| EPSRC Reference: |
GR/H12348/01 |
| Title: |
DECISION SUPPORT SYSTEMS: SAFETY AND LIABILITY (DECSAFE - IED4/1/9053) |
| Principal Investigator: |
Bell, Dr J |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Computer Science |
| Organisation: |
Queen Mary University of London |
| Scheme: |
Standard Research (Pre-FEC) |
| Starts: |
22 June 1992 |
Ends: |
21 December 1995 |
Value (£): |
94,075
|
| EPSRC Research Topic Classifications: |
|
| EPSRC Industrial Sector Classifications: |
| Financial Services |
Information Technologies |
| No relevance to Underpinning Sectors |
R&D |
|
| Related Grants: |
|
| Panel History: |
|
|
Summary on Grant Application Form |
|
To enable development and deployment of knowledge-based Decision Support Systems (DSSs) which can meet stringent criteria of safety and soundness. The research has both technical and non-technical objectives including: a mathematical theory of decision making, prototype software tools, investigating areas of legal liability, demonstrator applications. Progress:The main project is now complete. ISL, ICRF, and QMW have some funds remaining to work on the asthma demonstrator, to disseminate the results, and to further investigate modelling safety with logic.The project has researched, published and implemented a new approach to decision support, based on an explicit model of knowledge-based decision-taking. Informally, a RED agent can observe the environment including states resulting from its own actions.It maintains a database of beliefs about the environment. Certain beliefs cause the agent to raise goals. These goals evoke problem solving to find candidate solutions. Arguments are constructed for and against candidate solutions. After reviewing arguments the agent may be able to commit to one hypothesis (a new belief), or to a preferred plan of action. Commitment to a plan, leads indirectly through new observations to a new beliefs, leading to new cycles of problem-solving.The project has defined a Red Representational Language R2L to be used by knowledge engineers. R2L explicitly supports key RED concepts such as beliefs, decisions, plans, actions, obligations, and constraints. QMWs work has been to provide a sound theoretical basis in the form of a formal Logic for R2L (LR2L). LR2L is essentially temporal propositional logic extended with modal operators to represent the concepts of argument support, beliefs, goals, and obligations. Metalevel reasoning is required both to formalise the aggregation of arguments, and to ensure that constraints satisfy the knowledge base. A prototype RED agent developed in Prolog has been used in a number of medical DSSs including asthma and cancer patient management. Analysis of Cancer protocols has shown the existence of generic safety rules.A preliminary version or RED was used by Lloyds Register in a software assistant for LR experts assessing the quality or safety of software. Masons Solicitors have investigated the legal implications of SCS DSSs and have prototyped a hypermedia tool to provide guidance to project managers. Over 20 papers have been published or submitted to journals and workshops.
|
|
Key Findings |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Potential use in non-academic contexts |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Impacts |
|
| Description |
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk |
| Summary |
|
| Date Materialised |
|
|
|
|
Sectors submitted by the Researcher |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
| Project URL: |
|
| Further Information: |
|
| Organisation Website: |
|