oppr (version 1.0.0)

solvers: Problem solvers

Description

Specify the software and configuration used to solve a project prioritization problem. By default, the best available exact algorithm solver will be used.

Details

The following solvers can be used to find solutions for a project prioritization problem:

add_default_solver

This solver uses the best software currently installed on the system.

add_gurobi_solver

Gurobi is a state-of-the-art commercial optimization software with an R package interface. It is by far the fastest solver that can be used by this package, however, it is also the only solver that is not freely available. That said, licenses are available to academics at no cost. The gurobi package is distributed with the Gurobi software suite. This solver uses the gurobi package to solve problems.

add_rsymphony_solver

SYMPHONY is an open-source integer programming solver that is part of the Computational Infrastructure for Operations Research (COIN-OR) project, an initiative to promote development of open-source tools for operations research (a field that includes linear programming). The Rsymphony package provides an interface to COIN-OR and is available on CRAN. This solver uses the Rsymphony package to solve problems.

add_lpsymphony_solver

The lpsymphony package provides a different interface to the COIN-OR software suite. Unlike the Rsymhpony package, the lpsymphony package is distributed through Bioconductor. The lpsymphony package may be easier to install on Windows or Max OSX systems than the Rsymphony package.

add_lpsolveapi_solver

lp_solve is an open-source integer programming solver. The lpSolveAPI package provides an interface to this solver and is available on CRAN. Although this solver is the slowest currently supported solver, it is also the only exact algorithm solver that can be installed on all operating systems without any manual installation steps.

add_heuristic_solver

Generate solutions using a backwards heuristic algorithm. Although these types of algorithms have conventionally been used to solve project prioritization problems, they are extremely unlikely to identify optimal solutions and provide no guarantees concerning solution quality.

add_random_solver

Generate solutions by randomly funding actions. This can be useful when evaluating the performance of a funding scheme---though it is strongly recommended to evaluate the performance of a funding scheme by comparing it to an optimal solution identified using exact algorithms (e.g. add_gurobi_solver, add_rsymphony_solver).

See Also

constraints, decisions, objectives, problem, targets.

Examples

# NOT RUN {
# load data
data(sim_projects, sim_features, sim_actions)

# build problem
p1 <- problem(sim_projects, sim_actions, sim_features,
             "name", "success", "name", "cost", "name") %>%
     add_max_richness_objective(budget = 200) %>%
     add_binary_decisions()

# build another problem, with the default solver
p2 <- p1 %>%
      add_default_solver()

# build another problem, with the gurobi solver
# }
# NOT RUN {
p3 <- p1 %>%
      add_gurobi_solver()
# }
# NOT RUN {
# build another problem, with the Rsympony solver
# }
# NOT RUN {
p4 <- p1 %>%
      add_rsymphony_solver()
# }
# NOT RUN {
# build another problem, with the lpsymphony solver
# }
# NOT RUN {
p5 <- p1 %>%
      add_lpsymphony_solver()
# }
# NOT RUN {
# build another problem, with the lpSolveAPI solver
p6 <- p1 %>%
      add_lpsolveapi_solver()

# build another problem, with the heuristic solver
p7 <- p1 %>%
      add_heuristic_solver()

# build another problem, with the random solver
p8 <- p1 %>%
      add_random_solver()

# }
# NOT RUN {
# generate solutions using each of the solvers
s <- rbind(solve(p2), solve(p3), solve(p4), solve(p5), solve(p6), solve(p7),
           solve(p8))
s$solver <- c("default", "gurobi", "Rsymphony", "lpsymphony", "lpSolveAPI",
              "heuristic", "random")

# print solutions
print(as.data.frame(s))
# }