The project brief was to enable all inbound post to be scanned, categorised, matched to its originating case and imported into the Solcase case management system.
Findings
While Solcase does include some scanning functionality, it is not built to handle bulk scanning. We designed and built a complete system to streamline this process which handles over a thousand items of post each morning.
A C# windows service written in Visual Studio 2005 running on the server picks up each inbound scan and creates an entry in the SQL Server 2005 workflow database.
Users access the scans via an ASP.NET 3.5 intranet website which is optimised for volume work. Here the scans get categorised according to their document type and move on in the workflow. The next step requires the scans to be matched to their cases, and it's here that there was a problem: the Progress database does not provide adequately fast access to its "word indexes" via ODBC. We developed an database for the required data optimised for searchability in Microsoft SQL Server and we populate it overnight using a SQL Server Integration Services (SSIS) package. This SQL database gives us sub-second response times for all combinations of search criteria and is a critical success factor for the project.
Some categories of data (such as mortgage redemption statements) undergo further workflow steps in which additional data is acquired and stored in the workflow system before it is finally read back into the Solcase case management system, the document attached to the case and the case scheduled for further work by a user.
This project has proved transformational for the business, enabling a "scan and destroy" policy for handling inbound post. The further implications of this are that all case data is now electronic, so case workers can work from home without need for paper files and files & documents never get lost. Overall this is a great improvement in both quality and cost.