National Grid has begun to employ modern condition assessment techniques and data management tools which enable statistical analyses of current condition and predictive computational modelling of future condition. Through advanced processing of multiple data sources and extensive engagement with National Grid stakeholders, we have been able to provide the most complete and accurate view of fittings condition and defect occurrence at a network level. This data permits robust analyses regarding asset degradation and expected failure, which reduces uncertainty about asset condition-related risk and enables cost savings from better informed investment planning. Our analysis addressed:
- Fittings asset lives – through advanced data analysis and a structured SME elicitation workshop;
- Fittings deterioration model – which incorporates defect occurrence and is calibrated through a Bayesian approach;
- Asset Health Index (AHI) assignment and fittings renewal policy – based on predicted probability of defect from the deterioration model above.
Our work demonstrates that asset data has great value and can be utilised in conjunction with structured engineering input to drive asset management policies which achieve best whole life cycle value. This allows National Grid to make better investment decisions and achieve savings for the UK consumer while maintaining a high safety and reliability target, as set out in the RIIO regulatory framework.
Amey SC degradation model output. Cumulative probability of defect for spacers, by age and operating environment. 95% credible interval (HPD = Highest Posterior Density) represented as shaded bands.