Our Research

Data Visualisation for Process Insight and Decision Making

The proliferation of data owing to the onset of Industry 4.0 (I4.0) has led to many traditional data analysis approaches becoming redundant. Novel and innovative solutions are required to facilitate the new era of data-driven manufacturing characteristics of I4.0. This work demonstrates one such approach in the formation of a bespoke web-based visualisation and machine learning analytics platform, designed to bridge the gap between the old ways and the new. Our unique I4.0 data analytics platform, called Wiz 4.0, enables advanced big data analytics in conjunction with user-friendly features and multivariate data visualisations. This allows for both a holistic overview of manufacturing processes as well as a detailed analysis of data. Wiz 4.0 lays the foundations of an industry-defining software to grant deep insight into the inner relationships between process variables to the everyday user. The software provides the ability to analyse data using a variety of machine learning algorithms and plot the data in high dimensional space through the innovative no-code platform hosted on the Siemens MindSphere. This approach is set to revolutionise the value creation of data in the new IoT and smart factory paradigms emerging from the transition towards I4.0.

Read more here!

Predictive Maintenance in Industry 4.0 using Machine Learning

Poor maintenance regimes often contribute to unplanned downtime, quality defects and accidents, thus it is crucial to apply an effective maintenance strategy to achieve an efficient and safe process. Industry 4.0 has brought about a proliferation of digital data and with it, new opportunities to advance and improve the way maintenance activities are planned. Here, we propose a novel methodology that utilises machine learning to predict the maintenance of both faults and the repair time and uses this to underpin the scheduling of maintenance activities. This can be used to plan maintenance, optimising the schedule for cost within the constraints of labour availability and plant layout. When applied to simulated data, using a simulated Fischertechnik (FT) model, this methodology reduces overall plant maintenance costs by reducing unplanned downtime and increasing maintenance efficiency. This work provides a promising first step toward improving the way maintenance tasks are approached in Industry 4.0.

Physical Property Prediction in the Steel Industry

Prediction of relevant mechanical properties is vital for the steel industry and for materials engineering. This work investigates a novel method to predict the Jominy hardness curve of a single grade of steel based solely on the composition of the alloying elements. Further consideration is given to the optimisation of the chemical profile of the alloying elements with respect to the overall cost. The ultimate aim of the work is to be able to reduce spending on alloying elements by offering cheaper methods of achieving the same Jominy profile based upon accurate prediction. A feature auto-encoder neural network is employed to reduce the dimensionality of 19 alloying elements from some 1000 steel samples to extract relevant property information. Gaussian Process (GP) regression is used to predict the Jominy hardness profile of the steel up to a depth of 10mm. Results showed that a GP fed with extracted features was able to predict Jominy hardness profiles with high accuracy (RMSEunseen= 1.35, MAPE = 3.3%). Constrained optimisation using the GP model, and historical market data for the price of alloying elements showed that an average cost reduction of 18% on alloying element spend is possible. This work shows not only that novel machine learning approaches can be used to accurately predict mechanical properties of a given grade of steel, but crucially that use of these in conjunction with optimisation can generate considerable financial savings.