This paper reports on on-going research in the design of choice function hyperheuristics, the modelling of general-purpose low level heuristics, and the exploitation of parallel computing platforms for hyperheuristics

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.

Abstract-This paper investigates a tabu assisted genetic algorithm based hyper-heuristic (hyper-TGA) for personnel scheduling problems. We recently introduced a hyper-heuristic genetic algorithm (hyper-GA) with an adaptive length chromosome which aims to evolve an ordering of low-level heuristics in order to find good quality solutions to given problems. The addition of a tabu method, the focus of this paper, extends that work. The aim of adding a tabu list to the hyper-GA is to indicate the efficiency of each gene within the chromosome. We apply the algorithm to a geographically distributed training staff and course scheduling problem and compare the computational results with our previous hyper-GA. 1.

Hyper-GA was introduced by the authors as a genetic algorithm based hyperheuristic which aims to evolve an ordering of low-level heuristics so as to find a good quality solution to a given problem. The adaptive length chromosome hyper-GA, let's call it ALChyper-GA, is an extension of the authors previous work, in which the chromosome was of fixed length. The aim of a variable length chromosome is two fold; 1) it allows dynamic removal and insertion of heuristics 2) it allows the GA to find a good chromosome length which could otherwise only be found by experimentation. We apply the ALChyper-GA to a geographically distributed training staff and courses scheduling problem, and report that good quality solution can be found. We also present results for four versions of the ALChyper-GA, applied to five test data sets.

Hyperheuristics can be defined to be heuristics which choose between heuristics in order to solve a given optimisation problem. A number of hyperheuristics have been developed over the past few years. Here we propose a new hyperheuristic framework within which heuristics compete against one another. The rules for competition are motivated by the principles of reinforcement learning. We analyse the differences between a previously published choice function hyperheuristic and the new hyperheuristic. We demonstrate how the new hyperheuristic can make further improvements when a number of features are incorporated, including a dynamic tabu list which forbids the use of certain heuristics at certain times. The result is an algorithm which is competitive with the choice function hyperheuristic when applied to a comprehensive suite of nurse scheduling problems at a major UK hospital, featuring a wide variety of solution landscapes.

A hyperheuristic is a high-level heuristic which adaptively chooses between several low-level knowledge-poor heuristics so that while using only cheap, easy-to-implement low-level heuristics, we may achieve solution quality approaching that of an expensive knowledge-rich approach, in a reasonable amount of CPU time. For certain classes of problems, this generic method has been shown to yield high-quality practical solutions in a much shorter development time than that of other approaches such as tabu search and genetic algorithms, and using relatively little domain-knowledge. Hyperheuristics have previously been successfully applied by the authors to two real-world problems of personnel scheduling. In this paper, a hyperheuristic approach is used to solve 52 instances of an NP-hard nurse scheduling problem occuring at a major UK hospital. Compared with tabu-search and genetic algorithms, which have previously been used to solve the same problem, the hyperheuristic proves to be as robust as the former and more reliable than the latter in terms of solution feasibility. The hyperheuristic also compares favourably with both methods in terms of ease-of-implementation of both the approach and the low-level heuristics used.

This paper investigates a tabu search based hyper-heuristic for solving examination timetabling problems. The hyper-heuristic framework uses a tabu list to monitor the performance of a collection of low-level heuristics and then make tabu heuristics that have been applied too many times, thus allowing other heuristics to be applied. Experiments carried out on examination timetabling datasets from the literature show that this approach is able to produce good quality solutions.

Abstract not found

Abstract. Hyper-heuristics are proposed as a higher level of abstraction as compared to the metaheuristics. Hyper-heuristic methods deploy a set of simple heuristics and use only nonproblem-specific data, such as, fitness change or heuristic execution time. A typical iteration of a hyper-heuristic algorithm consists of two phases: heuristic selection method and move acceptance. In this paper, heuristic selection mechanisms and move acceptance criteria in hyperheuristics are analyzed in depth. Seven heuristic selection methods, and five acceptance criteria are implemented. The performance of each selection and acceptance mechanism pair is evaluated on fourteen well-known benchmark functions and twenty-one exam timetabling problem instances. 1

A hyperheuristic is a high-level heuristic which adaptively controls the combination of several low-level knowledgepoor heuristics so that while using only cheap and easyto -implement low-level heuristics, we may achieve solution quality approaching that of an expensive knowledgerich approach. Hyperheuristics have been successfully applied by the authors to three real-world problems of personnel scheduling. In this paper, the low-level behaviour of the choice-function based hyperheuristic is investigated and compared with a range of other heuristics and hyperheuristics. We show that the choice-function hyperheuristic makes an effective and realistic combination of the lowlevel heuristics at hand. Furthermore the combination of the low-level heuristics is intelligently adapted to both the problem being solved and the region of the search space currently being explored.