EPSRC logo

Details of Grant 

EPSRC Reference: GR/S45188/01
Title: Indexability and index heuristics for the intractable restless bandit problem
Principal Investigator: Glazebrook, Professor KD
Other Investigators:
Ansell, Dr P
Researcher Co-Investigators:
Project Partners:
Department: Sch of Economics
Organisation: University of Edinburgh
Scheme: Standard Research (Pre-FEC)
Starts: 07 April 2004 Ends: 20 September 2005 Value (£): 137,203
EPSRC Research Topic Classifications:
Mathematical Aspects of OR
EPSRC Industrial Sector Classifications:
No relevance to Underpinning Sectors
Related Grants:
Panel History:
Panel DatePanel NameOutcome
10 Jun 2003 Mathematics Prioritisation Panel Science Deferred
Summary on Grant Application Form
Whittle (1988) introduced a class of decision processes called restless bandit problems (RBPs) and proposed an approach to the development of index heuristics via Langrangian relaxations. RBPs generalise the class of multi-armed bandit problems (MABs) and are rich in modelling power. They produce a natural framework for the study, inter alia, of problems concerning the control of multiclass queueing systems, machine maintenance, multi-location inventory problems and military service control problems. The intractability of RBPs (they are PSPACE-hard) has discouraged much further research effort since the seminal contributions of Weber and Weiss in the early 90's. However, recent work by the investigators and Nino-Mora has suggested that the achievable region approach has the power to elucidate fundamental issues concerning both when indices exist (indexability) and how close index heuristics are to being optimal. The research will seek both to progress these general theoretical questions and also to develop and evaluate structured index heuristics (in simple closed form where possible) for the above application areas. The programme will also consider ways of deploying the restless bandit formulation and the associated notions of indexability/indices in more complex systems where the exact application of the ideas is either not oossible or requires further develooment
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: http://www.ed.ac.uk