How do we deal with the increasing vast data from complex systems? Engineers used to lament that the sensors, telemetry and storage were all too limited for the tasks ahead in control and monitoring. That is no longer the case, but getting the best information and decisions from very rich data remains a challenge, and is a commercial and ethical imperative – to optimize decision making, and not least because storing the data costs money too. The presentation will examine how autonomy can be built-in to infrastructure monitoring, to get the best out of rich data for planning and scheduling interventions by integrating sources, rules and information. It will be illustrated with examples from track monitoring, train location, and overhead line equipment incidents.
Speaker: Professor Andrew Starr works in novel sensing, e-maintenance systems, and decision-making strategies. At Cranfield he is head of the Through – life Engineering Services Institute, and Director of Education for the School of Aerospace, Transportation and Manufacturing. He has led a wide range of technical and asset management projects, e.g. two current projects in CleanSky2, on gears for the fast rotorcraft (Avio Aero, Active Space Technologies) and bearings for the high-speed generator for the more-electric-aircraft (Thales, Schaeffler, Active Space Technologies. Projects are under way for Innovate UK in railway business systems, with Unipart et al, and advanced powder flow monitoring using novel AE measurement and signal processing for in-line real time control, with Procter & Gamble et al. In infrastructure, work is under way for Network Rail et al in autonomous work prioritization based on data analytics, cost modelling and genetic-algorithm-based planning & scheduling.
Andrew is a Chartered Engineer, a Fellow of the Institution of Mechanical Engineers, a Fellow of the Royal Aeronautical Society, a member of the British Institute of Non-destructive Testing, a member of the Institute of Asset Management, a member of the International Society for Condition Monitoring