RISK Management Technologies (RMT), a recognised leader globally in incident and risk management technologies, is expanding its industry leading First Priority platform to include an augmented predictive incident capability.
This means that for the first-time, data mining and machine learning across all internal and external datasets will occur providing companies with insights to forewarn them before significant and catastrophic incidents occur.
No one accepts significant or catastrophic events, yet they continue to occur.
Whether they are one-person fatalities or catastrophic events such as Fukushima, Piper Alpha, Air France 447, Samarco Dam Wall collapse, Deepwater Horizon, or Texas City, the reality remains current technology is incapable to foresee or forewarn despite the unequivocal human, environmental and financial consequences.
Given most significant and or catastrophic events are preceded by soft – hard warnings, why are we unable of seeing or hearing them?
Leveraging the dramatic advances in AI, RMT scientists have been able to completely re-interpret incident precursors by interrogating, iterating, and learning from multisource corporate data using recognition algorithms on a scale, depth and speed impossible to replicate.
Deep Learning, situational predictability and sentiment analysis can now provide the technological platform capable of “listening” within the “noisy” and high complex interdependent environments that high risk industries operate within.
Combine these technologies with the precious strategic assets of structured enterprise incident and risk data from a number of industry leaders that represent the accumulated experiences of their organisation and we have an unprecedented opportunity to benefit from deep learning predictive models for insights and actions.
Design of a user-friendly interface for best predictions data dashboards
Equally important to identifying these hidden relationships in the data and making predictions is to design a User Interface (UI) where users can easily interpret the results. RMT are currently investigating topological data analysis alongside 3-D modelling tools to complement the existing sophisticated dashboards available in First Priority.
Understanding these hidden relationships enables predictions to be made. RMT are now exploring the prediction of number and severity of events when the equivalent precursors are observed, with a measured accuracy.
With release planned in early 2020, RMT’s team of data scientists, mathematicians and software engineers are confident their technical platform of integrated, Random Forests, Recurrent Neural Networks and Deep Categorical Classifiers will significantly reduce the occurrence, severity and impact of significant and catastrophic events in high risk industries.