Service Quality - Service Process & Solutions
Service Process & Solutions
In IT system development projects, document artifacts have an important role as a form of communication. As a result, many documents such as use case documents and design documents are prepared and updated frequently in a large IT system development project. In such a situation, it is difficult to analyze and review many documents manually because documents are written in a project specific format and we can not apply analysis technologies easily.
To address such a situation, we developed the document modeling technology that extract logical structures from project specific formatted documents and a tool called IBM Document Analytics Designer and Runtime where we can generated customized analysis engine. By this tool, for various types of documents, we can apply the quality analysis such as the cross documents consistency checking and also apply a context-aware content analysis such as the use case description modeling and the test case generation from use case descriptions.
Our team is working on program comprehension, software test, and vulnerability detection based on the technologies (both theory and practice) of program analysis and formal method. Regarding program comprehension, we developed a technology for extracting business rules from COBOL programs, and applied it to an application code used in a real core insurance system. In the area of software test, we are now working on automated high-coverage test input data generation. These technologies contribute to reducing both cost and risk of legacy migration tasks, in which we transform a legacy system to a new system running on a new platform and/or architecture). In addition, we developed a new kind of string analysis for Rational AppScan, a Web application vulnerability scanner.
Service Ticket Analysis
Ticket analysis is ticket (service request) analysis service provided by research by an IBM unique tool developed by IBM Research. This analysis service takes as an input troubleshooting, information request from the users, etc. and performs workload analysis, peak analysis, troubleshooting time analysis, utilization rate analysis and outputs the report.
We work together with Application Management Service (AMS) and by making use of text analysis technology we are performing research on adding more depth to the ticket analysis. Currently, we are providing this service with cooperation of AMS department.
From the ticket analysis our goal is to perform research on ways to reduce service cost, reduce time to respond, raise the skill level of maintenance personals, and provide better service.
For large scale maintenance teams processing tickets, it is very important to aggregate resolution methods of such tickets into knowledge information and provide them as FAQs (Frequently Asked Question), from cost reduction perspective. Expected outcome of such FAQ service are:
- The number of opened tickets can be reduced since end users can reference these FAQ.
- The average ticket resolution time can be reduced since maintenance personals can effectively share knowledge among themselves.
From such background, we are conducting research mainly on these 3 points.
- The method to cluster ticket information with high accuracy and aggregate them into highly required FAQ
- The method to search the FAQ for intended information
- The method to provide intuitive method of creating and updating the FAQ.