Job Shop Scheduling Problem

Most of the research on job shop scheduling problem are concerned with minimization of a single objective. However, the real world applications of job shop scheduling problems are involved in optimizing multiple objectives. Therefore, in recent years ant colony optimization algorithms have been proposed to solve job shop scheduling problems with multiple objectives. In this paper, some recent multi-objective ant colony optimization algorithms are reviewed and are applied to the job shop scheduling problem by considering two, three and four objectives. Also in this study, four criteria: makespan, mean flow time, mean tardiness and mean machine idle time are considered for simultaneous optimization. Two types of models are used by changing the number of ants in a colony and each multi-objective ant colony optimization algorithm is applied to sixteen benchmark problem instances of up to 20 jobs × 5 machines, for evaluating the performances of these algorithms. A detailed analysis is performed using the performance indicators, and the experimental results have shown that the performance of some multi-objective ant colony optimization algorithms depend on the number of objectives and the number of ants.

Performance Comparison between the Multi-Colony and Multi-Pheromone ACO Algorithms for Solving the Multi-objective Routing Problem in a Public Transportation Network

Routing in a multimodal urban public transportation network, according to the user's preferences, can be considered as a multi-objective optimisation problem. Solving this problem is a complicated task due to the different and incompatible objective functions, various modes in the network, and the large size of the network. In this research, two optimisation algorithms are considered for solving this problem. The multi-colony and multi-pheromone Ant Colony Optimisation (ACO) algorithms are two different modes of the Multi-Objective ACO (MOACO) algorithm. Moreover, according to the acquired information, the algorithms implemented in the public transportation network of Tehran consist of four modes. In addition, three objective functions have been simultaneously considered as the problem's objectives. The algorithms are run with different initial parameters and afterwards, the results are compared and evaluated based on the different obtained routes and with the aid of the convergence and repeatability tests, diversity and convergence metrics.

Generalization of the MOACS algorithm for Many Objectives. An application to motorcycle distribution

To solve many-objective routing problems, this paper generalizes the Multi-Objective Ant Colony System (MOACS) algorithm, a well-known Multi-Objective Ant Colony Optimization (MOACO) metaheuristic proposed in 2003. This Generalized MOACS algorithm is used to solve a Split-Delivery/Mixed-Fleet Vehicle Routing Problem (SD/MF-VRP) under different constraints, resulting from the mathematical modeling of a logistic problem: the distribution of motorcycles by a Paraguayan factory, considering several objective functions as: (1) total distribution cost, (2) total traveled distance, (3) total traveled time, and (4) nsatisfied demand. Experimental results using the proposed algorithm in weekly operations of the motorcycle factory prove the advantages of using the proposed algorithm, facilitating the work of the logistic planner, reducing the distribution cost and minimizing the time needed to satisfy customers.

Incorporating Bidirectional Heuristic Search and Improved ACO in Route Planning

A multi-objective multi-node dynamic route planning system for a vehicle is presented in this paper. In this system, a bidirectional heuristic search algorithm is designed to perform path planning between two nodes in a topological map created by OpenStreetMap for urban scenarios. And then an improved multi-objective Ant Colony Optimization (ACO) algorithm considering the timeliness of goal nodes is proposed to generate the node sequence. Experimental results validated the proposed approach.

Multi-objective Topology Optimization with Ant Colony Optimization and Genetic Algorithms

In this work, we present a multi-objective approach for Topology Optimization applied to the design of devices with several materials. The first stage consists of applying a Multi-Objective Ant Colony Optimization (MOACO) to find tradeoff topologies with different material distributions. In the second stage, we parameterize the boundaries of the topologies found by using NURBS. A Multi-objective Genetic Algorithm is applied as a heuristic optimization engine to optimize the control points, weights and knots of the curves in order to smooth and refine the boundaries of the topology. The main advantage of a multi-objective approach is that the designer can identify, explore and refine a number of tradeoff topologies. The proposed methodology is illustrated in the design of a C-core magnetic actuator.

Performance analysis of the multi-objective ant colony optimization algorithms for the traveling salesman problem

Most real world combinatorial optimization problems are difficult to solve with multiple objectives which have to be optimized simultaneously. Over the last few years, researches have been proposed several ant colony optimization algorithms to solve multiple objectives. The aim of this paper is to review the recently proposed multi-objective ant colony optimization (MOACO) algorithms and compare their performances on two, three and four objectives with different numbers of ants and numbers of iterations. Moreover, a detailed analysis is performed for these MOACO algorithms by applying them on several multi-objective benchmark instances of the traveling salesman problem. The results of the analysis have shown that most of the considered MOACO algorithms obtained better performances for more than two objectives and their performance depends slightly on the number of objectives, number of iterations and number of ants used.

Optimum reliable operation of water distribution networks by minimising energy cost and chlorine dosage

In recent decades much attention has been paid to optimal operation of water distribution networks (WDNs). In this regard, the system operation costs, including energy and disinfection chemicals, as well as system reliability should be simultaneously considered in system performance optimisation, to provide the minimum required level of performance in failure condition and to manage economic limitations. In this study, multi-objective optimisation of water distribution network performance in 3 different scenarios was considered. In these scenarios the effects of time-dependent chlorine injection and pump speed, as well as different combinations of objective functions for minimising energy and disinfection costs and for maximising hydraulic reliability and quality-based reliability are incorporated. As the optimisation method, a multi-objective ant colony optimisation (ACO) algorithm was used because of its high efficiency. For better managing the hydraulic behaviour and water quality in the WDN, considering temporal variations of demand, it is suggested to use variable speed pumps (VSP) as well as to inject chlorine at a variable rate. Application of VSP and time-dependent chlorine injection results in improvements such as reduction in energy and disinfection costs, and decrease in disinfection costs in application of HDSM (head-driven simulation method). In HDSM simulation of WDN, a decrease in hydraulic reliability because of shortages in water supply can be mitigated through extra chlorine injection and increase in quality-based reliability. To deal with this challenge, it is recommended to satisfy the hydraulic reliability first and then to evaluate the quality reliability. Furthermore it is necessary to modify the hydraulic reliability relationship to incorporate different components of the WDN other than pumps. This will provide more reliable results for evaluation of the system performance.Keywords: multi-objective optimisation, aco algorithm, energy cost, disinfection costs, hydraulic reliability, quality reliability