Showing papers on "Ant colony published in 2003"

The Ant Colony Optimization Metaheuristic: Algorithms, Applications, and Advances

Abstract: The field of ACO algorithms is very lively, as testified, for example, by the successful biannual workshop (ANTS—From Ant Colonies to Artificial Ants: A Series of International Workshops on Ant Algorithms; http://iridia.ulb.ac.be/~ants/) where researchers meet to discuss the properties of ACO and other ant algorithms, both theoretically and experimentally.

Speed versus accuracy in collective decision making.

Abstract: We demonstrate a speed versus accuracy trade-off in collective decision making. House-hunting ant colonies choose a new nest more quickly in harsh conditions than in benign ones and are less discriminating. The errors that occur in a harsh environment are errors of judgement not errors of omission because the colonies have discovered all of the alternative nests before they initiate an emigration. Leptothorax albipennis ants use quorum sensing in their house hunting. They only accept a nest, and begin rapidly recruiting members of their colony, when they find within it a sufficient number of their nest-mates. Here we show that these ants can lower their quorum thresholds between benign and harsh conditions to adjust their speed-accuracy trade-off. Indeed, in harsh conditions these ants rely much more on individual decision making than collective decision making. Our findings show that these ants actively choose to take their time over judgements and employ collective decision making in benign conditions when accuracy is more important than speed.

A Multiobjective Ant Colony System for Vehicle Routing Problem with Time Windows.

Abstract: This paper proposes a variation of the Multiple Ant Colony System for Vehicle Routing Problem with Time Windows (MACS-VRPTW) algorithm, which is based on an Ant Colony System approach, using two ant colonies to minimize first the number of vehicles and then the total traveled distance. As an improvement, the present work proposes to use a modified version specialized for a multiobjective context, using just one colony to get a set of Pareto optimal solutions considering three objectives at the same time, the number of vehicles, the total traveling time and the total delivery time. Experimental results validate the new approach with very good results when compared to the original MACS-VRPTW.

Ant algorithms for the university course timetabling problem with regard to the state-of-the-art

Abstract: Two ant algorithms solving a simplified version of a typical university course timetabling problem are presented -Ant Colony System and MAX-MIN Ant System The algorithms are tested over a set of instances from three classes of the problem Results are compared with recent results obtained with several metaheuristics using the same local search routine (or neighborhood definition), and a reference random restart local search algorithm Further, both ant algorithms are compared on an additional set of instances Conclusions are drawn about the performance of ant algorithms on timetabling problems in comparison to other metaheuristics Also the design, implementation, and parameters of ant algorithms solving the university course timetabling problem are discussed It is shown that the particular implementation of an ant algorithm has significant influence on the observed algorithm performance

Nestmate discrimination in ants: effect of bioassay on aggressive behavior

Abstract: Aggression assays are commonly used to study nestmate recognition in social insects. Methods range from detailed behavioral observations on small numbers of insects to counts of individuals fighting in group interactions. These assays vary in the equipment used and the intensity and duration of observations. We used the Argentine ant, Linepithema humile, to compare four aggression bioassays for consistency between replicates, similarity between assays, and ability to predict whole colony interactions. The assays included were 1 live – 1 dead ant interactions, live 1-1 battles, live 5-5 battles, and 1 ant introduced to a foreign colony. We tested six ant colonies in all pairwise combinations using four different assays and two to three scoring methods per assay. We also conducted a colony merging experiment to see which assays were capable of predicting this ecologically important event. We found that scoring methods within assays yielded very similar results, giving us no reason to favor observationally intense procedures, such as continuous scanning, over less observationally intense systems, such as snapshot surveys. Assays differed greatly in their consistency between replicates. No two replicates of the 1 live – 1 dead assay were significantly correlated. The live 5-5 and the colony introduction assays were the most consistent across replicates. The mean scores of the live 1-1, live 5-5 and colony introduction assays were all significantly correlated with each other; only the live 5-5 assay was significantly correlated with the 1 live – 1 dead assay. Assays that utilized the greatest number of live ants were the most likely to reveal high levels of aggression. The aggression scores of all but the 1 live – 1 dead assay were positively correlated with the number of ants that died during whole colony encounters and negatively associated with colony merging. We conclude that all live ant assays tested are useful tools for analyzing aggressive interactions between colonies, but that the pairing of a live and dead ant produced inconsistent results and generally lower levels of aggression. We found relatively low consistency between trials using the live 1-1 assay, but found that with sufficient replication its results were highly correlated with the assays using more interacting ants. We suggest that isolated aggressive acts in assays do not necessarily predict whole colony interactions : some colonies that fought in bioassays merged when the entire colonies were allowed to interact.

Web usage mining using artificial ant colony clustering and linear genetic programming

Abstract: The rapid e-commerce growth has made both business community and customers face a new situation. Due to intense competition on the one hand and the customer's option to choose from several alternatives, the business community has realized the necessity of intelligent marketing strategies and relationship management. Web usage mining attempts to discover useful knowledge from the secondary data obtained from the interactions of the users with the Web. Web usage mining has become very critical for effective Web site management, creating adaptive Web sites, business and support services, personalization, network traffic flow analysis and so on. The study of ant colonies behavior and their self-organizing capabilities is of interest to knowledge retrieval/management and decision support systems sciences, because it provides models of distributed adaptive organization, which are useful to solve difficult optimization, classification, and distributed control problems, among others [Ramos, V. et al. (2002), (2000)]. In this paper, we propose an ant clustering algorithm to discover Web usage patterns (data clusters) and a linear genetic programming approach to analyze the visitor trends. Empirical results clearly show that ant colony clustering performs well when compared to a self-organizing map (for clustering Web usage patterns) even though the performance accuracy is not that efficient when compared to evolutionary-fuzzy clustering (i-miner) [Abraham, A. (2003)] approach.

Optimization of multi-pass turning operations using ant colony system

Abstract: This paper proposes a new optimization technique based on the ant colony algorithm for solving multi-pass turning optimization problems. The cutting process has roughing and finishing stages. The machining parameters are determined by minimizing the unit production cost, subject to various practical machining constraints. The results indicate that the proposed ant colony framework is effective compared to other techniques carried out by different researchers.

High rates of army ant raids in the Neotropics and implications for ant colony and community structure

Abstract: Army ants form nomadic insect colonies whose chief food is other social insects. Here we compare the rate of army ant raids with the average density of their potential prey from 28 New World subtropical and tropical localities. We estimate that army ant raids occur at the rate of 1.22 m 2 per day in tropical forests. Army ant raid rates increased with primary productivity, and with the density of potential prey (litter-nesting ant colonies), across sites. Our estimates of raid rates for army ant guilds are much higher than previously published estimates based solely on surface-raiding Eciton. Life-history theory predicts that high rates of predation on insect societies will select for both smaller average colony sizes and indeterminate colony growth, and these traits have been documented for tropical ant litter-nesting ants. Our results suggest that army ant predation can affect both patterns.

Ant colony search algorithm for unit commitment

Abstract: In this paper, the ant colony search algorithm (ACSA) is proposed to solve the thermal unit commitment problem ACSA is a new cooperative agents approach, which is inspired by the observation of the behaviors of real ant colonies on the topic of ant trial formation and foraging methods In the ACSA, a set of cooperating agents called "ants" cooperates to find good solution for unit commitment problem of thermal units The merits of ACSA are parallel search and optimization capabilities The problem is decomposed in two sub-problems The unit commitment sub-problem is solved by the ant colony search algorithm method and the economic dispatch sub-problem is solved by the lambda-iteration method The unit commitment problem is formulated as the minimization of the performance index, which is the sum of objectives (fuel cost, start-up cost) and some constraints (power balance, generation limits, spinning reserve, minimum up time and minimum down time) This proposed approach is tested and compared to conventional Lagrangian relaxation (LR), genetic algorithm (GA), evolutionary programming (EP), Lagrangian relaxation and genetic algorithm (LRGA) on the 10 unit system

Nest Defense and Conspecific Enemy Recognition in the Desert Ant Cataglyphis fortis

Abstract: This study focuses on different factors affecting the level of aggression in the desert ant Cataglyphis fortis. We found that the readiness to fight against conspecific ants was high in ants captured close to the nest entrance (0- and 1-m distances). At a 5-m distance from the nest entrance the level of aggression was significantly lower. As the mean foraging range in desert ants by far exceeds this distance, the present account clearly shows that in C. fortis aggressive behavior is displayed in the context of nest, rather than food-territory defense. In addition, ants were more aggressive against members of a colony with which they had recently exchanged aggressive encounters than against members of a yet unknown colony. This finding is discussed in terms of a learned, enemy-specific label-template recognition process.

Intelligent selective disassembly using the ant colony algorithm

Abstract: Selective disassembly is an important issue in industrial and mechanical engineering for environmentally conscious manufacturing. This paper presents an intelligent selective disassembly approach based on ant colony algorithms, which take inspiration from the behavior of real ant colonies and are used to solve combinatorial optimization problems. For diverse assemblies, the algorithm generates different amounts of ants cooperating to find disassembly sequences for selected components, minimizing the reorientation of assemblies and removal of components. A candidate list that is composed of feasible disassembly operations, which are derived from a disassembly matrix of products, guides sequence construction in the implicit solution space and ensures the geometric feasibility of sequences. Preliminary implementation results show the effectiveness of the proposed method.

An improved ant colony optimisation algorithm for the 2D HP protein folding problem

Abstract: The prediction of a protein's structure from its amino-acid sequence is one of the most important problems in computational biology. In the current work, we focus on a widely studied abstraction of this problem, the 2-dimensional hydrophobic-polar (2D HP) protein folding problem. We present an improvedv ersion of our recently proposed Ant Colony Optimisation (ACO) algorithm for this NP-hardcom binatorial problem and demonstrate its ability to solve standard benchmark instances substantially better than the original algorithm; the performance of our new algorithm is comparable with state-of-the-art Evolutionary andMon te Carlo algorithms for this problem. The improvements over our previous ACO algorithm include long range moves that allows us to perform modification of the protein at high densities, the use of improving ants, ands elective local search. Overall, the results presented here establish our new ACO algorithm for 2D HP protein folding as a state-of-the-art methodf or this highly relevant problem from bioinformatics.

Patterns of the Crematogaster-Macaranga association: The ant partner makes the difference

Abstract: One of the most species-rich ant-plant mutualisms worldwide is the palaeotropical Crematogaster-Macaranga system. Although the biogeography and ecology of both partners have been extensively studied, little is known about the temporal structuring and the dynamics of the association. In this study we compared life-history traits of the specific Crematogaster (Decacrema) partner-ants and followed the development of ant colonies on eight different Macaranga host plant species, from colony founding on saplings to adult trees in a snapshot fashion. We found differences in the onset of alate production, queen number and mode of colony founding in the ant species and examined the consequences of these differences for the mutualism with the host plant. The lifespan of some host plants and their specific ant partners seemed to be well matched whereas on others we found an ontogenetic succession of specific partner ants. The partner ants of saplings or young plants often differed from specific partner ants found on larger trees of the same species. Not all specific Crematogaster species can re-colonize the crown region of adult trees, thus facilitating a change of ant species. Therefore lifespan of the ant colony as well as colony founding behaviour of the different partner ant species are important for these ontogenetic changes. The lifespan of a colony of two species can be prolonged via secondary polygyny. For the first time, also primary polygyny (pleometrosis) is reported from this myrmecophytic system.

MRF-based image segmentation using Ant Colony System

Abstract: In this paper, we propose a novel method for image segmentation that we call ACS-MRF method. ACS-MRF is a hybrid ant colony system coupled with a local search. We show how a colony of cooperating ants are able to estimate the labels field and minimize the MAP estimate. Cooperation between ants is performed by exchanging information through pheromone updating. The obtained results show the efficiency of the new algorithm, which is able to compete with other stochastic optimization methods like Simulated annealing and Genetic algorithm in terms of solution quality.

Dynamic ant colony optimisation for TSP

Abstract: Ants exhibit collective behaviour in performing tasks that cannot be carried out by an individual ant. When ants are working, they must communicate with each other through a kind of chemical substance—pheromones. Ants look for food and lay the way back to their nest with pheromones, and the other ants can follow the pheromone to find the food efficiently. Using the analogy of foraging behaviour and pheromones, Marco Dorigo proposed the ant algorithm and applied it to solving the travelling salesman problem (TSP) and solving job-shop scheduling. In this paper, we simulate real ants with more aspects. Updating of pheromones is more likely to be the real situation in the natural world. Our algorithm shows a better performance than the original algorithm.

Multi-robot cooperation method based on the ant algorithm

Abstract: The ant algorithm is an optimization algorithm that is gained by observing the real ant colonies, and it is very useful in solving difficult optimization problems and distributed control problems. The algorithm is modeled on a key concept called "stigmergy" of the ant societies, which is used in our algorithm to design the cooperation of multi-robots. In an unknown environment, one of the most important problems in the multi-robot system is to decide how many robots are needed to complete a task. With the concept "stigmergy", the number of the robots cooperating on a task is decided according to the difficulty of the task. Given the definition of the "task deadlock", an adaptive attenuation factor to eliminate task deadlock is introduced in the cooperation algorithm. Simulations show the effectiveness of the algorithm.

DNA fragment assembly using an ant colony system algorithm

Abstract: This work presents the use of an ant colony system algorithm in a DNA (deoxyribonucleic acid) fragment assembly. The assembly problem is a combinatorial optimisation problem where the aim of the search is to find the right order and orientation of each fragment in the fragment ordering sequence that leads to the formation of a consensus sequence. In this paper, an asymmetric ordering representation is proposed where a path cooperatively generated by all ants in the colony represents the search solution. The optimality of the fragment layout obtained is then determined from the sum of overlap scores calculated for each pair of consecutive fragments in the layout. Two types of assembly problem are investigated: single-contig and multiple-contig problems. The simulation results indicate that in single-contig problems, the performance of the ant colony system algorithm is approximately the same as that of a nearest neighbour heuristic algorithm. On the other hand, the ant colony system algorithm outperforms the nearest neighbour heuristic algorithm when multiple-contig problems are considered.

Ants: A food source sought by Slovenian brown bears (Ursus arctos)?

Abstract: In the heavily managed boreal forest of Scandinavia, ants, especially large colonies of red forest ants (Formica spp.), are abundant and brown bears (Ursus arctos) intensively feed on them. In contrast, the beech (Fagus sylvatica) forests of Slovenia provide only suboptimal habitat for ants and large ant colonies are virtually absent. To quantify how much ant use by brown bears is a matter of availability or preference, we quantified ant availability, species composition, and ant use. The estimated biomass of ants available to brown bears was very low in Slovenia compared with those in Sweden, averaging 135 vs. 9600 g/ha, respectively. Nevertheless, the frequency of occurrence of ants in Slovenian brown bear scats was high, averaging 85% and accounting for 25% of the ingested dry mass during the summer, which was nearly as much as their frequency of occurrence in Swedish brown bear scats during the summer. Although brown bears in Slovenia had year-round access to artificial feeding sites and the availabil...

Allometric scaling of ant foraging trail networks

Abstract: The aggregation of individuals into colonies raises important questions about scaling of structure and function. We model the metabolic benefits and costs of two-dimensional, fractallike foraging trails, such as those used by ant colonies. Total area foraged by the colony and, consequently, resource flow to the nest and rate of colony metabolism, increase non-linearly with number of foragers (F) as F 2/3 . Since the cost of foraging increases linearly with F, the model predicts an optimal number of foragers and, therefore, total foraging area that maximize colony fitness or energy allocation to reproduction. The scaling of foraging may influence evolution of coloniality.

Effects of group composition and level of selection in the evolution of cooperation in artificial ants

Abstract: Since ants and other social insects have long generation time, it is very difficult for biologists to study the origin of complex social organization by guided evolution (a process where the evolution of a trait can be followed during experimental evolution). Here we use colonies of artificial ants implemented as small mobile robots with simple vision and communication abilities to explore these issues. In this paper, we present results concerning the role of relatedness (genetic similarity) and levels of selection (individual and colony-level selection) on the evolution of cooperation and division of labor in simulated ant colonies. In order to ensure thorough statistical analysis, the evolutionary experiments, herein reported, have been carried out using “minimalist” simulations of the collective robotics evolutionary setup. The results show that altruistic behaviors have low probability of emerging in heterogeneous colonies evolving under individual-level selection and that colonies with high genetic relatedness display better performance.

Making the most of two heuristics: breaking transposition ciphers with ants

Abstract: Multiple anagramming is a general method for the cryptanalysis of transposition ciphers, and has a graph theoretic representation. Inspired by a partially mechanised approach used in World War II, we consider the possibility of a fully automated attack. Two heuristics based on measures of natural language are used - one to recognise plaintext, and another to guide construction of the secret key. This is shown to be unworkable for cryptograms of a certain difficulty due to random variation in the constructive heuristic. A solver based on an ant colony optimisation (AGO) algorithm is then introduced, increasing the range of cryptograms that can be treated; the pheromone feedback provides a mechanism for the recognition heuristic to correct the noisy constructive heuristic.

Static Multi-processor Scheduling with Ant Colony Optimisation & Local Search

Abstract: Efficient multi-processor scheduling is essentially the problem of allocating a set of computational jobs to a set of processors to minimise the overall execution time. There are many variations of this problem, most of which are NP-hard, so we must rely on heuristics to solve real world problem instances. This dissertation describes several novel approaches using the ant colony optimisation (ACO) meta-heuristic and local search techniques, including tabu search, to two important versions of the problem: the static scheduling of independent jobs onto homogeneous and heterogeneous processors. Finding good schedules for jobs allocated on homogeneous processors is an important consideration if efficient use is to be made of a multiple-CPU machine, for example. An ACO algorithm to solve this problem is presented which, when combined with a fast local search procedure, can outperform traditional approaches on benchmark problems instances for the closely related bin packing problem. The algorithm cannot compete, however, with more modern specialised techniques. Scheduling jobs onto hetereogeneous processors is a more general problem which has potential applications in domains such as grid computing. A fast local search procedure for this problem is described which can quickly and effectively improve solutions built by other techniques. When used in conjunction with a well-known heuristic, Min-min, it can find shorter schedules on benchmark problems than other solution techniques found in the literature, and in significantly less time. A tabu search algorithm is also presented which can improve on solutions found by the local search procedure but takes longer. Finally a hybrid ACO algorithm which incorporates the local and tabu searches is described which outperforms both, but takes significantly longer to run.

Ant colony system with local search for Markov random field image segmentation

Abstract: In this paper, we propose a new algorithm for image segmentation based on the Markov random field (MRF) and the ant colony optimization (AGO) metaheuristic. The underlying idea is to take advantage from the ACO metaheuristic characteristics and the MRF theory to develop a novel agents-based approach to segment an image. The proposed algorithm is based on a population of simple agents which construct a candidate partition by a relaxation labeling with respect to the contextual constraints. The obtained results show the efficiency of the new algorithm and that it competes with other global stochastic optimization methods like simulated annealing and genetic algorithm.

An ant colony system approach for scheduling problems

Abstract: Over the past 50 years, researchers have developed many simple constructive heuristics for the scheduling problem. A major defect of these heuristics is the non-robustness of their solutions. An ant colony system (ACS) approach is presented to continuously improve the constructive heuristics. To verify the developed ACS approach, a computational study is conducted on the single machine total weighted tardiness problem. The results show that the proposed approach can effectively improve the robustness of various constructive heuristics, and outperform the existing heuristics for a well-known benchmark problem set. From the viewpoints of both the solution quality and computational expenses, the proposed ACS approach is an efficient and effective method for scheduling problems.

An adaptive ant colony system algorithm for continuous-space optimization problems.

Abstract: Ant colony algorithms comprise a novel category of evolutionary computation methods for optimization problems, especially for sequencing-type combinatorial optimization problems. An adaptive ant colony algorithm is proposed in this paper to tackle continuous-space optimization problems, using a new objective-function-based heuristic pheromone assignment approach for pheromone update to filtrate solution candidates. Global optimal solutions can be reached more rapidly by self-adjusting the path searching behaviors of the ants according to objective values. The performance of the proposed algorithm is compared with a basic ant colony algorithm and a Square Quadratic Programming approach in solving two benchmark problems with multiple extremes. The results indicated that the efficiency and reliability of the proposed algorithm were greatly improved.

Interhabitat differences in ant activity on plant foliage: ants at extrafloral nectaries of Hibiscus pernambucensis in sandy and mangrove forests

Abstract: The association between visiting ants and the extrafloral nectaries (EFN)-bearing shrub Hibiscus pernambucensis Arruda (Malvaceae) was investigated in two different coastal habitats ‐ a permanently dry sandy forest and a regularly inundated mangrove forest. In both habitats the frequency of plants with ants and the mean number of ants per plant were much higher on H. pernambucensis than on non-nectariferous neighbouring plants. In the sandy forest the proportion of live termite baits attacked by ants on H. pernambucensis was much higher than on plants lacking EFNs. In the mangrove, however, ants attacked equal numbers of termites on either plant class. Ant attendance to tuna/ honey baits revealed that overall ant activity in the sandy forest is higher than in the mangrove area. The vertical distribution (ground vs. foliage) of ant activity also differed between habitats. While in the mangrove foraging ants were more frequent at baits placed on foliage, in the sandy forest ant attendance was higher at ground baits. Plants housing ant colonies were more common in the mangrove than in the sandy forest. Frequent flooding in the mangrove may have resulted in increased numbers of ant nests on vegetation and scattered ant activity across plant foliage, irrespective of possession of EFNs. Thus plants with EFNs in the mangrove may not experience increased ant aggression towards potential herbivores relative to plants lacking EFNs. The study suggests that the vertical distribution of ant activity, as related to different nest site distribution (ground vs. foliage) through a spatial scale, can mediate ant foraging patterns on plant foliage and probably affect the ants’ potential for herbivore deterrence on an EFN-bearing plant species.

The K-means clustering algorithm based on density and ant colony

Abstract: The ant algorithm is a new evolutional method, k-means and the density-cluster are familiar cluster analysis In this paper, we proposed a new K-means algorithm based on density and ant theory, which resolved the problem of local minimal by the random of ants and handled the initial parameter sensitivity of k-means In addition it combined idea of density and made the ants searching selectable With the experiments it was proved that the algorithm we proposed improved the efficiency and precision of cluster

The accumulated experience ant colony for the travelling salesman problem

Abstract: Ant colony optimization techniques are usually guided by pheromone and heuristic cost information when choosing the next element to add to a solution. However, while an individual element may be attractive, usually its long term consequences are neither known nor considered. For instance, a short link in a traveling salesman problem may be incorporated into an ant's solution, yet, as a consequence of this link, the rest of the path may be longer than if another link was chosen. The Accumulated Experience Ant Colony uses the previous experiences of the colony to guide in the choice of elements. This is in addition to the normal pheromone and heuristic costs. Two versions of the algorithm are presented, the original and an improved AEAC that makes greater use of accumulated experience. The results indicate that the original algorithm finds improved solutions on problems with less than 100 cities, while the improved algorithm finds better solutions on larger problems.