Random Instance Generators
Multiobjective and Biobjective Multi Dimensional Knapsack Problems
Modolib.generate_knapsack_instances — Function.generate_knapsack_instances(num_obj::Int64=2, num_var::Int64=10, num_cons::Int64=1, sparse::Bool=true)Returns a MOBPInstance of a random Multi Objective Knapsack Problem with num_obj objectives num_var binary variables and num_cons inequality constraints.
Modolib.generate_bokp_instances — Function.generate_bokp_instances(num_var::Int64, num_cons::Int64, sparse::Bool)Returns a BOBPInstance of a random Bi Objective Knapsack Problem with num_var binary variables and num_cons inequality constraints.
The method for generating a random biobjective multi dimensional knapsack instance is as follows:
When sparse = true
\[\begin{align*}
c^1_i & \in [1:100], & \forall i \in \{1,..., n\} \\
c^2_i & \in [1:100], & \forall i \in \{1,..., n\} \\
A_{i,j} & \in [0:100], & \forall i \in \{1,..., n\}, j \in \{1,..., m\} \\
b_j & = \lceil \dfrac{\sum_i A_{ij}}{2} \rceil, & \forall j \in \{1,..., m\}
\end{align*}\]
When sparse = false
\[\begin{align*}
c^1_i & \in [1:100], & \forall i \in \{1,..., n\} \\
c^2_i & \in [1:100], & \forall i \in \{1,..., n\} \\
A_{i,j} & \in [1:100], & \forall i \in \{1,..., n\}, j \in \{1,..., m\} \\
b_j & = \lceil \dfrac{\sum_i A_{ij}}{2} \rceil, & \forall j \in \{1,..., m\}
\end{align*}\]
Multiobjective General Mixed Binary Problem
Modolib.generate_mixed_binary_instances — Function.generate_mixed_binary_instances(num_obj::Int64=2, num_cont_var::Int64=10, num_bin_var::Int64=10, num_cons::Int64=1)Returns a MOMBLPInstance or a BOMBLPInstance of a random Multiobjective Mavrotas Mixed Binary Problem.