Table 2 Evaluation conditions setting.

From: Quantum annealing-based route optimization for commercial AGV operating systems in large-scale logistics warehouses

Benchmark evaluation conditions settings

Parameter

Gurobi or Gurobi-MILP

Openjij_SA

QA

AGV Numbers (N)

10,20,50,100,200,...,1000

10,20,50,100,200,...,1000

10,20,50,100,200,...,1000

Variables

10,100,500,1000,...,30,000

10,100,500,1000,...,30,000

10,100,500,1000,...,30,000

Candidate Route Method

DCG/RCG

DCG/RCG

DCG/RCG

Clustering Method

NCO/SCO/CCO

NCO/SCO/CCO

NCO/SCO/CCO

Max_tags (tags)

18

18

18

Max_remain_tags (tags)

10

10

10

Search_Time (ms)

\(3 \times 10^8\)

\(3 \times 10^8\)

\(3 \times 10^8\)

Max_Candidate_Search

100

100

100

Penalty_On_Tag

True/False

True/False

True/False

Used_Tag_Penalty

15

15

15

N_Samples

1000

1000

1000

Num_Sweeps

100

100

100

Annealing time (\(\mu s\))

-

-

20

Number of experiments

1

5

5/15

\(\lambda _1\)

From Table 4

From Table 4

From Table 4

\(\lambda _2\)

From Table 4

From Table 4

From Table 4

  1. The solver SA is selected in the problem generation step. All solvers are used in the evaluation experiment with the settings shown above. The number of experimental repetitions for both QA_RCG+CCO and QA_DCG+CCO is 15, respectively.