The SMPS format
explained
e-mail: hgassman@mgmt.dal.ca
Bjarni
Kristjansson
Maximal
Software, Inc.,
e-mail: bjarni@maximalsoftware.com
Synopsis
Stochastic programming is an active area of research due to
recent advances in computing power. However, for benchmarking and comparison of
algorithms it is essential to have a way of exchanging test problem sets. The
SMPS format is available to describe stochastic linear and quadratic programs.
It is based on the MPS format, the de
facto standard for linear programs.
While it is widely used, SMPS does not enjoy universal
acceptance. Part of the reason is that the record-based structure of MPS is
deemed to be overly rigid and limiting, but we feel that at least in part the
reason is a lack of examples that describe the many and varied constructs of
the SMPS format.
This web site illustrates most of the current features of
SMPS using sample problems. Some of these problems have appeared in the
literature, others were designed specially for this collection.
Unlike other collections of stochastic programs, whose main
focus is archiving and benchmarking, the problems offered on this site have
been selected mainly for variety, both of application area and of model type
and features.
Download the paper (IMA Journal of Management Mathematics, DOI: 10.1093/imaman/dpm007)
SCENARIOS format: Investment problem
NODES format: Production problem
INDEP and BLOCKS format: Energy problem
Linear transformations: Forestry
problem
ARMA process: Multistage production problem
NETWORK format: Asset management
Mixed LP and network: Fisheries and production
SIMPLE recourse: Airline fleet allocation
CHANCE constraints: Water management