Case Study: Anglian Water
Background
Anglian Water is a water company that operates in the East of England supplying water and sewage treatment.
They work with multiple alliance partners proving water services.
Situation/Challenge
Every month, each partner sends detailed lists of work performed and assets utilised. This information is then used to produce the Work Payment schedule and forms the basis for payment of the partners.
Due to their outdated systems, it was deemed unrealistic to request access to partner back office systems for data extraction, so they were supplied with guidelines for file format and content.
This led to too much manual processing. This produced many errors in the received files and were quite often formatted differently from month to month – meaning that the system was constantly being patched to cope with incorrectly formatted data.
The system became so complex that all previous developers refused to continue. Lack of user testing missed defects and a poor choice of business intelligence database system compounded the problem.
Solution
The poor quality of data being received was the major problem with the current system.
We produced an AI engine to interpret the supplied files and upload content based on Anglian Water’s project, asset and attribute lists – even when the supplied information was outside the scope of the requested file format.
Trial imports showing any unmatched data and reported by exception.
Successful imports were used as a feed by the Business Intelligence system.
Benefits
Almost 60% of the current work had to be re-coded, but this resulted in major time savings and provided lower on-going support costs as the manual processes associated with file processing were not required.
Each partner file was originally taking an average of 2.5 days to upload due to verification, validation and data manipulation tasks. We reduced this to hours, saving 0.75 of a person per month and facilitating the timely reporting of project data.
Before our updates only 15% of attribute names were being matched – we increased this to 97%.
Subsequently, we were asked to provide a critique of the project along with suggested improvements on design decisions, integration with external systems, software development standards and best practices.
Let’s Start A Project
DiggleNet Ltd
Unit 3, Sovereign Way, Trafalgar Industrial Estate,
Downham Market, Norfolk. PE38 9SW
