Projects
Projects: Projects for Investigator |
||
| Reference Number | NIA_NGN_030 | |
| Title | Predictive Analytics | |
| Status | Completed | |
| Energy Categories | Fossil Fuels: Oil Gas and Coal(Oil and Gas, Refining, transport and storage of oil and gas) 100%; | |
| Research Types | Applied Research and Development 100% | |
| Science and Technology Fields | ENGINEERING AND TECHNOLOGY (Mechanical, Aeronautical and Manufacturing Engineering) 100% | |
| UKERC Cross Cutting Characterisation | Not Cross-cutting 100% | |
| Principal Investigator |
Project Contact No email address given Northern Gas Networks |
|
| Award Type | Network Innovation Allowance | |
| Funding Source | Ofgem | |
| Start Date | 01 December 2012 | |
| End Date | 01 April 2014 | |
| Duration | 16 months | |
| Total Grant Value | £220,000 | |
| Industrial Sectors | Technical Consultancy | |
| Region | Yorkshire & Humberside | |
| Programme | Network Innovation Allowance | |
| Investigators | Principal Investigator | Project Contact , Northern Gas Networks (100.000%) |
| Web Site | http://www.smarternetworks.org/project/NIA_NGN_030 |
|
| Objectives | Stage 1 is a proof of concept. Focusing on external escape reports, using existing easily available data from across the business. Success criteria for the Predictive Maintenance project have been identified: 1. Accurate/repeatable identification of when and where a gas leak is most likely to be reported. 2. Understanding of the main factors affecting a case. 3. Demonstration of the benefit of deploying predictive modelling to business areas to enable more effective fleet/resource planning. 4. Capability to take ownership of the data models created, as part of the NGN roll out of Predictive AnalyticsStage 2 Undertake Research & Development of full live data trial Select full area model to extend trial to use a more sophisticated workable model using: Use structured and unstructured data Internal and external data Apply those predictions to real-world constraints to optimise decisions Develop "What If" simulations to measure risks and benefits Refresh live models with new data to "learn" from past decisions. Financial Forecast Stage 1Based on our experience, the propensity models will take 35 days to complete. Presidion will deliver a report and repeatable outputs that can be immediately used as part of the NGN Predictive Analytics implementation. The challenge is to test if Predictive Analytics can outline, using available NGN data, the factors associated with reported gas leaks and predict where the next geographic location for a leak is most likely to be. Would such a model enable more effective deployment of the resources by the Area managers, and give valuable additional insights towards planning and decision making. Success criteria for the Predictive Maintenance project have been identified: Accurate/repeatable identification of when and where a gas leak is most likely to be reported Understanding of the main factors affecting a case Demonstration of the benefit of deploying predictive modelling to business areas to enable more effective fleet/resource planning Capability to take ownership of the data models created, as part of the NGN roll out of Predictive Analytics. | |
| Abstract | While NGN have an understanding of the existing reported gas escapes, they cannot forecast the number, type or locations of the next probable leaks which makes fleet management/resource planning difficult. Specifically this means putting the right resources/ equipment in the most likely locations to respond as quickly and effectively as possible to the incoming customer calls. NGN has the following performance targets within the current safety case: 85% of all escapes from the Network repaired within 7 days 98. 5% of all escapes from the Network repaired within 28 days Any Escapes older than 28 days, with exception reporting on the actions being taken and reasons for any escapes over 40 days old. At present, there are approximately 81% of cases closed within 7 days and 94% within 28 days. In order to hit the 7/28 target, NGN needs to consistently perform above 85% and 98. 5% respectively over the majority of the year to prevent underperformance annually. For the proof of concept stage, Predictive Maintenance Solutions from IBM will access multiple data sources in real time to predict equipment failure so the organization can avoid costly downtime and reduce maintenance costs. Driven by predictive analytics, these solutions can detect even minor anomalies and failure patterns to determine the areas that are at the greatest risk of failure. This early identification of issues will help NGN deploy limited maintenance resources more cost-effectively, maximize equipment uptime, and improve service levels for customers. Once proof of concept has been established, other areas of investigation and analysis will be identified so that a range of methods (and potentially partner organisations) can be used to identify the best solutions for a range of problems. Once suitable sample areas across a wide range of business activities have been identified, intensive investigation will be carried out on the available data to confirm its availability, quality, sufficiency, suitability for a range of analytical techniques / solutions and any data currently not currently available that would have a significant positive impact if it could be obtained.Note : Project Documents may be available via the ENA Smarter Networks Portal using the Website link above | |
| Publications | (none) |
|
| Final Report | (none) |
|
| Added to Database | 26/10/18 | |
