Predictive Analytics in IoT: Lessons Learnt from Aerospace

By Sophie Curtis on September 18, 2015

Original
The rapid spread of Internet of Things (IoT) technologies is producing a massive volume of sensor data, and growing exponentially each year. As the value for this kind of data is impacting multiple industries, more and more products are developed with sensors already installed – which newly founded company Senseye calls "smart assets".

Founded in late 2014, Senseye is an easy-to-use, online service that takes data from equipment, sensors and the environment to provide predictions and forecasts to help keep businesses one step ahead. By exploiting the Internet of Things and analytics technology, Senseye aims to deliver fast business insights without the hard work and expense that other products demand.

The CTO of Senseye, Robert Russell, has spent the last 15 years designing and deploying asset management and condition monitoring systems within the aerospace, defence and transport sectors. At the RE.WORK Future Technology Summit, Robert will present his vision for predictive analytics in the IoT sector, and how Senseye is using experience from other sectors to guide the implementation of that vision. We caught up with him ahead of the summit to hear more.

What was the motivation behind founding Senseye?
To bring predictive analytics to the masses in an affordable and easy-to-use package. Being tired of big, complex and costly systems, we saw a gap in the market to exploit the emergence of the IoT and build an a predictive analytics application for sectors and companies that have sensors and data, but do not have the knowledge on how to derive additional benefit. The product we envisaged needed to be simple to use and provide output that requires no interpretation.

What is the 'Internet of Assets'?
It is out vision for the connected world of business critical equipment that we will monitor and let the operator know about problems before they happen.

How is using experience from other sectors useful in the implementation of your vision?
Having worked in a number of sectors where the techniques have been applied in a bespoke manner has let us see the commonality in the problem and that what the end user wants to achieve with their assets is fundamentally the same, only the language and method of describing the problem are different. This has allowed us to identify other ways of approaching the problem of utilising sensor data to generate forecasts of failures. Rather than approaching the problem in the traditional way of fully understanding each individual machine type and building specific analytics, at Senseye our vision is to use machine learning in it truest form (i.e. it learns!) and have a solution that is sector and machine agnostic. By utilising user feedback we will quickly identify the data features that are interesting for the user and the sector to provide insights and forecasts.

What challenges have you met with the development and integration of Aerospace tech?
Most of the main challenges have not been technical, but related to organisational change and the demonstration of the return of investment in condition monitoring systems - as they are known in aerospace. Simple things like the level of human interaction required to gather and transfer data. These features need to be designed out. In many cases the user community is resistant to changes to working practices without seeing a direct benefit. A key challenge was to enable data sharing between organisation while ensuring trust and security in the data was retained.

What do you see in the future for Senseye?
In the short term the growth of an innovative and creative team with a passion for what we do. In the long term to be the “go-to” service for IoT predictive analytics. Between these there will be sector specific focus that will allow us to build the product and mature the vision.

Which areas do you feel could benefit from cross-industry collaboration?
There are no bounds to the benefits from collaboration. For the IoT to reach its full potential it is key that technology and data are open enough to be utilised across sectors. Examples being in Smart Cities, Health and Social Care, Transport and Utilities. All of these areas will need to make use of data from each other in order for the challenges of the next few decades to be overcome. We refer to this as needing a cross-vertical approach.

What new developments can we expect to see in IoT in the next 5 years?
IoT needs to become part of the background infrastructure in our lives and will develop in a way that makes it less intrusive and providing genuine benefits to all of society. Techniques will emerge for true machine to machine discoverability that will remove the need for human intervention in the deployment of new devices and services. A key aspect that I don’t believe has been cracked yet is how to build IoT business models that can address the question on how to monetise solutions to make them commercially viable. There are opportunities for virtual currencies to find their place.

Robert Russell will be speaking at the Future Technology Summit in London on 24-25 September. Other IoT speakers and exhibitors include Andrew Hudson-Smith, University College London; Eric Yeatman, Imperial College London; Toposens and 3DPlex. View the agenda here.

The Future Technology Summit is taking place alongside the Deep Learning Summit For more information and to register, please visit the event website here.
    
  

Machine Learning IoT Agtech Connected Devices Future Technology Summit Sensors Predictive Intelligence


0 Comments

    As Featured In

    Original
    Original
    Original
    Original
    Original
    Original

    Partners & Attendees

    Intel.001
    Nvidia.001
    Acc1.001
    Ibm watson health 3.001
    Rbc research.001
    Mit tech review.001
    Kd nuggets.001
    Facebook.001
    Graphcoreai.001
    Maluuba 2017.001
    Twentybn.001
    Forbes.001
    This website uses cookies to ensure you get the best experience. Learn more