Kernel/lib/branch_and_bound.pl

7 % in compliance with the License.  You may obtain a copy of the License
38     optimization. It can be used with any nondeterministic search routine
56         Found a solution with cost 500
57         Found a solution with cost 400
58         Found a solution with cost 200
59         Found no solution with cost -1.0Inf .. 199
67     generator of solutions with different values for the variable Cal,
70     good solutions (i.e. solutions with smaller cost), and finally delivers
71     the minimum-cost solution with Cal=200.
76         Found a solution with cost 1.5
77         Found a solution with cost 0.99
78         Found no solution with cost -1.0Inf .. 0.94
88 <H3>Optimization with Constraints</H3>
90     This library is designed to work together with arbitrary constraint
103         Found a solution with cost 5
104         Found a solution with cost 0
105         Found a solution with cost -2
106         Found a solution with cost -4
107         Found a solution with cost -6
108         Found no solution with cost -15.0 .. -7.0
123     succeeds with a solution.  Moreover, good, early bounds on the cost
133     with continuous variables.  When the variable domains have been
329 	    % this prevents looping with overlapping costs
474     % We assume that there is a solution with cost Best.
475     % If FromNoSol==true then we know there is no solution with cost From
530 	    % this prevents looping with overlapping costs
588 	    fail	% to undo the call-effect and succeed with 2nd clause
597 % the bounds from propagation with the user option input (if any)
607 	report_result(Spec, "Found a solution with cost %q%n", Cost, Handle, Module).
610 	report_result(Spec, "Found no solution with cost %q%n", Range, Handle, Module).
659 	remembered and the search is continued or restarted with an
669 	    	<DD>after finding a solution, continue search with the newly
672 	    	<DD>after finding a solution, restart the whole search with
701 	    <DD>number - minimal improvement ratio (with respect to the lower
778 % simple minimization with default options
780     Found a solution with cost 9
781     Found a solution with cost 6
782     Found a solution with cost 4
783     Found a solution with cost 2
784     Found no solution with cost -1.0Inf .. 1
791     Found a solution with cost 9
792     Found a solution with cost 4
793     Found no solution with cost -1.0Inf .. 0
800     Found a solution with cost 99
801     Found a solution with cost 40
802     Found no solution with cost 0.0 .. 20.0
803     Found a solution with cost 30
804     Found no solution with cost 20.0 .. 25.0
805     Found no solution with cost 25.0 .. 27.5
806     Found no solution with cost 27.5 .. 28.75
807     Found no solution with cost 28.75 .. 29.0
811 % examples with library(ic) constraints
817     Found a solution with cost 5
818     Found a solution with cost 0
819     Found a solution with cost -2
820     Found a solution with cost -4
821     Found a solution with cost -6
822     Found no solution with cost -15.0 .. -7.0
835     Found a solution with cost 15
836     Found a solution with cost 8
837     Found a solution with cost 1
838     Found a solution with cost -4
839     Found a solution with cost -6
840     Found no solution with cost -15.0 .. -7.0
919 	remembered and the search is continued or restarted with an
929 	    	<DD>after finding a solution, continue search with the newly
932 	    	<DD>after finding a solution, restart the whole search with
957 	    <DD>number - minimal improvement ratio (with respect to the lower
1012 	in Optimum, and the Solution argument gets unified with an instance of
1024 	"Solution":"A term which will be unified with the optimized Template",