Case Studies - Energy & Environment

The Problem
Utilities in the UK are under constant pressure from their regulators to improve their operational efficiency whilst maintaining quality and infrastructure condition. Our client, a Top 10 Water Company, had recently responded by introducing a major change in maintenance policy, reducing the total maintenance budget whilst putting more effort into the most critical pumping stations.

The client wished to know whether the change in policy was affecting the reliability of the pumping stations, and whether there was any evidence of deterioration in the performance of the assets.

The Solution
We provided a detailed analysis of trends in pumping station reliability and performance for the periods before and after the policy change. The analysis used multiple segmentation, by zone, size of pumping station and degree of criticality.

It provided clear evidence that the policy had been implemented successfully across all zones, with no change in the overall levels of reliability, power consumption or performance.

The Approach
We performed a two month study working closely with the client to understand the data and interpret the results.

The client provided 20 data sets from disparate data and information warehouses. The quality and format of the data sets were varied, and much of the data had previously been thought unusable.

We reformatted the data and cleansed it to remove duplicate records and add missing data. We then developed a series of statistical models to analyse the data. The outputs included trends of scheduled and unscheduled maintenance and time series analyses of power consumption against rainfall.

We used the DEX^TM software tool to carry out the analysis. DEX is a uniquely powerful data analysis and modelling environment, which we originally developed to support the military.

The Benefits
Our analysis gave the client confidence that their change in policy is working, with no detrimental effects observed.

The data cleansing and statistical modelling also provided a powerful new suite of tools for analysing performance down to individual site level.

In addition, the study identified a number of areas for improvement in data management and working practices. For example the financial data and work history data were found to have no common basis, with no way of linking them.