1. Introduction
Optimization of the transport system problems is cur- rently required by science community and also by trans- port and telecommunication companies. There is a ne- cessity of achieving high quality solutions of combinato- rial problems in a very short time. The goal of research in this field is to find effective method to find solution of supply problems. Looking for optimal routes of larger problems with many customers is too expensive and there- fore using of heuristic method is the only option. There exists several heuristic methods which can be used. One of them is called antcolonyoptimization (ACO). This modern method has been inspired by behaviour of ants and was discovered in late nineties. This method is used by scientists and there are many publications which dis- cuss it. On its base there were developed many applica- tions. This method was successfully used to solve NP- complete problems, where to get solution requires expo- nential time. These problems include: travelling salesman problem, quadratic assignment problem, solving of many transport problems, schedule problems, telecommunica- tion problems, protein folding problems and so on. Mod- ern computing technologies bring new opportunities and thesis for research. Currently available multi-core archi- tecture allows running parallel computations also on per- sonal computers. It makes sense to design new optimiza- tion methods and investigate existing parallel algorithms for this new computer architectures. In our work we have focused on possibilities of effective parallel implementa- tion of the antcolonyoptimization method which is ap- plied on vehicle routing problem. Antcolonyoptimization and its variations are suitable for using in parallel com- puting environment including multi-core architectures.
moedjiono@budiluhur.ac.id 3
denni.kurniawan@budiluhur.ac.id
Abstraksi — Metode Optimasi Abstrak lebih dikenal salah satunya adalah metode pendekatan. Metode pendekatan tidak menjamin solusi optimal. Namun, meskipun metode ini tidak menjamin hasil yang optimal, tetapi hasil penyelesaian umumnya cukup baik. Metaheuristik sebenarnya adalah pendekatan yang didasarkan pada metode heuristik. Jadi tidak mengherankan bahwa metode heuristik sering terintegrasi dalam metode metaheuristik. AntColonyOptimization ( ACO ) adalah salah satu contoh dari metode metaheuristik yang dapat digunakan untuk masalah Traveling Salesman Problem .
Therefore, the shorter path is more and more frequently selected by the ants until, eventually all ants end up using the short path. (see [3], [5])
In this report simple antcolonyoptimization techniques are discussed. Some stress is laid on, how it is engineered and put to work so that a colony of artificial ants can find good solutions to difficult optimization problem. Also how these ant algorithms can be modified in order to enlarge the domain of application of these algorithms. In the last chapter a comparative study has been done to compare the performance of ACO with other optimization algorithms.
Keywords AntColonyOptimization (ACO), Swarm intelligence, meta-heuristics, OpenCL
I. INTRODUCTION
With a rapidly growing world and technology there are many challenges to solve complex problems in an efficient way and in a quick time. Antcolonyoptimization is one of the optimization techniques that have the capability of combining with other algorithms and improving their efficiency. It gives even better results when it is parallelized as it is intrinsically parallel. ACO is one of the naturally inspired metaheuristic algorithms. ACO takes the inspiration from ants. In ACO, colony of ants moves from nest to food source through different paths initially. While moving they lay a special chemical substance called pheromone. This pheromone is used
Unfortunately, since every real life problem are dynamic problem, thus their behaviors are much complex, GP suffers from serious weaknesses .random systems. Chaos is important, in part, because it helps us to cope with unstable system by improving our ability to describe, to understand, perhaps even to forecast them. AntColonyOptimization (ACO) is the result of research on computational intelligence approaches to combinatorial optimization. The fundamental approach underlying ACO is an iterative process in which a population of simple agents repeatedly construct candidate solutions; this construction process is probabilistically guided by heuristic information on the given problem instance as well as by a shared memory containing experience gathered by the ants in previous iteration. ACO has been applied to a broad range of hard combinatorial problems.
Index Terms—AntColonyOptimization, rule based classifiers, unstructured rules.
I. I NTRODUCTION
After sufficient training, a classifier’s task is to try to correctly predict the classes of a number of given instances of data. By examining the way that a particular system has been trained to respond, they can also be used to enhance our knowledge of a particular phenomenon. One such system, that directly fits this bill is the rule based classifier.
Abstract. The optimum solution of the production scheduling problem for manufacturing
processes at an enterprise is crucial as it allows one to obtain the required amount of production within a specified time frame. Optimum production schedule can be found using a variety of optimization algorithms or scheduling algorithms. Antcolonyoptimization is one of well-known techniques to solve the global multi-objective optimization problem. In the article, we present a solution of the production scheduling problem by means of an antcolonyoptimization algorithm. A case study of the algorithm efficiency estimated against some others production scheduling algorithms is presented. Advantages of the antcolonyoptimization algorithm and its beneficial effect on the manufacturing process are provided.
AntColonyOptimization (ACO) is a meta-heuristic inspired by the emergent behaviour of real ants [4]. Classification is a central problem in the fields of data mining and machine learning, which attempts to learn the relationship between the input attributes and a class in a training dataset so as to be able to predict the class of instances in unforeseen datasets. Because real world datasets will often contain noise, erroneous data, outliers, and mis-labeled and irrelevant instances, the data presented to a data mining algorithm will commonly first go through a phase that either removes attributes, removes instances, or both before it is processed. This pre-processing phase, known as data reduction, aims to improve the predictive effectiveness of the produced classifiers by reducing the number of aforementioned data anomalies. In addition, data reduction also has the benefit of decreasing the dataset processed by the learning algorithm by retaining only the most representative instances. Due to the high utility of performing data reduction, it is usually included in the pipeline of most real world classification applications.
Abstract: This paper attempts to overcome stagnation problem of AntColonyOptimization (ACO) algorithms. Stagnation is undesirable state which occurs at a later phases of the search process. Excessive pheromone values attract more ants and make further exploration hardly possible. This problem has been addressed by Genetic operations (GO) incorporated into ACO framework. Crossover and mutation operations have been adapted for use with ant generated strings which still have to provide feasible solutions. Genetic operations decrease selection pressure and increase probability of finding the global optimum. Extensive simulation tests were made in order to determine influence of genetic operation on algorithm performance.
Abstract: This contribution introduces an AntColonyOptimization (ACO) algorithm with re-initialization mechanism. The whole search process is broken by re-initialization into shorter semi-independent steps called “macro cycles”. The length of macro cycle depends on pheromone accumulation and can be adjusted by a user parameter. It is shown that re- initialization mechanism prevents ACO algorithm from pheromone saturation and consecutive stagnation. This approach avoids overhead caused by algorithm run with excessive pheromone values where further exploration is hardly possible. The solution offers lower CPU cost of the search process and enables automation of heuristic search especially in changing environments like dynamic networks. The efficiency of proposed method is demonstrated on a path minimization problem on 50 node graph.
Solving Sudoku with AntColonyOptimization
Huw Lloyd, Member, IEEE, and Martyn Amos
Abstract—In this paper we present a new algorithm for the well-known and computationally-challenging Sudoku puzzle game. Our AntColonyOptimization-based method significantly out-performs the state-of-the-art algorithm on the hardest, large instances of Sudoku. We provide evidence that – compared to traditional backtracking methods – our algorithm offers a much more efficient search of the solution space, and demonstrate the utility of a novel anti-stagnation operator. This work lays the foundation for future work on a general-purpose puzzle solver, and establishes Japanese pencil puzzles as a suitable platform for benchmarking a wide range of algorithms.
Recently, an extension of the ACO to continuous domains without any major conceptual change is proposed, which is called ACO R [10].
A variety of application problems in engineering, science and technology can be formu- lated as CnOPs having local as well as global optima. Mostly, the user is interested in determining the global minima. However, it is more difficult to determine the global min- ima rather than local minima. Especially, for a lot of multimodal problems, the number of the local minima is large. As a result, most algorithms are very easy to be trapped in the local minima. In order to avoid falling into local minima, a proper trade-off between the exploration and exploitation is necessary for the algorithm. In this paper an antcolonyoptimization based algorithm is proposed to solve the CnOPs efficiently. An operation similar to the crossover in the GA is introduced into the antcolonyoptimization. In
1, 2, 3
Department of Civil Engineering, MNIT Jaipur, (India)
ABSTRACT
Due to rapid urbanization coupled with population growth facilities, enactment of pollution control laws and increasing awareness towards sanitation, the problem of waste water collection and disposal is becoming difficult and requires large amount of money .The cost of a sewage collection network constitutes a major fraction of the overall cost of waste disposal. Thus, substantial sums of money can be saved by improving sewerage system design. Diameter and slope are the two major components that contribute to the cost of sewerage system. In this paper, a new and powerful intelligent evolution method, called antcolonyoptimization (ACO) is adopted for solving the optimization problem by formulating an objective function. The proposed algorithm is coded using FORTRAN. Then, ACO algorithm has been applied to the design of sewerage system through the optimization of the objective function. The obtained results reveal that the proposed method is promising in the optimal design of the sewerage system.
Shashvat Sanadhya
Student ,Dept .of CTA NITTTR Bhopal ,India
ABSTRACT
We are living world of virtual community, where people connect to each other through any kind of relationship. Social networking is platform where people share emotions, activities, area of interest etc. Communities in social network are deployed in user nodes with connecting people, second generation social network come in existence lot of emerging application which create service oriented environment, this paper is summarization of antcolonyoptimization with social networking, which is a growing discipline .Ant take decision with the help of special type of chemical „pheromone'. Base on this concept ACO meta-heuristic which helps to solve combinational problems such as TSP, Graph color, job shop Network routing, machine learning etc. Hence social networking may be a new platform with antcolonyoptimization, to solve complex task in social phenomena.
A hybrid antcolonyoptimization technique to solve the stagnation problem in grid computing is proposed in this paper. The proposed algorithm combines the techniques from AntColony System and Max – Min Ant System and focused on local pheromone trail update and trail limit. The agent concept is also integrated in this proposed technique for the purpose of updating the grid resource table. This facilitates the hybrid antcolonyoptimization technique in solving the stagnation problem in two ways within one cycle, thus minimize the total computational time of the jobs.
– It utilises a reducing search range.
2 Constrained AntColonyOptimizationAntColonyOptimization with Different Favor (ACODF ) applies ACO for use in data clustering. The difference between the ACODF and ACO is that each ant in ACODF only visits a fraction of the total clustering objects and the number of visited objects decreases with each cycle. ACODF also incorporates the strategies of simulated annealing and tournament selection and results in an algorithm which is effective for clusters with clearly defined boundaries. However, ACODF does not handle clusters with arbitrary shapes, clusters with outliers and bridges between clusters well. In order to improve the effectiveness of the clustering the following four strategies are applied:
Abstract: Opinion Mining could be a sort of process in Natural Language Processing to track the disposition or supposition of individuals about a particular item, subject or service. This can be too called as Opinion Investigation or sentiment analysis which includes building a framework to gather and look at the suppositions, feelings, around the item, subject and administrations made in web journal posts, comments, surveys or tweets. This paper thinks about opinion and opinion based classification for movie surveys. To begin with, feature extraction has been done with Inverse document frequency strategy. Information Gain based feature selection process has been done from the reviews for effective feature selection. At long last, Multi Objective function based AntColonyOptimization (MOFACO) procedure has been utilized for viable classification of surveys with optimized feature selection strategy. This inquire about work accomplishes the classification accuracy of 90.89%.
AntColonyOptimization (ACO) algorithms [1] have been successful at discovering classification rules, most notably Ant-Miner and its derivatives [13, 6, 12]. ACO algorithms use a colony of artificial ants, where each ant creates a can- didate solution by selecting individual components based on pheromone and heuristic information derived from the prob- lem domain. Components with a greater pheromone level are more likely to be selected by an ant. After creation, each ant solution quality is evaluated and components that are used in good quality solutions have their pheromone level increased while those that do not will have their pheromone reduced. Every iteration will produce a more refined so- lution until the colony converges on a set of near optimal solutions. Ant-Miner and its derivatives employ an ACO procedure to create classification rules, while the applica- tion of ACO algorithms to discover regression rules remains an unexplored area to the best of our knowledge.
Abstract—Seeking multiple optima simultaneously, which multimodal optimization aims at, has attracted increasing atten- tion but remains challenging. Taking advantage of antcolonyoptimization (ACO) algorithms in preserving high diversity, this paper intends to extend ACO algorithms to deal with multimodal optimization. First, combined with current niching methods, an adaptive multimodal continuous ACO algorithm is intro- duced. In this algorithm, an adaptive parameter adjustment is developed, which takes the difference among niches into consider- ation. Second, to accelerate convergence, a differential evolution mutation operator is alternatively utilized to build base vec- tors for ants to construct new solutions. Then, to enhance the exploitation, a local search scheme based on Gaussian distribution is self-adaptively performed around the seeds of niches. Together, the proposed algorithm affords a good bal- ance between exploration and exploitation. Extensive experiments on 20 widely used benchmark multimodal functions are con- ducted to investigate the influence of each algorithmic component and results are compared with several state-of-the-art multi- modal algorithms and winners of competitions on multimodal optimization. These comparisons demonstrate the competitive efficiency and effectiveness of the proposed algorithm, especially in dealing with complex problems with high numbers of local optima.
* Corresponding author. Tel.: +385 42 390872; email: Nikola.ivkovic@foi.hr Manuscript submitted September 24, 2015; accepted December 10, 2015.
doi: 10.17706/jcp.11.6.528-536
Abstract: DNA microarrays are manufactured by synthesizing probes on a solid surface with the help of light and a sequence of lithographic masks. Unintentional illumination can create defects on the microarray due to small dimensions and light properties, but a suitable arrangement of probes can reduce the probability of defects. The problem of designing DNA microarrays is computationally hard and there is no publicly available algorithm that can solve this problem exactly, in polynomial time. This study investigates the suitability of the antcolonyoptimization (ACO) metaheuristic for finding optimal or at least good microarray designs. This research is based on a MAX-MIN ant system variant that is enhanced with 2-opt local optimization and max-κ-best pheromone reinforcement strategy. Experiments were conducted on problem instances based on border length and conflict index models. The proposed algorithm found solutions that are better than the best solutions previously published for 10 out of 14 problem instances.