Lyndon While

An evolution strategy with probabilistic mutation for multi-objective optimisation

Evolutionary algorithms have been applied with great success to the difficult field of multiobjective optimisation. Nevertheless, the need for improvements in this field is still strong. We present a new evolutionary algorithm, ESP (the Evolution Strategy with Probabilistic mutation). ESP extends traditional evolution strategies in two principal ways: it applies mutation probabilistically in a GA-like fashion, and it uses a new hyper-volume based, parameterless, scaling independent measure for resolving ties during the selection process. ESP outperforms the state-of-the-art algorithms on a suite of benchmark multiobjective test functions using a range of popular metrics.

An adaptive learning model for simplified poker using evolutionary algorithms

Evolution is the process of adapting to a potentially dynamic environment. By utilising the implicit learning characteristic of evolution in our algorithms, we can create computer programs that learn, and evolve, in uncertain environments. We propose to use evolutionary algorithms to learn to play games of imperfect information-in particular, the game of poker. We describe a new adaptive learning model using evolutionary algorithms that is suitable for designing adaptive computer poker players. We identify several important principles of poker play and use these as the basis for a hypercube of evolving populations in our model. We report experiments using this model to learn a simplified version of poker; results indicate that our new approach demonstrates emergent adaptive behaviour in evolving computer poker players. In particular, we show that our evolving poker players develop different techniques to counteract the variety of strategies employed by their opponents in order to maximise winnings. We compare the strategies evolved by our evolved poker players with a competent static player to demonstrate the importance of adaptation to achieve this end. Comparison with our existing evolutionary poker model highlights the improved performance of this approach.

Throughput maximization of queueing networks with simultaneous minimization of service rates and buffers.

The throughput of an acyclic, general-service time queueing network was optimized, and the total number of buffers and the overall service rate was reduced. To satisfy these conflicting objectives, a multiobjective genetic algorithm was developed and employed. Thus, our method produced a set of efficient solutions for more than one objective in the objective function. A comprehensive set of computational experiments was conducted to determine the efficacy and efficiency of the proposed approach. Interesting insights obtained from the analysis of a complex network may assist practitioners in planning general-service queueing networks.

A multi-objective evolutionary algorithm approach for crusher optimisation and flowsheet design

The performance of crushing equipment in mineral processing circuits is often critical to the generation of final product. A multiobjective evolutionary algorithm has been developed that allows the crusher internal geometry to be created and evaluated against multiple performance objectives. The multiple-objective approach is particularly important in mineral processing, as the optimum performance of single machines is often a trade-off between competing process drivers. A case study is presented that demonstrates the application of the technique to the design of cone crusher liners. New crusher liner profiles resulting from the application of the evolutionary algorithm suggest that significant improvements in the generation of lump product can be obtained. The extension of the approach to wider process plant design is discussed in terms of the objectives and issues to be addressed. 2004 Elsevier Ltd. All rights reserved.

Cyclic scheduling for an ethylene cracking furnace system using diversity learning teaching-learning-based optimization

Abstract The ethylene cracking furnace system is central to an olefin plant. Multiple cracking furnaces are employed for processing different hydrocarbon feeds to produce various smaller hydrocarbon molecules, such as ethylene, propylene, and butadiene. We develop a new cyclic scheduling model for a cracking furnace system, with consideration of different feeds, multiple cracking furnaces, differing product prices, decoking costs, and other more practical constraints. To obtain an efficient scheduling strategy and the optimal operational conditions for the best economic performance of the cracking furnace system, a diversity learning teaching-learning-based optimization (DLTLBO) algorithm is used to simultaneously determine the optimal assignment of multiple feeds to different furnaces, the batch processing time and sequence, and the optimal operational conditions for each batch. The performance of the proposed scheduling model and the DLTLBO algorithm is illustrated through a case study from a real-world ethylene plant: experiments show that the new algorithm out-performs both previous studies of this set-up, and the basic TLBO algorithm.

Super 14 Rugby Fixture Scheduling Using a Multi-Objective Evolutionary Algorithm

Super 14 Rugby is not only a popular game, but also a hugely profitable business. However, determining a schedule for games in the competition is very difficult, as a number of different, often conflicting, factors must be considered. We propose the use of a multi-objective evolutionary algorithm for deciding such a schedule. We detail the technical details needed to apply a multi-objective evolutionary algorithm to this problem and report on experiments that show the effectiveness of this approach. We compare solutions found by our approach with recent fixtures employed by the organising authority; our results showing significant improvements over the existing solutions

Improving the IWFG algorithm for calculating incremental hypervolume

We describe an optimised version of the incremental hypervolume algorithm IWFG that achieves new levels of performance for this class of algorithm. The principal changes are the use of an adaptive slicing scheme that works well both for points that need to be fully-evaluated, and for those that need only a small amount of evaluation; and the incorporation of an existing heuristic for ordering objectives independently for each point. The new algorithm can process in substantially less than a second sets containing a thousand points in 10-13 objectives, with much typical data; it is therefore a significant contribution to optimisers that use incremental hypervolume for selection, archiving, or diversity.

Scheduling English Football Fixtures: Consideration of Two Conflicting Objectives

In previous work the distance travelled by UK football clubs, and their supporters, over the Christmas/New Year period was minimised. This is important as it is not only a holiday season but, often, there is bad weather at this time of the year. Whilst searching for good quality solutions for this problem, various constraints have to be respected. One of these relates to clashes, which measures how many paired teams play at home on the same day. Whilst the supporters have an interest in minimising the distance they travel, the police also have an interest in having as few pair clashes as possible. This is due to the fact that these fixtures are more expensive, and difficult, to police. However, these two objectives (minimise distance and minimise pair clashes) conflict with one another in that a decrease in one intuitively leads to an increase in the other. This chapter explores this question and shows that there are compromise solutions which allow fewer pair clashes but does not statistically increase the distance travelled. We present a detailed set of computational experiments, on datasets covering seven seasons. We conclude that it is sometimes possible to reduce the number of pair clashes whilst not significantly increasing the overall distance that is travelled.

Improving and extending the HV4D algorithm for calculating hypervolume exactly

We describe extensions to the 4D hypervolume algorithm HV4D that greatly improve its performance in 4D, and that enable an extension of the algorithm to 5D. We add a facility to cope with dominated points, reducing the number of contribution calculations required; and a new representation of the front between slices, eliminating significant repeated work. The former also allows the algorithm to work efficiently with 5D data. The new algorithms can process sets containing 1,000 points in around 1 ms in 4D, and around 5–10 ms in 5D. They make a significant contribution to the state-of-the-art.

Transistor Sizing Using Particle Swarm Optimisation

We describe an application of particle swarm optimisation to the problem of determining the optimal sizing of transistors in an integrated circuit. The algorithm minimises the total area of silicon utilised by a given circuit, whilst maintaining the propagation delay of the circuit within a hard limit. It assesses designs using the well-known circuit simulation engine SPICE, making allowance for the inability of SPICE to assess poorly-designed circuits within a reasonable timeframe. Experiments on three different types of circuits demonstrate that the algorithm is able to derive excellent designs for a range of problem instances, including several problems where the Monte Carlo method is unable to find any feasible solutions at all.