Feb 2015
Gas Distribution
Predictive Analytics Part Two
NIA_NGN_120
Complete
Feb 2015
Nov 2016
Northern Gas Networks
Tony Pearson, Predictive Analytic's Leader
Click here to send a question to the contact.
Network Innovation Allowance
None
£932,400.00
  • To maximise learning and knowledge transfer, organisations invited to submit offers will cover a wide range of modelling types and approaches including open source, closed source, freeware, platform based and bespoke / custom modelling.

  • Development of structured data analysis models for up to six opportunities covering a wide range of business activities (detailed above) using a mix of different approaches and engaging with a variety of external partners

  • Transfer of knowledge and learning. Maximisation of this transfer is a key aspect of this exercise and so, rather than appointing organisations simply to develop solutions, NGN will dedicate internal resources to work alongside the selected partners, gain understanding not just of how the specific models function but also to gain deeper insights into structured analytics as a process, the strengths and weaknesses of various different approaches and how these vary across a wide range of business processes. Acquiring this learning, insight and understanding is the core of this project and will be shared on completion of the works.

The objectives of the project are as follows :-

  • Development of analytical models covering a wide variety of business activities
  • Understanding of the strengths and weaknesses of the available data and how this impacts on a variety of modelling approaches
  • Understanding the potential improvements in output benefits that a structured analytical approach can deliver compared with a “traditional” approach across a wide variety of business areas (including developing an understanding of where a structured analytical approach is unlikely to deliver significant improvements)
  • Compare and understand the strengths and weaknesses of a range of modelling approaches and techniques
  • Develop understanding as to how to effectively and efficiently engage with specialist external providers of modelling solutions
  • Transfer knowledge, learning and experience
The success of the project will be measured across Technical Success and Learning & Knowledge Transfer.

Technical Success

Across the opportunities explored:-

  • Were successful analytical models developed (i.e. a model developed using “training” data was able to model relationships using “test” data previously unseen by the model)?

  • Was their “success” at providing forecasts and insights able to be measured?

  • Were confidence intervals able to be produced for the models developed?

  • Were a variety of technical approaches tested and were the advantages / disadvantages of these assessed?

Learning & Knowledge Transfer

Was learning and knowledge successfully transferred to allow / improve the following:-

  • Improved technical understanding of structured data analytics

  • Appreciation of the different techniques, approaches and methodologies that can be employed

  • Ability to identify areas where structured analytical models and solutions could potentially deliver improvements compared with the traditional approaches

  • Ability to select the approach that best matched the opportunity, data and potential benefit.

  • Ability to engage with expert external providers in an informed, effective and efficient way.

  • Successful transfer of knowledge and learning to other organisations through the NIA process