| EPSRC Reference: |
GR/J90855/01 |
| Title: |
LARGE SCALE STOCHASTIC PROGRAMMING ALGORITHMS |
| Principal Investigator: |
Dempster, Professor MAH |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Mathematical Sciences |
| Organisation: |
University of Essex |
| Scheme: |
Standard Research (Pre-FEC) |
| Starts: |
01 May 1994 |
Ends: |
14 March 1996 |
Value (£): |
45,019
|
| EPSRC Research Topic Classifications: |
| Mathematical Aspects of OR |
|
|
| EPSRC Industrial Sector Classifications: |
|
| Related Grants: |
|
| Panel History: |
|
|
Summary on Grant Application Form |
|
The research will develop, code and test algorithms for solving large scale linearly constrained dynamic stochastic programming (DSP) problems which typically arise from strategic, tactical and operational planning problems in a variety of real world applications. In such applications the number of possible alternative problem data paths (scenarios) can become astronomical, requiring that only a representative sample be used in the optimisation process. A new DSP algorithm for the linear case will iterate importance sampling of data paths, solution of the resulting large scale deterministic equivalent linear programme using nested Benders' decomposition (MSLiP) and reduction of the problem using expected value of perfect information techniques (COMPACTER), in order to converge to an unbiased minimum variance estimate of the problem expected value based on a fixed sample size. The algorithm will be tested on an existing extensive set of representative real world test problems and the experience gained used to extend it to the practically important case of (differentiable) nonlinear (convex) objective functions in which at each iteration of the decomposition procedure first order approximations are substituted for true objective values. Tools will be developed in C++ to implement the procedures developed and for DSP model specification and management which augment previously developed software. Parallelisation of the developed algorithms will be pursued concurrently under separate funding.
|
|
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.sx.ac.uk |