The version presented here may differ from the published version or from the version of the record. Please see the repository URL above for details on accessing the published version and note that access may require a subscription.

Accurate and reliable forecasting plays a key role in the planning and designing of municipal water supply infrastructures. Recent studies related to water demand prediction have shown that water demand is driven by weather variables, but the results do not clearly show to what extent. The principal aim of this research was to better understand the effects of weather variables on water demand. Additionally, it aimed to offer an appropriate and reliable technique to predict municipal water demand by using the Gravitational Search Algorithm (GSA) and Backtracking Search Algorithm (BSA) with Artificial Neural Network (ANN). Moreover, eight weather factors were adopted to evaluate their impact on the water demand. The principal findings of this research are that the hybrid GSA-ANN (Agent=40) model is superior in terms of fitness function (based on RMSE) for yearly and seasonal phases. In addition, it is evidently clear from the findings that the GSA-ANN model has the ability to simulate both seasonal and yearly patterns for daily data water consumption.

Experiments on individual cells require a range of extremely precise tools to permit their selection, manipulation, stimulation and analysis. This is further complicated by the cells' sensitivity to their environment, meaning that such tools must also be very gentle (or at least very localised) to minimise the generation of artefacts. Optical tools provide ideal performance in a number of such roles, exhibiting high spatial and temporal selectivity while causing minimal non-specific effects. This review will focus upon the optical tools that have been developed for these purposes, ranging from optical trapping systems which provide a contact-free technique for the manipulation of micron-scale objects, through to a selection of the different optically-mediated cell membrane disruption methods available for lysis and/or delivery of material.Optical trapping techniques provide the means to manipulate matter at approximately a 1-100 µm scale, without requiring direct contact with the cell [1]. The use of infrared wavelengths minimises the amount of light absorbed by biological targets, while a range of light-sculpting approaches are available to generate a wide array of complex beams which can be dynamically modified. This means that cells (or micro-structured probes) can be directly manipulated, either to build arrays or to perform mechanical measurements of the properties of the cell membrane.However, to modulate or measure processes occurring beyond the cell membrane, a mechanism of controlled membrane rupture must be utilised. These can either be destructive, for selective lysis experiments, or reversible, to allow the introduction of material to stimulate responses with minimal disruption to the cell. Both approaches feature a range of modalities: for example, lysis can be induced by direct plasma formation using high-energy pulsed lasers to induce catastrophic damage to anything within a defined radius [2], or can be combined with electrical fields to provide lysis with single-cell resolution [3]. Similarly, photoporation can be accomplished directly with very high precision (although conditions must be finely tuned to minimise cell disruption), or in combination with materials with specific absorption characteristics to deliver similar effects using much longer wavelengths and commensurately lower cell damage [4].The theories underpinning these techniques will be discussed, and illuminated using examples of recent research to provide first-hand examples of their successful application. The advantages and disadvantages of each approach will be comprehensively debated, and directions of promising research will be presented to give insight into the tools and techniques likely to be available in the future.