Florian Schwendinger Updated: 2025-04-20
This repository contains an R interface to the HiGHS solver. The HiGHS solver, is a high-performance open-source solver for solving linear programming (LP), mixed-integer programming (MIP) and quadratic programming (QP) optimization problems.
The package can be installed from CRAN
install.packages("highs")
or GitLab.
::install_gitlab("roigrp/solver/highs") remotes
It is possible to use a precompile HiGHS library by providing the
system variable R_HIGHS_LIB_DIR
. For example I used
mkdir build
cd build
cmake .. -DCMAKE_INSTALL_PREFIX=/Z/bin/highslib -DCMAKE_POSITION_INDEPENDENT_CODE:bool=ON -DSHARED:bool=OFF -DBUILD_TESTING:bool=OFF
make install
to install the HiGHS library to
/Z/bin/highslib
Sys.setenv(R_HIGHS_LIB_DIR = "/Z/bin/highslib")
install.packages("highs")
# or
# remotes::install_gitlab("roigrp/solver/highs")
The highs package provides an API similar to
Rglpk and a low level API to the HiGHS
solver. For most users using highs_solve
as shown below
should be the best choice.
The package functions can be grouped into the following categories:
highs_solve
.highs_model
and
highs_solver
.highs_control
to construct
the control object for highs_solve
and
highs_solver
.hi_new_model
and other functions starting
with hi_model_
.hi_new_solver
and other functions starting
with hi_solver_
.example_model
and solvers
example_solver
for the documentation
examples.highs_available_solver_options
to get the
available solver options.highs_write_model
to write
the model to a file.library("highs")
The the example models and solvers are included to have small examples available for the manual.
writeLines(ls("package:highs", pattern = "^example"))
#> example_model
#> example_solver
The low-level model functions allow to create and modify models. More details and examples can be found in the manual.
writeLines(ls("package:highs", pattern = "^hi(|_new)_model"))
#> hi_model_get_ncons
#> hi_model_get_nvars
#> hi_model_set_constraint_matrix
#> hi_model_set_hessian
#> hi_model_set_lhs
#> hi_model_set_lower
#> hi_model_set_ncol
#> hi_model_set_nrow
#> hi_model_set_objective
#> hi_model_set_offset
#> hi_model_set_rhs
#> hi_model_set_sense
#> hi_model_set_upper
#> hi_model_set_vartype
#> hi_new_model
The low-level solver functions allow to create and modify solvers. More details and examples can be found in the manual.
writeLines(ls("package:highs", pattern = "^hi(|_new)_solver"))
#> hi_new_solver
#> hi_solver_add_cols
#> hi_solver_add_rows
#> hi_solver_add_vars
#> hi_solver_change_constraint_bounds
#> hi_solver_change_variable_bounds
#> hi_solver_clear
#> hi_solver_clear_model
#> hi_solver_clear_solver
#> hi_solver_get_bool_option
#> hi_solver_get_constraint_bounds
#> hi_solver_get_constraint_matrix
#> hi_solver_get_dbl_option
#> hi_solver_get_int_option
#> hi_solver_get_lp_costs
#> hi_solver_get_model
#> hi_solver_get_num_col
#> hi_solver_get_num_row
#> hi_solver_get_option
#> hi_solver_get_options
#> hi_solver_get_sense
#> hi_solver_get_str_option
#> hi_solver_get_variable_bounds
#> hi_solver_get_vartype
#> hi_solver_infinity
#> hi_solver_info
#> hi_solver_run
#> hi_solver_set_coeff
#> hi_solver_set_constraint_bounds
#> hi_solver_set_integrality
#> hi_solver_set_objective
#> hi_solver_set_offset
#> hi_solver_set_option
#> hi_solver_set_options
#> hi_solver_set_sense
#> hi_solver_set_variable_bounds
#> hi_solver_solution
#> hi_solver_status
#> hi_solver_status_message
#> hi_solver_write_basis
#> hi_solver_write_model
The high level functions allow to work with models and solvers. More details and examples can be found in the manual.
args(highs_model)
#> function (Q = NULL, L, lower, upper, A = NULL, lhs = NULL, rhs = NULL,
#> types = rep.int(1L, length(L)), maximum = FALSE, offset = 0)
#> NULL
args(highs_solver)
#> function (model, control = highs_control())
#> NULL
args(highs_control)
#> function (threads = 1L, time_limit = Inf, log_to_console = FALSE,
#> ...)
#> NULL
args(highs_write_model)
#> function (model, file)
#> NULL
The main function highs_solve
.
library("highs")
args(highs_solve)
#> function (Q = NULL, L, lower, upper, A = NULL, lhs = NULL, rhs = NULL,
#> types = rep.int(1L, length(L)), maximum = FALSE, offset = 0,
#> control = highs_control())
#> NULL
# Minimize
# x_0 + x_1 + 3
# Subject to
# x_1 <= 7
# 5 <= x_0 + 2 x_1 <= 15
# 6 <= 3 x_0 + 2 x_1
# 0 <= x_0 <= 4
# 1 <= x_1
<- rbind(c(0, 1), c(1, 2), c(3, 2))
A <- highs_solve(L = c(1.0, 1), lower = c(0, 1), upper = c(4, Inf),
s A = A, lhs = c(-Inf, 5, 6), rhs = c(7, 15, Inf),
offset = 3)
str(s)
#> List of 6
#> $ primal_solution: num [1:2] 0.5 2.25
#> $ objective_value: num 5.75
#> $ status : int 7
#> $ status_message : chr "Optimal"
#> $ solver_msg :List of 6
#> ..$ value_valid: logi TRUE
#> ..$ dual_valid : logi TRUE
#> ..$ col_value : num [1:2] 0.5 2.25
#> ..$ col_dual : num [1:2] 0 0
#> ..$ row_value : num [1:3] 2.25 5 6
#> ..$ row_dual : num [1:3] 0 0.25 0.25
#> $ info :List of 18
#> ..$ valid : logi TRUE
#> ..$ mip_node_count : num -1
#> ..$ simplex_iteration_count : int 0
#> ..$ ipm_iteration_count : int 5
#> ..$ qp_iteration_count : int 0
#> ..$ crossover_iteration_count : int 0
#> ..$ primal_solution_status : chr "Feasible"
#> ..$ dual_solution_status : chr "Feasible"
#> ..$ basis_validity : int 1
#> ..$ objective_function_value : num 5.75
#> ..$ mip_dual_bound : num 0
#> ..$ mip_gap : num Inf
#> ..$ num_primal_infeasibilities: int 0
#> ..$ max_primal_infeasibility : num 0
#> ..$ sum_primal_infeasibilities: num 0
#> ..$ num_dual_infeasibilities : int 0
#> ..$ max_dual_infeasibility : num 0
#> ..$ sum_dual_infeasibilities : num 0
# Minimize
# 0.5 x^2 - 2 x + y
# Subject to
# x <= 3
<- .Machine$double.eps * 100
zero <- rbind(c(1, 0), c(0, zero))
Q <- c(-2, 1)
L <- t(c(1, 0))
A
<- list(log_dev_level = 0L)
cntrl <- highs_solve(Q = Q, L = L, A = A, lhs = 0, rhs = 3, control = cntrl)
s str(s)
#> List of 6
#> $ primal_solution: num [1:2] 3 0
#> $ objective_value: num -6
#> $ status : int 10
#> $ status_message : chr "Unbounded"
#> $ solver_msg :List of 6
#> ..$ value_valid: logi TRUE
#> ..$ dual_valid : logi TRUE
#> ..$ col_value : num [1:2] 3 0
#> ..$ col_dual : num [1:2] 0 1
#> ..$ row_value : num 3
#> ..$ row_dual : num -2
#> $ info :List of 18
#> ..$ valid : logi TRUE
#> ..$ mip_node_count : num -1
#> ..$ simplex_iteration_count : int 1
#> ..$ ipm_iteration_count : int 0
#> ..$ qp_iteration_count : int 0
#> ..$ crossover_iteration_count : int 0
#> ..$ primal_solution_status : chr "Feasible"
#> ..$ dual_solution_status : chr "Infeasible"
#> ..$ basis_validity : int 1
#> ..$ objective_function_value : num -6
#> ..$ mip_dual_bound : num 0
#> ..$ mip_gap : num Inf
#> ..$ num_primal_infeasibilities: int 0
#> ..$ max_primal_infeasibility : num 0
#> ..$ sum_primal_infeasibilities: num 0
#> ..$ num_dual_infeasibilities : int 1
#> ..$ max_dual_infeasibility : num 1
#> ..$ sum_dual_infeasibilities : num 1
The HiGHs C++ library internally supports the matrix formats csc (compressed sparse column matrix) and csr (compressed Sparse Row array). The highs package currently supports the following matrix classes:
"matrix"
dense matrices,"dgCMatrix"
compressed sparse column matrix from the
Matrix package,"dgRMatrix"
compressed sparse row matrix from the
Matrix package,"matrix.csc"
compressed sparse column matrix from the
SparseM package,"matrix.csr"
compressed sparse row matrix from the
SparseM package,"simple_triplet_matrix"
coordinate format from the
slam package.If the constraint matrix A
is provided as
dgCMatrix
, dgRMatrix
, matrix.csc
or matrix.csr
the underlying data is directly passed to
HiGHs otherwise it is first transformed into the csc
format an afterwards passed to HiGHs
library("Matrix")
<- rbind(c(0, 1), c(1, 2), c(3, 2))
A <- as(A, "CsparseMatrix") # dgCMatrix
csc <- highs_solve(L = c(1.0, 1), lower = c(0, 1), upper = c(4, Inf),
s0 A = csc, lhs = c(-Inf, 5, 6), rhs = c(7, 15, Inf),
offset = 3)
<- as(A, "RsparseMatrix") # dgRMatrix
csr <- highs_solve(L = c(1.0, 1), lower = c(0, 1), upper = c(4, Inf),
s1 A = csr, lhs = c(-Inf, 5, 6), rhs = c(7, 15, Inf),
offset = 3)
library("SparseM")
<- as.matrix.csc(A)
csc <- highs_solve(L = c(1.0, 1), lower = c(0, 1), upper = c(4, Inf),
s2 A = csc, lhs = c(-Inf, 5, 6), rhs = c(7, 15, Inf),
offset = 3)
<- as.matrix.csr(A)
csr <- highs_solve(L = c(1.0, 1), lower = c(0, 1), upper = c(4, Inf),
s3 A = csr, lhs = c(-Inf, 5, 6), rhs = c(7, 15, Inf),
offset = 3)
library("slam")
<- as.simple_triplet_matrix(A)
stm <- highs_solve(L = c(1.0, 1), lower = c(0, 1), upper = c(4, Inf),
s4 A = stm, lhs = c(-Inf, 5, 6), rhs = c(7, 15, Inf),
offset = 3)
The function highs_available_solver_options
lists the
available solver options
<- highs_available_solver_options()
d "option"]] <- sprintf("`%s`", d[["option"]])
d[[::kable(d, row.names = FALSE) knitr
option | type |
---|---|
presolve |
string |
solver |
string |
parallel |
string |
run_crossover |
string |
time_limit |
double |
read_solution_file |
string |
read_basis_file |
string |
write_model_file |
string |
solution_file |
string |
write_basis_file |
string |
random_seed |
integer |
ranging |
string |
infinite_cost |
double |
infinite_bound |
double |
small_matrix_value |
double |
large_matrix_value |
double |
primal_feasibility_tolerance |
double |
dual_feasibility_tolerance |
double |
ipm_optimality_tolerance |
double |
primal_residual_tolerance |
double |
dual_residual_tolerance |
double |
objective_bound |
double |
objective_target |
double |
threads |
integer |
user_bound_scale |
integer |
user_cost_scale |
integer |
highs_debug_level |
integer |
highs_analysis_level |
integer |
simplex_strategy |
integer |
simplex_scale_strategy |
integer |
simplex_crash_strategy |
integer |
simplex_dual_edge_weight_strategy |
integer |
simplex_primal_edge_weight_strategy |
integer |
simplex_iteration_limit |
integer |
simplex_update_limit |
integer |
simplex_min_concurrency |
integer |
simplex_max_concurrency |
integer |
log_file |
string |
write_model_to_file |
bool |
write_presolved_model_to_file |
bool |
write_solution_to_file |
bool |
write_solution_style |
integer |
glpsol_cost_row_location |
integer |
write_presolved_model_file |
string |
output_flag |
bool |
log_to_console |
bool |
timeless_log |
bool |
ipm_iteration_limit |
integer |
pdlp_native_termination |
bool |
pdlp_scaling |
bool |
pdlp_iteration_limit |
integer |
pdlp_e_restart_method |
integer |
pdlp_d_gap_tol |
double |
qp_iteration_limit |
integer |
qp_nullspace_limit |
integer |
iis_strategy |
integer |
blend_multi_objectives |
bool |
log_dev_level |
integer |
log_githash |
bool |
solve_relaxation |
bool |
allow_unbounded_or_infeasible |
bool |
use_implied_bounds_from_presolve |
bool |
lp_presolve_requires_basis_postsolve |
bool |
mps_parser_type_free |
bool |
use_warm_start |
bool |
keep_n_rows |
integer |
cost_scale_factor |
integer |
allowed_matrix_scale_factor |
integer |
allowed_cost_scale_factor |
integer |
ipx_dualize_strategy |
integer |
simplex_dualize_strategy |
integer |
simplex_permute_strategy |
integer |
max_dual_simplex_cleanup_level |
integer |
max_dual_simplex_phase1_cleanup_level |
integer |
simplex_price_strategy |
integer |
simplex_unscaled_solution_strategy |
integer |
presolve_reduction_limit |
integer |
restart_presolve_reduction_limit |
integer |
presolve_substitution_maxfillin |
integer |
presolve_rule_off |
integer |
presolve_rule_logging |
bool |
presolve_remove_slacks |
bool |
simplex_initial_condition_check |
bool |
no_unnecessary_rebuild_refactor |
bool |
simplex_initial_condition_tolerance |
double |
rebuild_refactor_solution_error_tolerance |
double |
dual_steepest_edge_weight_error_tolerance |
double |
dual_steepest_edge_weight_log_error_threshold |
double |
dual_simplex_cost_perturbation_multiplier |
double |
primal_simplex_bound_perturbation_multiplier |
double |
dual_simplex_pivot_growth_tolerance |
double |
presolve_pivot_threshold |
double |
factor_pivot_threshold |
double |
factor_pivot_tolerance |
double |
start_crossover_tolerance |
double |
less_infeasible_DSE_check |
bool |
less_infeasible_DSE_choose_row |
bool |
use_original_HFactor_logic |
bool |
run_centring |
bool |
max_centring_steps |
integer |
centring_ratio_tolerance |
double |
icrash |
bool |
icrash_dualize |
bool |
icrash_strategy |
string |
icrash_starting_weight |
double |
icrash_iterations |
integer |
icrash_approx_iter |
integer |
icrash_exact |
bool |
icrash_breakpoints |
bool |
mip_detect_symmetry |
bool |
mip_allow_restart |
bool |
mip_max_nodes |
integer |
mip_max_stall_nodes |
integer |
mip_max_start_nodes |
integer |
mip_max_leaves |
integer |
mip_max_improving_sols |
integer |
mip_lp_age_limit |
integer |
mip_pool_age_limit |
integer |
mip_pool_soft_limit |
integer |
mip_pscost_minreliable |
integer |
mip_min_cliquetable_entries_for_parallelism |
integer |
mip_report_level |
integer |
mip_feasibility_tolerance |
double |
mip_rel_gap |
double |
mip_abs_gap |
double |
mip_heuristic_effort |
double |
mip_min_logging_interval |
double |
mip_heuristic_run_rins |
bool |
mip_heuristic_run_rens |
bool |
mip_heuristic_run_root_reduced_cost |
bool |
mip_heuristic_run_zi_round |
bool |
mip_heuristic_run_shifting |
bool |
mip_improving_solution_save |
bool |
mip_improving_solution_report_sparse |
bool |
mip_improving_solution_file |
string |
mip_root_presolve_only |
bool |
mip_lifting_for_probing |
integer |
for additional information see the HiGHS homepage.
HiGHS currently has the following status codes defined in
HConst.h"
.
enumerator | status | message |
---|---|---|
kNotset |
0 | "Not Set" |
kLoadError |
1 | "Load error" |
kModelError |
2 | "Model error" |
kPresolveError |
3 | "Presolve error" |
kSolveError |
4 | "Solve error" |
kPostsolveError |
5 | "Postsolve error" |
kModelEmpty |
6 | "Empty" |
kOptimal |
7 | "Optimal" |
kInfeasible |
8 | "Infeasible" |
kUnboundedOrInfeasible |
9 | "Primal infeasible or unbounded" |
kUnbounded |
10 | "Unbounded" |
kObjectiveBound |
11 | "Bound on objective reached" |
kObjectiveTarget |
12 | "Target for objective reached" |
kTimeLimit |
13 | "Time limit reached" |
kIterationLimit |
14 | "Iteration limit reached" |
kUnknown |
15 | "Unknown" |
kMin |
0 | "Not Set" |
kMax |
15 | "Unknown" |