Offline Learning with a Selection Hyper-Heuristic: An Application to Water Distribution Network Optimisation.

A sequence-based selection hyper-heuristic with online learning is used to optimise 12 water distribution networks of varying sizes. The hyper-heuristic results are compared with those produced by five multiobjective evolutionary algorithms. The comparison demonstrates that the hyper-heuristic is a computationally efficient alternative to a multiobjective evolutionary algorithm. An offline learning algorithm is used to enhance the optimisation performance of the hyper-heuristic. The optimisation results of the offline trained hyper-heuristic are analysed statistically, and a new offline learning methodology is proposed. The new methodology is evaluated, and shown to produce an improvement in performance on each of the 12 networks. Finally, it is demonstrated that offline learning can be usefully transferred from small, computationally inexpensive problems, to larger computationally expensive ones, and that the improvement in optimisation performance is statistically significant, with 99% confidence.

A statistical analysis on the effect of preceding dry weather on sewer blockages in South Wales.

Water and sewerage companies (WaSC) in the UK are under increasing pressures to improve customer satisfaction. The biggest cause for customer dissatisfaction in the wastewater sector is a service failure caused by a blockage. There is therefore a need to understand the factors which influence blockage processes in order to prevent them. This work demonstrates how preceding rainfall impacts the sewer system of two highly populated regions within South Wales that have differing gradients. The total rainfall, number of dry days and consecutive number of dry days prior to a blockage were investigated using statistical analysis in order to determine the impact that rainfall has on blockages. The results obtained demonstrate the importance that dry weather has on blockage rates in both steep and flat catchments. Future work will incorporate predicted rainfall impact into a proactive maintenance scheduling model.

Derivation and validation of a simple, accurate and robust prediction rule for risk of mortality in patients with Clostridium difficile infection.

BACKGROUND: Clostridium difficile infection poses a significant healthcare burden. However, the derivation of a simple, evidence based prediction rule to assist patient management has not yet been described. METHOD: Univariate, multivariate and decision tree procedures were used to deduce a prediction rule from over 186 variables; retrospectively collated from clinical data for 213 patients. The resulting prediction rule was validated on independent data from a cohort of 158 patients described by Bhangu et al. (Colorectal Disease, 12(3):241-246, 2010). RESULTS: Serum albumin levels (g/L) (P = 0.001), respiratory rate (resps /min) (P = 0.002), C-reactive protein (mg/L) (P = 0.034) and white cell count (mcL) (P = 0.049) were predictors of all-cause mortality. Threshold levels of serum albumin ≤ 24.5 g/L, C- reactive protein >228 mg/L, respiratory rate >17 resps/min and white cell count >12 × 10(3) mcL were associated with an increased risk of all-cause mortality. A simple four variable prediction rule was devised based on these threshold levels and when tested on the initial data, yield an area under the curve score of 0.754 (P < 0.001) using receiver operating characteristics. The prediction rule was then evaluated using independent data, and yield an area under the curve score of 0.653 (P = 0.001). CONCLUSIONS: Four easily measurable clinical variables can be used to assess the risk of mortality of patients with Clostridium difficile infection and remains robust with respect to independent data.

Bi-objective design-for-control for improving the pressure management and resilience of water distribution networks.

This paper investigates control and design-for-control strategies to improve the resilience of sectorized water distribution networks (WDN), while minimizing pressure induced pipe stress and leakage. Both evolutionary algorithms (EA) and gradient-based mathematical optimization approaches are investigated for the solution of the resulting large-scale non-linear (NLP) and bi-objective mixed-integer non-linear programs (BOMINLP). While EAs have been successfully applied to solve discrete network design problems for large-scale WDNs, gradient-based mathematical optimization methods are more computationally efficient when dealing with large search spaces associated with continuous variables in optimal network control problems. Considering the advantages of each method, we propose a sequential hybrid method for the optimal design-for-control of large-scale WDNs, where a refinement stage relying on gradient-based mathematical optimization is used to solve continuous optimal control problems corresponding to design solutions returned by an initial EA search. The proposed method is applied to compute the Pareto front of a bi-objective design-for-control problem for the operational network BWPnet, where we consider reopening closed connections between isolated supply areas. The results show that the considered design-for-control strategy increases the resilience of BWPnet while minimizing pressure induced leakage. Moreover, the refinement stage of the proposed hybrid method efficiently improves the coarse approximation computed by the initial EA search, returning a continuous and even Pareto front approximation.

Short-term forecasting of turbidity in trunk main networks.

Water discolouration is an increasingly important and expensive issue due to rising customer expectations, tighter regulatory demands and ageing Water Distribution Systems (WDSs) in the UK and abroad. This paper presents a new turbidity forecasting methodology capable of aiding operational staff and enabling proactive management strategies. The turbidity forecasting methodology developed here is completely data-driven and does not require hydraulic or water quality network model that is expensive to build and maintain. The methodology is tested and verified on a real trunk main network with observed turbidity measurement data. Results obtained show that the methodology can detect if discolouration material is mobilised, estimate if sufficient turbidity will be generated to exceed a preselected threshold and approximate how long the material will take to reach the downstream meter. Classification based forecasts of turbidity can be reliably made up to 5 h ahead although at the expense of increased false alarm rates. The methodology presented here could be used as an early warning system that can enable a multitude of cost beneficial proactive management strategies to be implemented as an alternative to expensive trunk mains cleaning programs.

Discovering gene networks with a neural-genetic hybrid.

Recent advances in biology (namely, DNA arrays) allow an unprecedented view of the biochemical mechanisms contained within a cell. However, this technology raises new challenges for computer scientists and biologists alike, as the data created by these arrays is often highly complex. One of the challenges is the elucidation of the regulatory connections and interactions between genes, proteins and other gene products. In this paper, a novel method is described for determining gene interactions in temporal gene expression data using genetic algorithms combined with a neural network component. Experiments conducted on real-world temporal gene expression data sets confirm that the approach is capable of finding gene networks that fit the data. A further repeated approach shows that those genes significantly involved in interaction with other genes can be highlighted and hypothetical gene networks and circuits proposed for further laboratory testing.

Artificial intelligence techniques for bioinformatics.

This review provides an overview of the ways in which techniques from artificial intelligence (AI) can be usefully employed in bioinformatics, both for modelling biological data and for making new discoveries. The paper covers three techniques: symbolic machine learning approaches (nearest neighbour and identification tree techniques), artificial neural networks and genetic algorithms. Each technique is introduced and supported with examples taken from the bioinformatics literature. These examples include folding prediction, viral protease cleavage prediction, classification, multiple sequence alignment and microarray gene expression analysis.