Showing papers on "Ant colony published in 2005"

Ant colony system for a dynamic vehicle routing problem

Abstract: An aboundant literature on vehicle routing problems is available. However, most of the work deals with static problems, where all data are known in advance, i.e. before the optimization has started. The technological advances of the last few years give rise to a new class of problems, namely the dynamic vehicle routing problems, where new orders are received as time progresses and must be dynamically incorporated into an evolving schedule. In this paper a dynamic vehicle routing problem is examined and a solving strategy, based on the Ant Colony System paradigm, is proposed. Some new public domain benchmark problems are defined, and the algorithm we propose is tested on them. Finally, the method we present is applied to a realistic case study, set up in the city of Lugano (Switzerland).

Training feed-forward neural networks with ant colony optimization: an application to pattern classification

Abstract: Ant colony optimization (ACO) is an optimization technique that was inspired by the foraging behaviour of real ant colonies. Originally, the method was introduced for the application to discrete optimization problems. Research efforts led to the development of algorithms for the application to continuous optimization problems. In this paper we extend and apply one of the most successful variants for the training of feed-forward neural networks. For evaluating our algorithm we apply it to pattern classification problems from the medical field. The results show that our algorithm is comparable to specialized algorithms for neural network training, and that it has advantages over other general purpose optimizers.

Insect communication: 'no entry' signal in ant foraging.

Abstract: Forager ants lay attractive trail pheromones to guide nestmates to food1, 2, but the effectiveness of foraging networks might be improved if pheromones could also be used to repel foragers from unrewarding routes3, 4. Here we present empirical evidence for such a negative trail pheromone, deployed by Pharaoh's ants (Monomorium pharaonis) as a 'no entry' signal to mark an unrewarding foraging path. This finding constitutes another example of the sophisticated control mechanisms used in self-organized ant colonies

Supercolonies of the invasive yellow crazy ant, Anoplolepis gracilipes, on an oceanic island: Forager activity patterns, density and biomass

Abstract: Ants have the capacity to reach unusually high densities, mostly in their introduced ranges. Numerical dominance is often cited as key to the ability of exotic ants to displace native ant species, reduce the abundance of invertebrates and negatively impact upon bird, land crab and other vertebrate populations. On Christmas Island, Indian Ocean, the yellow crazy ant, Anoplolepis gracilipes (Jerdon), forms supercolonies, where extremely high densities of foraging ants have contributed to ‘invasional meltdown’ in rainforest areas. Densities of up to 2254 foraging ants per m2 and a biomass of 1.85 g per m2 were recorded, and nest densities reached 10.5 nest entrances per m2. Populations of A. gracilipes can overcome and kill red endemic land crabs (Gecarcoidea natalis) over 100 times their own biomass. This is the highest recorded density of foraging ants, and adds another element to the definition of ‘supercolony’ of unicolonial ants. This paper documents one extreme in a continuum of densities of unicolonial, invasive ant species and highlights the need to incorporate forager densities into invasive ant research.

Using ant agents to combine reactive and proactive strategies for routing in mobile ad hoc networks

Abstract: This paper describes AntHocNet, an algorithm for routing in mobile ad-hoc networks based on ideas from the ant colony optimisation framework. In AntHocNet a source node reactively sets up a path to a destination node at the start of each communication session. During the course of the session, the source node uses ant agents to proactively search for alternatives and improvements of the original path. This allows to adapt to changes in the network, and to construct a mesh of alternative paths between source and destination. The proactive behaviour is supported by a lightweight information bootstrapping process. Paths are represented in the form of distance-vector routing tables called pheromone tables. An entry of a pheromone table contains the estimated goodness of going over a certain neighbour to reach a certain destination. Data are routed stochastically over the different paths of the mesh according to these goodness estimates. In an extensive set of simulation tests, we compare AntHocNet to AODV, a reactive algorithm which is an important reference in this research area. We show that AntHocNet can outperform AODV for different evaluation criteria in a wide range of different scenarios. AntHocNet is also shown to scale well with respect to the number of nodes.

Division of labour and worker size polymorphism in ant colonies: the impact of social and genetic factors

Abstract: Division of labour among workers is central to the organisation and ecological success of insect societies. If there is a genetic component to worker size, morphology or task preference, an increase in colony genetic diversity arising from the presence of multiple breeders per colony might improve division of labour. We studied the genetic basis of worker size and task preference in Formica selysi, an ant species that shows natural variation in the number of mates per queen and the number of queens per colony. Worker size had a heritable component in colonies headed by a doubly mated queen (h2=0.26) and differed significantly among matrilines in multiple-queen colonies. However, higher levels of genetic diversity did not result in more polymorphic workers across single- or multiple-queen colonies. In addition, workers from multiple-queen colonies were consistently smaller and less polymorphic than workers from single-queen colonies. The relationship between task, body size and genetic lineage appeared to be complex. Foragers were significantly larger than brood-tenders, which may provide energetic or ergonomic advantages to the colony. Task specialisation was also often associated with genetic lineage. However, genetic lineage and body size were often correlated with task independently of each other, suggesting that the allocation of workers to tasks is modulated by multiple factors. Overall, these results indicate that an increase in colony genetic diversity does not increase worker size polymorphism but might improve colony homeostasis.

Ant colony system-based algorithm for constrained load flow problem

Abstract: This paper presents the ant colony system (ACS) method for network-constrained optimization problems. The developed ACS algorithm formulates the constrained load flow (CLF) problem as a combinatorial optimization problem. It is a distributed algorithm composed of a set of cooperating artificial agents, called ants, that cooperate among them to find an optimum solution of the CLF problem. A pheromone matrix that plays the role of global memory provides the cooperation between ants. The study consists of mapping the solution space, expressed by an objective function of the CLF on the space of control variables [ant system (AS)-graph], that ants walk. The ACS algorithm is applied to the IEEE 14-bus system and the IEEE 136-bus system. The results are compared with those given by the probabilistic CLF and the reinforcement learning (RL) methods, demonstrating the superiority and flexibility of the ACS algorithm. Moreover, the ACS algorithm is applied to the reactive power control problem for the IEEE 14-bus system in order to minimize real power losses subject to operating constraints over the whole planning period.

Ant Colony Algorithms for Steiner Trees: An Application to Routing in Sensor Networks

Abstract: This chapter introduces ant colony optimization as a method for computing minimum Steiner trees in graphs. Tree computation is achieved when multiple ants, starting out from different nodes in the graph, move towards one another and ultimately merge into a single entity. A distributed version of the proposed algorithm is also described, which is applied to the specific problem of data-centric routing in wireless sensor networks. This research illustrates how tree based graph theoretic computations can be accomplished by means of purely local ant interaction. The authors hope that this work will demonstrate how innovative ways to carry out ant interactions can be used to design effective ant colony algorithms for complex optimization problems. This chapter appears in the book, Recent Developments in Biologically Inspired Computing, edited by Leandro N. de Castro and Fernando J. Von Zuben. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. 701 E. Chocolate Avenue, Suite 200, Hershey PA 17033-1240, USA Tel: 717/533-8845; Fax 717/533-8661; URL-http://www.idea-group.com IDEA GROUP PUBLISHING 182 Singh, Das, Gosavi & Pujar Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited. INTRODUCTION Ants live in colonies and have evolved to exhibit very complex patterns of social interaction. Such interactions are clearly seen in the foraging strategy of ants. Despite the extremely simplistic behavior of individual ants, they can communicate with one another through secretions called pheromones, and this cooperative activity of the ants in a nest gives rise to an emergent phenomenon known as swarm intelligence (Bonabeau et al., 1999). Ant Colony Optimization (ACO) algorithms are a class of algorithms that mimic the cooperative behavior of real ant behavior to achieve complex computations. Ant colony optimization was originally introduced as a meta-heuristic for the wellknown traveling salesman problem (TSP), which is a path based optimization problem. This problem is proven to be NP-complete, which is a subset of a class of difficult optimization problems that are not solvable in polynomial time (unless P=NP). Since an exponential time algorithm is infeasible for larger scale problems in class NP, much research has focused on applying stochastic optimization algorithms such as genetic algorithms and simulated annealing to obtain good (but not necessarily globally optimal) solutions. The ant colony approach was subsequently shown to be a very effective technique for approaching a variety of other combinatorial optimization problems in class NP. An intrinsic advantage of ACO is the relative ease of implementation in a decentralized environment. These algorithms have therefore been applied to distributed network based problems that involve optimal path computations, such as routing, load balancing, and multicasting in computer networks (Bonabeau et al., 1998; Das et al., 2002; Navarro-Varela & Sinclair, 1999; Schoonderwoerd, 1997). In the rest of this chapter, we will use the terms distributed algorithm, online algorithm and decentralized algorithm interchangeably to imply algorithms that do not require any form of global computation. Algorithms that do require it will be referred to as centralized, or offline algorithms. This chapter explores the application of ant colony algorithms to the data-centric routing in sensor networks. This problem involves establishing paths from multiple sources in a sensor network to one or more destinations, where data are aggregated at intermediate stages in the paths for optimal dissemination. When only a single destination is involved, the optimal path amounts to a minimum Steiner tree in the sensor network. The minimum Steiner tree problem is a classic NP-complete problem that has numerous applications. It is a problem of extracting a sub-tree from a given graph with certain properties. A formal description of the problem is postponed until later. The second section introduces the ant colony optimization approach. The Steiner tree problem is introduced here and its applicability to sensor networks taken up in detail. The third section provides the details of the algorithm. It first describes an offline algorithm that can be used to compute Steiner trees of any graph. A preliminary set of simulations carried out to demonstrate the algorithm’s effectiveness is included. This is followed in the fourth section by a detailed description of the online algorithm to establish optimal paths for data-centric routing. Simulation results for three separate randomly generated networks are analyzed. In the fifth section, further extensions and applications of the present algorithm are suggested. Conclusions are provided in the last section. 24 more pages are available in the full version of this document, which may be purchased using the "Add to Cart" button on the product's webpage: www.igi-global.com/chapter/ant-colony-algorithms-steinertrees/28328?camid=4v1 This title is available in InfoSci-Books, InfoSci-Medical, Communications, Social Science, and Healthcare. Recommend this product to your librarian: www.igi-global.com/e-resources/libraryrecommendation/?id=1

Solving multiobjective multicast routing problem with a new ant colony optimization approach

Abstract: This work presents two multiobjective algorithms for Multicast Traffic Engineering. The proposed algorithms are new versions of the Multi-Objective Ant Colony System (MOACS) and the Max-Min Ant System (MMAS), based on Ant Colony Optimization (ACO). Both ACO algorithms simultaneously optimize maximum link utilization and cost of a multicast routing tree, as well as average delay and maximum end-to-end delay, for the first time using an ACO approach. In this way, a set of optimal solutions, know as Pareto set is calculated in only one run of the algorithms, without a priori restrictions. Experimental results show a promising performance of both proposed algorithms for a multicast traffic engineering optimization, when compared to a recently published Multiobjective Multicast Algorithm (MMA), specially designed for Multiobjective Multicast Routing Problems.

Self-organized digging activity in ant colonies

Abstract: Many ant species adjust the volume of their underground nest to the colony size. We studied whether the regulation of the volume of excavated sand could result from an interplay between recruitment processes and ant density. Experiments were performed with different group sizes of workers in the ant Messor sancta. When presented with a thin homogeneous sand disk, these groups excavated networks of galleries in less than 3 days. The excavation dynamics were logistic shaped, which suggests the existence of a double feedback system: a positive one resulting in an initial exponential growth phase, and a negative one leading the dynamics to a saturation phase. The total volume of excavated sand was almost proportional to the number of workers. We then developed a model in which we incorporated the quantitative behavioral rules of the workers’ digging activity. A positive feedback was introduced in the form of a recruitment process mediated by pheromones. The model predicts that the excavation dynamics should be logistic shaped and the excavation should almost stop despite the absence of any explicit negative feedback. Moreover, the model was able to reproduce the positive linear relationship between nest volume and colony size.

Importance of large colony size for successful invasion by Argentine ants (Hymenoptera: Formicidae): Evidence for biotic resistance by native ants

Abstract: The Argentine ant, Linepithema humile (Mayr), is a widespread invasive ant species that has been associated with losses of native ant species and other invertebrates from its introduced range. To date, various abiotic conditions have been associated with limitations to the spread of Argentine ants, however, competitive interactions with native ant fauna may also affect the spread of Argentine ants. Here, we experimentally manipulated colony sizes of Argentine ants in the laboratory to assess whether Argentine ants were able to survive and compete for resources with a widespread, dominant native ant, Iridomyrmex 'rufoniger'. The results showed that over 24 h, the proportions of Argentine ants that were alive, at baits, and at sugar water decreased significantly in the presence of Iridomyrmex. In addition, Argentine ant mortality increased over time, however, the proportion of the colony that was dead decreased with the largest colony size. Argentine ants were only able to overcome Iridomyrmex when their colony sizes were 5-10 times greater than those of the native ants. We also conducted trials in which colonies of Argentine ants of varying sizes were introduced to artificial baits occupied by Iridomyrmex in the field. The results showed that larger Argentine ant colonies significantly affected the foraging success of Iridomyrmex after the initial introduction (5 min). However, over the first 20 min, when the Argentine ants were present at the baits, and over the entire 50 min experimental period, the numbers of Iridomyrmex at baits did not differ significantly with the size of the Argentine ant colony. This is the first experimental study to investigate the role of colony size in the invasion biology of Argentine ants in Australia, and the results suggest that Iridomyrmex may reduce the spread of Argentine ants, and that Argentine ants may need to attain large colony sizes in order to survive in the presence of Iridomyrmex. We address the implications of these findings for the invasion success of Argentine ants in Australia, and discuss the ability of Argentine ants to attain large colony sizes in introduced areas.

The Ant Algorithm for Solving Robot Path Planning Problem

Abstract: Using ant algorithm, robot path planning in two-dimension environment is studied. It introduces the intelligent finding optimum mechanism of ant colony. It solves the drawback of local optimization and expedites searching speed. The mathematical model is established and the algorithm is achieved with VB language, the result shows it is valid with the capability of robust and extensibility

Evolutionary methods for ant colony paintings

Abstract: We investigate evolutionary methods for using an ant colony optimization model to evolve “ant paintings.” Our model is inspired by the recent work of Monmarche et al. The two critical differences between our model and that of Monmarche's are: (1) we do not use an interactive genetic algorithm, and (2) we allow the pheromone trail to serve as both a repelling and attracting force. Our results show how different fitness measures induce different artistic “styles” in the evolved paintings. Moreover, we explore the sensitivity of these styles to perturbations of the parameters required by the genetic algorithm. We also discuss the evolution and interaction of various castes within our artificial ant colonies.

Self-Regulated Artificial Ant Colonies on Digital Image Habitats

Abstract: Artificial life models, swarm intelligent and evolutionary computation algorithms are usually built on fixed size populations Some studies indicate however that varying the population size can increase the adaptability of these systems and their capability to react to changing environments In this paper we present an extended model of an artificial ant colony system designed to evolve on digital image habitats We will show that the present swarm can adapt the size of the population according to the type of image on which it is evolving and reacting faster to changing images, thus converging more rapidly to the new desired regions, regulating the number of his image foraging agents Finally, we will show evidences that the model can be associated with the Mathematical Morphology Watershed algorithm to improve the segmentation of digital grey-scale images KEYWORDS: Swarm Intelligence, Perception and Image Processing, Pattern Recognition, Mathematical Morphology, Social Cognitive Maps, Social Foraging, Self-Organization, Distributed Search

Developing a diagnostic system through integration of fuzzy case-based reasoning and fuzzy ant colony system

Abstract: This study intends to propose a hybrid Case-Based Reasoning (CBR) system with the integration of fuzzy sets theory and Ant System-based Clustering Algorithm (ASCA) in order to enhance the accuracy and speed in case matching. The cases in the case base are fuzzified in advance, and then grouped into several clusters by their own similarity with fuzzified ASCA. When a new case occurs, the system will find the closest group for the new case. Then the new case is matched using the fuzzy matching technique only by cases in the closest group. Through these two steps, if the number of cases is very large for the case base, the searching time will be dramatically saved. In the practical application, there is a diagnostic system for vehicle maintaining and repairing, and the results show a dramatic increase in searching efficiency.

Large neighborhood local search for the p-median problem

Abstract: In this paper we consider the well known p-median problem. We introduce a new large neighborhood based on ideas of S.Lin and B.W. Kernighan for the graph partition problem. We study the behavior of the local improvement and Ant Colony algorithms with new neighborhood. Computational experiments show that the local improvement algorithm with the neighborhood is fast and finds feasible solutions with small relative error. The Ant Colony algorithm with new neighborhood as a rule finds an optimal solution for computationally difficult test instances.

Optimal Capacitor Placement in Distribution Systems Employing Ant Colony Search Algorithm

Abstract: This article introduces an ant colony search algorithm (ACSA) to solve the optimal capacitor placement problem. This ACSA is a relatively new meta-heuristic for solving hard combinational optimization problems. It is a population-based approach that uses exploration of positive feedback as well as greedy search. The ACSA was inspired from the natural behavior of the ant colonies on how they find the food source and bring them back to their nest by building the unique trail formation. Therefore, through a collection of cooperative agents called ants, the near-optimal solution to the capacitor placement problem can be effectively achieved. In addition, in the algorithm, the state transition rule, local pheromone-updating rule, and global pheromone-updating rule are all added to facilitate the computation. Through operating the population of agents simultaneously, the process stagnation can be effectively prevented. Namely the optimization capability can thus be significantly enhanced. Moreover, the capacito...

Machine cell formation for cellular manufacturing systems using an ant colony system approach

Abstract: The aim of a cellular manufacturing system is to group parts that have similar processing needs into part families and machines that meet these needs into machine cells. This paper addresses the problem of grouping machines with the objective of minimizing the total cell load variation and the total intercellular moves. The parameters considered include demands for number of parts, routing sequences, processing time, machine capacities, and machine workload status. For grouping the machines, an ant colony system (ACS) approach is proposed. The computational procedure of the approach is explained with a numerical illustration. Large problems with up to 40 machines and 100 part types are tested and analyzed. The results of ACS are compared with the results obtained from a genetic algorithm (GA), and it is observed that its performance is better than that of GA.

An Ant Colony Algorithm Based on Grid Method for Mobile Robot Path Planning

Abstract: A bionics algorithm for robot path planning in static environment is proposed, in which the environmental models are established with grid method, the foraging behavior of ant colonies is simulated and optimal path search is finished by many ants cooperatively. Furthermore, the strategies of probabilistic search, nearest neighbor search and a goal guiding function are applied to enable the searching to be rapid and efficient. Results of simulation experiments demonstrate that the best path can be found in short time, real-time planning can be achieved, and the effect is very satisfying even if the geographic conditions with obstacles are exceedingly complicated.

Parallel ant colony optimization for 3D protein structure prediction using the HP lattice model

Abstract: The protein folding problem studies the way in which a protein - a chain of amino acids - will 'fold' into its natural state. Predicting the way in which various proteins fold can be fundamental in developing treatments of diseases such as Alzeihmers and Systic Fibrosis. Classical solutions to calculating the final conformation of a protein structure are resource-intensive. The Hydrophobic-Hydrophilic (HP) method is one way of simplifying the problem. We introduce a novel method of solving the HP protein folding problem in both two and three dimensions using ant colony optimizations and a distributed programming paradigm. Tests across a small number of processors indicate that the multiple colony distributed ACO (MACO) approach is scalable and outperforms single colony implementations.

Parallel Ant Colony Algorithms

Abstract: Ant colony algorithms are computational methods for solving problems that are inspired by the behaviour of real ant colonies. One particularly interesting aspect of the behaviour of ant colonies is that relatively simple individuals perform complicated tasks. Examples for such collective behavior are: i) the foraging behaviour that guides ants on short paths to their food sources, ii) the collective transport of food where a group of ants can transport food particles that are heavier than the sum of what all members of the group can transport individually, and iii) the brood sorting behavior of ants to place larvae and eggs into brood chambers of the nest that have the best environmental conditions. In this chapter we concentrate on the Ant Colony Optimization (ACO) metaheuristic for solving combinatorial optimization problems. ACO is inspired by the foraging behaviour of ants. An essential aspect thereby is the indirect communication of the ants via pheromones, i.e., chemical substances which are released into the environment and influence the behavior or the development of other individuals of the same species. In a famous biological experiment called double bridge experiment ([9, 23]) it was shown how trail pheromone lead ants along short paths to their food sources. In this experiment a double bridge with two branches of different lengths connected a nest of the Argentine ant with a food source. It was found that after a few minutes nearly all ants use the shorter branch. This is interesting because Argentine ants can not see very well. The explanation of this behavior has to do with the fact that the ants lay pheromone along their path. It is likely that ants which randomly chose the shorter branch arrive earlier at the food source. When they go back to the nest they smell some pheromone on the shorter branch

Defensive behavior of ants in a mutualistic relationship with aphids

Abstract: Mutualistic relationships between ants and aphids are well studied but it is unknown if aphid-attending ants place a greater relative importance on defending aphids from aphid-predators or from competing ant colonies. We tested the hypothesis that aphid-attending ants defend their aphids against aphid-predators more aggressively than against ants from neighboring colonies. We conducted introduction trials by placing an individual non-predatory insect, an aphid-predator, or a foreign conspecific ant on the leaf of a resident ant. We found that ants did not attack non-predatory insects, but did attack competing ants and aphid-predators. When we presented resident ants with both the threats (i.e., predator and competitor) at the same time, residents always attacked potential competitors as opposed to aphid-predators. We suggest this behavior may reduce the likelihood of raids by neighboring colonies. Ants appear to balance both the energetic costs of making an attack and the costs associated with losing aphids to a predator, against the benefits of signaling their defensive ability to rivals and/or preventing rivals from gaining knowledge of a potential food resource.

A model based on ant colony system and rough set theory to feature selection

Abstract: In this paper we propose a hybrid approach to feature selection based on Ant Colony System algorithm and Rough Set Theory. Rough Set Theory offers the heuristic function to measure the quality of a single subset. We have studied the influence of the setting of the parameters for this problem, in particular for finding reducts. Experimental results show this hybrid approach is a promising method for features selection.

A hybrid ant colony system for vehicle routing problem with time windows

Abstract: The Vehicle Routing Problem with Time Windows (VRPTW) is an important problem occurring in many logistics systems. The objective of VRPTW is to serve a set of customers within their predefined time windows at minimum cost. Ant Colony System algorithm (ACS) that is capable of searching multiple search areas simultaneously in the solution space is good in diversification. On the other hand, Simulated Annealing algorithm (SA) is a local search technique that has been successfully applied to many NP-hard problems. A hybrid algorithm (IACS-SA) that combines an improved ACS with SA is proposed in this paper. The algorithm has been tested on 56 Solomon benchmark problems. The results show that our IACS-SA is competitive with other meta-heuristic approaches in the literature. The results also indicate that such a hybrid algorithm outperforms the individual heuristic alone.