AssignmentProblems.jl

AssignmentProblems reads data files in gap format for Generalized Assignmnent Problem (GAP) instances.

Usage

The main type used by the package is AssignmentProblem, defined as follows:

struct AssignmentProblem
    name::String                # Instance name

    capacities::Vector{Int64}   # Agents' capacities
    costs::Matrix{Int64}        # Assigments costs (agents x jobs)
    consumptions::Matrix{Int64} # Assigments consumptions (agents x jobs)

    lb::Int64                   # Lower bound (typemin(Int64) if not known)
    ub::Int64                   # Upper bound (typemax(Int64) if not known)
end

The package also defines the functions na and nj returning the number of agents and jobs, respectively.

Some classical GAP instances from the literature are preloaded. For example, to load GAP instance a05100:

data = loadAssignmentProblem(:a05100)

There is a second optional parameter to set the objective function used (default :Min or :Max). This is only used to populate lb and ub fields.

See the full instance list.

AssignmentProblems also loads custom GAP instances (following ORLib format):

data = loadAssignmentProblem("/path/to/your/GAP/instance.txt")

Installation

AssignmentProblems is not yet a registered Julia Package. You can install AssignmentProblems through the Julia package manager. Open Julia's interactive session (REPL) and type:

] add https://github.com/rafaelmartinelli/AssignmentProblems.jl

Related links

• Mutsunori Yagiura's GAP Page
• ORLib's GAP page

Other packages

• Knapsacks.jl: Knapsack algorithms in Julia
• FacilityLocationProblems.jl: Facility Location Problem Lib
• BPPLib.jl: Bin Packing and Cutting Stock Lib
• CARPData.jl: Capacitated Arc Routing Problem Lib