Mitek Analytics make advanced analytics for Industrial Internet easy. We unlock business value of machines with networked sensors and software through integrated analytical software applications.
We analyze customer data and generate analytical reports. Billions of dollars are spent on Business Intelligence (BI) software projects that also require a substantial customer effort. For many industrial internet applications, Mitek Analytics can provide the end result – a BI report – at a fraction of a typical BI software project cost, as a service.
We processed tens of terabytes of machine data to generate portable hyperlinked report documents with executive dashboards and drill-down to detailed actionable reports. Our BI reports contain predictive analytics results and help to uncover the business value latent in your machine data. Full service BI report does not require the customers to have data science capability. If such capability is available, it can be incorporated through a collaborative development of the BI.
As an example, Mitek Analytics processed aircraft and jet engine data collected over three years and million flights from flight data recorders for a commercial airline fleet. The 10Tb data were distilled into 1GB of hyperlinked HTML documents. The executive fleet-level reports allow drill-down access to unit (aircraft-level) reports, record (single-flight-level) reports of anomalies and specific data channel problems. The reports provided visibility into fuel efficiency, safety, and data quality issues at fleet scale.
Mitek Analytics provides consulting in analytics application architecture. We work in close collaboration with the customers and use proven tools of management consulting, requirement analysis, and systems engineering. The analytics architecture can include both IT and OT (operations technology) applications.
Machine analytics applications are mission critical and require high confidence analytics. Most data science groups use machine learning and data mining methods for low-confidence social network, web advertising, and marketing applications. Proven analytics approaches for industrial internet applications rely on engineering models. The models have explicit engineering meaning and allow predictive simulations. The model parameters are identified (calibrated) using the historical data. The model-based data exploitation analytics include fleet monitoring, reliability, risk analytics, model identification, predictive trending, anomaly detection, fault isolation, forecasting, decision support, optimization, feedback control, distributed control, fault tolerance and redundancy management, and many others.
As one example, NASA contracted Mitek Analytics in a 4-year project to help establish analytics architecture for Integrated Vehicle Health Management (IVHM). In IVHM, the vehicle (such as aircraft, launch vehicle, or space vehicle) data are used to evaluate system health and provide optimized response to the detected anomalies. Mitek Analytics team included Boeing, GE, and Honeywell and used DoD Architecture Framework (DODAF) to formulate the high-level application architecture. The IT architecture was prototyped using open-source tools and followed service-oriented architecture (SOA) design.
Other examples include analytics applications architecture for a regional microgrid project with distributed energy resources (DER) for a US electrical utility and analytics application architecture for a high-tech startup. We developed analytics architectures for predictive monitoring of industrial turbomachinery fleet, for advanced large-scale distributed control in process industry, and for space systems.
Our solutions are based on cutting edge research in Data Science for industrial (machine) data. Some of the research can be accessed through the resources included below.