The MIIS Eprints Archive: No conditions. Results ordered -Date Deposited. 2019-09-17T16:48:18ZEPrintshttp://www.maths-in-industry.org/images/sitelogo.gifhttp://www.maths-in-industry.org/miis/2005-05-26Z2015-05-29T19:45:42Zhttp://www.maths-in-industry.org/miis/id/eprint/30This item is in the repository with the URL: http://www.maths-in-industry.org/miis/id/eprint/302005-05-26ZReal-time traffic monitoring using mobile phone dataEddie WilsonRE.Wilson@bristol.ac.uk2005-05-26Z2015-05-29T19:45:44Zhttp://www.maths-in-industry.org/miis/id/eprint/31This item is in the repository with the URL: http://www.maths-in-industry.org/miis/id/eprint/312005-05-26ZMeasuring glucose content in the aqueous humorMany diabetics must measure their blood glucose levels regularly to maintain good health. In principle, one way of measuring the glucose concentration in the human body would be by measuring optically the glucose content of the aqueous humor in the eye. Lein Applied Diagnostics wish to assess whether this is feasible by a linear confocal scan with an LED source, or by supplementing such a system with other measurements.Maria AguarelesJohn Fozardfozard@maths.ox.ac.ukJens GravesenRob Hinchhinch@maths.ox.ac.ukPoul HjorthKaterina Kaourikaouri@maths.ox.ac.ukDavid ParkerMorten Willatzen2005-05-26Z2015-05-29T19:45:45Zhttp://www.maths-in-industry.org/miis/id/eprint/32This item is in the repository with the URL: http://www.maths-in-industry.org/miis/id/eprint/322005-05-26ZModelling of melt on spinning wheelsThermal Ceramics manufacture high-temperature thermal insulation, much of which is made in the form of fibre, for use in furnaces, aeroengines, domestic appliances, fire protection systems and other applications. In the manufacturing process, a melt stream emerges from a circular orifice and falls on to two successive spinning wheels, resulting in fiberisation. However, the process also results in a significant fraction of unfiberised material in the form of shot particles. The presence of shot tends to increase the thermal conductivity of the final product. Moreover, some areas of application, such as the automotive industry, require clean fibre with zero shot. Reducing the amount of shot that is produced in the fiberisation process would also reduce the costs of 'cleaning' the fibres for these markets. The Study Group was asked to look at how melt transfers on to the spinning wheels, what kind of melt layer exists on the wheels, and how this layer breaks up. By using a mathematical model to understand the relative importance of various parameters, it is hoped to guide experimental trials and then the scale-up to production levels.Chris Brewardbreward@maths.ox.ac.ukRosemary Dysondysonr@maths.ox.ac.ukCarina Edwardsedwardsc@maths.ox.ac.ukPaul MetcalfeColin PleaseMaxim Zyskin2005-05-26Z2015-05-29T19:45:46Zhttp://www.maths-in-industry.org/miis/id/eprint/33This item is in the repository with the URL: http://www.maths-in-industry.org/miis/id/eprint/332005-05-26ZUncertainty of flow in porous mediaThe problem posed to the Study Group was, in essence, how to estimate the probability distribution of f(x) from the probability distribution of x. Here x is a large vector and f is a complicated function which can be expensive to evaluate. For Schlumberger's applications f is a computer simulator of a hydrocarbon reservoir, and x is a description of the geology of the reservoir, which is uncertain.Joyce AitchisonLinda CummingsGiles RichardsonJonathan Rougier2005-05-26Z2015-05-29T19:45:48Zhttp://www.maths-in-industry.org/miis/id/eprint/34This item is in the repository with the URL: http://www.maths-in-industry.org/miis/id/eprint/342005-05-26ZDesign of microfluidic networksMicrofluidics is a relatively new and fast growing research area in fluid mechanics. The devices in question are thin wafers containing etched or printed interconnecting channels through which fluids are pumped, which can mix and/or react at various nodes to produce an output product. Microfluidic devices have applications in many manufacturing and chemical detection processes. For example, they can be used to manufacture monodisperse droplets with very well defined properties for pharmaceutical applications; or form the basis for miniaturised ‘lab-on-a-chip’ sensor arrays for detecting biological substances or toxins.John BillinghamAndrew GriefDavid LeppinenNick Ovenden2005-05-26Z2015-05-29T19:45:49Zhttp://www.maths-in-industry.org/miis/id/eprint/35This item is in the repository with the URL: http://www.maths-in-industry.org/miis/id/eprint/352005-05-26ZNational Air Traffic ServicesNational Air Traffic Services (NATS) are concerned with ensuring low probabilities of errors in determining aircraft positions. In general, error probabilities depend on the tails of some probability distributions for which there has been no theoretical model. Analysis of radar performance is regularly undertaken by NATS to ensure radar performance is within safety limits, with the maximum range being dependent on the declared separation between aircraft. NATS brought two questions to the Study Group, involving the horizontal (azimuthal) errors in radar data and the vertical errors in altimetry system data. In both cases, NATS asked the Study Group to analyse the data and assess whether the probability distributions that are currently used are good models for the errors.Chris BuddAndrew HillAndre Andre Leger2005-05-26Z2015-05-29T19:45:50Zhttp://www.maths-in-industry.org/miis/id/eprint/37This item is in the repository with the URL: http://www.maths-in-industry.org/miis/id/eprint/372005-05-26ZData packet loss in a queue with limited bufferWhen channels with limited buffer space are shared by multiple bursty data sources then some fraction of submitted packets is lost when the buffer becomes full. Feedback signals enable the sources to reduce their burst sizes when the buffer becomes congested. On the other hand, if there is plenty of free space in the buffer then burst sizes are allowed to increase. Such systems may be designed with a variety of queueing disciplines and packet discard policies. The Study Group was asked to analyse the packet-loss process for packet submission using either first-in-first-out or weighted fair queueing, and for packet discard using either drop-tail or random early discard policies.David Allwrightallwrigh@maths.ox.ac.uk2005-05-26Z2015-05-29T19:45:52Zhttp://www.maths-in-industry.org/miis/id/eprint/38This item is in the repository with the URL: http://www.maths-in-industry.org/miis/id/eprint/382005-05-26ZModels of consumer behaviourThe problem posed to the Study Group was to construct models for consumer behaviour that might be useful in tools for brand management in markets for fast-moving consumer goods. Such models must take into account various psychological and sociological factors that describe respectively how consumers are influenced by what is on sale and who else is already buying. The outputs of the models should be predictions for the division of market share between competing products. Two phenomena of particular interest for assessing modelling options are the 'decoy effect' and 'lock-in'.Pablo CasasJon Chapmanchapman@maths.ox.ac.ukRobert HuntGregory KozyreffAndrew LaceyEmeline LarrieuRobert LeeseTiina Rooseroose@maths.ox.ac.ukDavid SchleyLydie Staron