The efficiency of current cargo screening processes at sea and air ports is largely unknown as few benchmarks exists against which they could be measured. Some manufacturers provide benchmarks for individual sensors but we found no benchmarks that take a holistic view of the overall screening procedures and no benchmarks that take operator variability into account. Just adding up resources and manpower used is not an effective way for assessing systems where human decision-making and operator compliance to rules play a vital role. Our aim is to develop a decision support tool (cargo-screening system simulator) that will map the right technology and manpower to the right commodity-threat combination in order to maximise detection rates. In this paper we present our ideas for developing such a system and highlight the research challenges we have identified. Then we introduce our first case study and report on the progress we have made so far.

Previous work has shown that robot navigation systems that employ an architecture based upon the idiotypic network theory of the immune system have an advantage over control techniques that rely on reinforcement learning only. This is thought to be a result of intelligent behaviour selection on the part of the idiotypic robot. In this paper an attempt is made to imitate idiotypic dynamics by creating controllers that use reinforcement with a number of different probabilistic schemes to select robot behaviour. The aims are to show that the idiotypic system is not merely performing some kind of periodic random behaviour selection, and to try to gain further insight into the processes that govern the idiotypic mechanism. Trials are carried out using simulated Pioneer robots that undertake navigation exercises. Results show that a scheme that boosts the probability of selecting highly-ranked alternative behaviours to 50% during stall conditions comes closest to achieving the properties of the idiotypic system, but remains unable to match it in terms of all round performance.

The analysis of system calls is one method employed by anomaly detection systems to recognise malicious code execution. Similarities can be drawn between this process and the behaviour of certain cells belonging to the human immune system, and can be applied to construct an artificial immune system. A recently developed hypothesis in immunology, the Danger Theory, states that our immune system responds to the presence of intruders through sensing molecules belonging to those invaders, plus signals generated by the host indicating danger and damage. We propose the incorporation of this concept into a responsive intrusion detection system, where behavioural information of the system and running processes is combined with information regarding individual system calls.

In this paper, we investigate output accuracy for a Discrete Event Simulation (DES) model and Agent Based Simulation (ABS) model. The purpose of this investigation is to find out which of these simulation techniques is the best one for modelling human reactive behaviour in the retail sector. In order to study the output accuracy in both models, we have carried out a validation experiment in which we compared the results from our simulation models to the performance of a real system. Our experiment was carried out using a large UK department store as a case study. We had to determine an efficient implementation of management policy in the store's fitting room using DES and ABS. Overall, we have found that both simulation models were a good representation of the real system when modelling human reactive behaviour.

Many probabilistic models introduce strong dependencies between variables using a latent multivariate Gaussian distribution or a Gaussian process. We present a new Markov chain Monte Carlo algorithm for performing inference in models with multivariate Gaussian priors. Its key properties are: 1) it has simple, generic code applicable to many models, 2) it has no free parameters, 3) it works well for a variety of Gaussian process based models. These properties make our method ideal for use while model building, removing the need to spend time deriving and tuning updates for more complex algorithms.

We consider a living organism as an observer of the evolution of its environment recording sensory information about the state space X of the environment in real time. Sensory information is sampled and then processed on two levels. On the biological level, the organism serves as an evaluation mechanism of the subjective relevance of the incoming data to the observer: the observer assigns excitation values to events in X it could recognize using its sensory equipment. On the algorithmic level, sensory input is used for updating a database, the memory of the observer whose purpose is to serve as a geometric/combinatorial model of X, whose nodes are weighted by the excitation values produced by the evaluation mechanism. These values serve as a guidance system for deciding how the database should transform as observation data mounts. We define a searching problem for the proposed model and discuss the model's flexibility and its computational efficiency, as well as the possibility of implementing it as a dynamic network of neuron-like units. We show how various easily observable properties of the human memory and thought process can be explained within the framework of this model. These include: reasoning (with efficiency bounds), errors, temporary and permanent loss of information. We are also able to define general learning problems in terms of the new model, such as the language acquisition problem.

In recent years genetic algorithms have emerged as a useful tool for the heuristic solution of complex discrete optimisation problems. In particular there has been considerable interest in their use in tackling problems arising in the areas of scheduling and timetabling. However, the classical genetic algorithm paradigm is not well equipped to handle constraints and successful implementations usually require some sort of modification to enable the search to exploit problem specific knowledge in order to overcome this shortcoming. This paper is concerned with the development of a family of genetic algorithms for the solution of a nurse rostering problem at a major UK hospital. The hospital is made up of wards of up to 30 nurses. Each ward has its own group of nurses whose shifts have to be scheduled on a weekly basis. In addition to fulfilling the minimum demand for staff over three daily shifts, nurses' wishes and qualifications have to be taken into account. The schedules must also be seen to be fair, in that unpopular shifts have to be spread evenly amongst all nurses, and other restrictions, such as team nursing and special conditions for senior staff, have to be satisfied. The basis of the family of genetic algorithms is a classical genetic algorithm consisting of n-point crossover, single-bit mutation and a rank-based selection. The solution space consists of all schedules in which each nurse works the required number of shifts, but the remaining constraints, both hard and soft, are relaxed and penalised in the fitness function. The talk will start with a detailed description of the problem and the initial implementation and will go on to highlight the shortcomings of such an approach, in terms of the key element of balancing feasibility, i.e. covering the demand and work regulations, and quality, as measured by the nurses' preferences. A series of experiments involving parameter adaptation, niching, intelligent weights, delta coding, local hill climbing, migration and special selection rules will then be outlined and it will be shown how a series of these enhancements were able to eradicate these difficulties. Results based on several months' real data will be used to measure the impact of each modification, and to show that the final algorithm is able to compete with a tabu search approach currently employed at the hospital. The talk will conclude with some observations as to the overall quality of this approach to this and similar problems.

Intelligent agents offer a new and exciting way of understanding the world of work. Agent-Based Simulation (ABS), one way of using intelligent agents, carries great potential for progressing our understanding of management practices and how they link to retail performance. We have developed simulation models based on research by a multi-disciplinary team of economists, work psychologists and computer scientists. We will discuss our experiences of implementing these concepts working with a well-known retail department store. There is no doubt that management practices are linked to the performance of an organisation (Reynolds et al., 2005; Wall & Wood, 2005). Best practices have been developed, but when it comes down to the actual application of these guidelines considerable ambiguity remains regarding their effectiveness within particular contexts (Siebers et al., forthcoming a). Most Operational Research (OR) methods can only be used as analysis tools once management practices have been implemented. Often they are not very useful for giving answers to speculative 'what-if' questions, particularly when one is interested in the development of the system over time rather than just the state of the system at a certain point in time. Simulation can be used to analyse the operation of dynamic and stochastic systems. ABS is particularly useful when complex interactions between system entities exist, such as autonomous decision making or negotiation. In an ABS model the researcher explicitly describes the decision process of simulated actors at the micro level. Structures emerge at the macro level as a result of the actions of the agents and their interactions with other agents and the environment. 3 We will show how ABS experiments can deal with testing and optimising management practices such as training, empowerment or teamwork. Hence, questions such as "will staff setting their own break times improve performance?" can be investigated.

This paper reports on continuing research into the modelling of an order picking process within a Crossdocking distribution centre using Simulation Optimisation. The aim of this project is to optimise a discrete event simulation model and to understand factors that affect finding its optimal performance. Our initial investigation revealed that the precision of the selected simulation output performance measure and the number of replications required for the evaluation of the optimisation objective function through simulation influences the ability of the optimisation technique. We experimented with Common Random Numbers, in order to improve the precision of our simulation output performance measure, and intended to use the number of replications utilised for this purpose as the initial number of replications for the optimisation of our Crossdocking distribution centre simulation model. Our results demonstrate that we can improve the precision of our selected simulation output performance measure value using Common Random Numbers at various levels of replications. Furthermore, after optimising our Crossdocking distribution centre simulation model, we are able to achieve optimal performance using fewer simulations runs for the simulation model which uses Common Random Numbers as compared to the simulation model which does not use Common Random Numbers.

We propose a novel algorithm for greedy forward feature selection for regularized least-squares (RLS) regression and classification, also known as the least-squares support vector machine or ridge regression. The algorithm, which we call greedy RLS, starts from the empty feature set, and on each iteration adds the feature whose addition provides the best leave-one-out cross-validation performance. Our method is considerably faster than the previously proposed ones, since its time complexity is linear in the number of training examples, the number of features in the original data set, and the desired size of the set of selected features. Therefore, as a side effect we obtain a new training algorithm for learning sparse linear RLS predictors which can be used for large scale learning. This speed is possible due to matrix calculus based short-cuts for leave-one-out and feature addition. We experimentally demonstrate the scalability of our algorithm and its ability to find good quality feature sets.