Dec 2016
Gas Transmission
Cognitive Technology for Technical Standards
NIA_NGGT0104
Complete
Dec 2016
Jun 2017
National Grid Gas Transmission
Richard Waine, box.GT.innovation@nationalgrid.com
Click here to send a question to the contact.
Network Innovation Allowance
None
Gas Transmission Networks
£460,000.00
Cognitive Technology is a significant opportunity to revolutionise analytics of unstructured data and decision making across Knowledge spheres in Gas Transmission. The vision is for an intelligent system that can capture knowledge, expertise and experience supplied in any format. In turn, this will provide a powerful capability to inform decisions based on the best practice across a range of business knowledge spheres. A key element to achieving this vision is to identify suitable methodologies for interacting with cognitive technologies, in particular when capturing from individual experts within the business who provide the information verbally based on their own wealth of knowledge and prior experience.

This project will develop a pilot using the IBM Watson core engine, which will capture knowledge from 630 Technical Standards Documents and use cognitive techniques to provide intelligent, reasoned responses to queries. In addition, the process for Third Party Enquiries (known as SSW/22) will be included, to demonstrate the ability of cognitive technology to remove the manually intensive element of information interrogation and cross referencing as part of key business processes, and to reduce the reliance on technical experts with more common enquiries. Natural language conversational interfaces will be demonstrated that will allow untrained users, whether internal or external, to interact with appropriate National Grid business processes.

The Third Party Enquiry Process is required to ensure that third parties carrying out work in the vicinity of National Grid high pressure gas pipelines (above 7 bar gauge) and associated installations take appropriate measures to prevent damage. Any damage to a high-pressure gas pipeline or its coating can affect its integrity and can result in failure of the pipeline with potential serious hazardous consequences for individuals located in the vicinity of the pipeline if it were to fail. It is therefore essential that the procedures are complied with when working near to a high pressure, above 7 bar gauge, pipeline. Third Party Enquiries which are received require our team of experts to carefully consider each case, cross referencing a number of technical standards, sources of information and experience and expertise of colleagues in order to process each case. The demand on knowledge from a range of sources, requiring intelligent cognitive technology makes this process an ideal candidate for the pilot programme.

The objective of the project is to assess and demonstrate the application of cognitive technology in an Asset Engineering sphere of knowledge. The project will build a corpus of knowledge across National Grid’s Technical Standard Documents and Third Party Enquiry process (SSW/22), allowing the application of three tools from the IBM Watson suite.  The tools will be trialled to demonstrate how it can engage intuitively with its users to learn from training and feedback to discover insights across the knowledge repository. These insights will aide in decision making by removing human biases and providing intelligent evidence based inferences based on industry best practices. The project will produce a test environment of both the engineering knowledge and standards advisor that can be transferred to Business as Usual at the end of the project. A demonstration of the Conversation Bot will be produced to demonstrate the application of this technology in a Gas Transmission Stakeholder environment. The system development and training elements supporting delivery of the three use cases will enable the business to develop a value assessment tool. This will enable Gas Transmission to assess the potential of cognitive technology in the engineering knowledge sphere and support targeted future deployment to fully leveraging the value of the tool. 

Success will be delivery of the three stages stated above. This in turn will enable the testing of decision making based on the reasoning and recommendations of cognitive technology. The technology should enable the business to better leverage subject matter experts through reduction of low level standards enquiries; enhance the technical standards development process driving value through more agile responsive and consistency standards; significantly enhance the ability of technical standards users to interrogate technical standards; drive efficiencies in the third party process and demonstrate the potential to provide a conversation bot to external stakeholders. Further, this proof of concept will enable the business to understand how cognitive technology can bring benefit to knowledge spheres; identify areas for application and assess business value.

Longer term, this project will inform the wider vision for an intelligent system that can capture knowledge, expertise and experience supplied in a variety of formats, and to provide a powerful capability inform decisions based on the best practice across a range of business processes.