Implementations of Hill Climbing and Iterated Local Search algorithms for the Next Release Problem. These implementations use a graphical pattern observed in problem instances to narrow the search space.
Fixed random seed rationale: miliseconds system time on 2016-04-26.
To run the tunning configurations, use the following commands:
- TIS -i classic -b 30 50 70 -o output_tis_classic.txt -s 253958446296927
To run the landscape reports for the "Cost-Risk as an Objective" (CRR), "Cost-Risk as a Constraint" (CCR), "Profit-Risk as an Objective" (PRR), and "Release Planning" (RPR), use the following commands:
-
LCRR -i classic -b 30 -r 10 20 40 -o results/landscape/cost-risk/%s.txt -s 253958446296927
-
LCCR -i classic -b 30 -r 8 -o results/landscape/cost-cap/%s.txt -s 253958446296927
-
LPRR -i classic -b 30 -r 10 20 40 -o results/landscape/profit-risk/%s.txt -s 253958446296927
-
LRPR -i nrp1 -b 15 -r 6 -o results/landscape/release/%s.txt -s 253958446296927
To run the optimizers for the "Profit-Only" (OP), "Cost-Risk Only" (OCO), "Cost-Risk as an Objective" (OCR), "Cost-Risk as a Constraint" (OCC), "Profit-Risk Only" (OPO), "Profit-Risk as an Objective" (PRR), and "Release Planning" (ORL), use the following commands:
-
OP -i classic -b 30 50 70 -o output_op_classic.txt -s 253958446296927
-
OCO -i classic -b 30 50 70 -o output_oco_classic.txt -s 253958446296927
-
OPO -i classic -b 30 50 70 -o output_opo_classic.txt -s 253958446296927
-
OCR -i classic -b 30 -r 10 20 40 -o output_ocr_classic.txt -s 253958446296927
-
OCC -i classic -b 30 -r 8 -o output_occ_classic.txt -s 253958446296927
-
OPR -i classic -b 30 -r 10 20 40 -o output_opr_classic.txt -s 253958446296927
-
ORL -i classic -b 15 -r 6 -t 5 -o output_orl_classic.txt -s 253958446296927