TRIAL DESIGN WITH CONVEX-HULL TECHNIQUES
A method, according to some implementations, includes obtaining trial design simulation results for a set of trial designs and determining a score for each trial design based on a performance criteria. The method may further include evaluating the designs in the set of trial designs to determine a convex hull for the set of trial designs, filtering designs based on the convex hull, and communicating the convex hull designs.
1 . A method comprising:
obtaining trial design simulation results for a set of trial designs;
determining a score for each trial design based on a performance criteria;
evaluating the designs in the set of trial designs to determine a convex hull for the set of trial designs;
filtering designs based on the convex hull; and
communicating the convex hull designs.
2 . The method of claim 1 , further comprising:
recommending designs within epsilon-distance from designs on the convex hull.
3 . The method of claim 1 , further comprising:
identifying facets in the convex hull and performing simulated annealing based on the facets.
4 . The method of claim 1 , further comprising:
identifying different design types on the convex hull and recommending different design types.
5 . The method of claim 1 , further comprising:
identifying a second level convex hull.
6 . The method of claim 2 , further comprising:
receiving feedback for recommended designs and determining epsilon values for additional recommendations.
7 . The method of claim 1 , further comprising:
clustering designs dominated by designs in the convex hull.
8 . The method of claim 1 , further comprising:
clustering designs that are within a margin of error.
9 . The method of claim 1 , further comprising:
updating the convex hull designs in response to at least one of addition or subtraction of designs from the set of trial designs.
10 . The method of claim 1 , further comprising:
updating the convex hull designs in response to an update of scenario probabilities.
11 . The method of claim 1 , further comprising:
evaluating historical trial design selections to identify one or more trial design parameters based at least in part on one or more trial design criteria determined from a user via an interactive interface, wherein obtaining the trial design simulation results is based at least in part on a quick search data structure and the one or more trial design parameters;
generating a substitute for at least some of the trial design simulation results based at least in part on a relationship between the trial design simulation results and supplemental data;
generating a performance surface based at least in part on the set of trial designs; and
evaluating one or more trial designs based at least in part on the performance surface,
wherein calculating the score is further based on normalized score component values corresponding to the trial design simulation results.
12 . An apparatus comprising:
a data processing circuit configured to obtain design data for a set of trial designs;
an optimality determining circuit configured to determine optimum designs using convex hull analysis from the set of trial designs; and
a design analysis circuit configured to analyze the convex hull designs and determine a modification to the convex hull analysis;
wherein the optimality determining circuit is further structured to receive the modification and determine a second set of optimum designs.
13 . The apparatus of claim 12 , wherein the set of trial designs includes all design options for a set of criteria.
14 . The apparatus of claim 12 , wherein the optimality determining circuit modifies the convex hull analysis to determine designs within epsilon-distance of convex hull designs.
15 . The apparatus of claim 12 , wherein the optimality determining circuit modifies the convex hull analysis to determine designs dominated by the convex hull designs.
16 . The apparatus of claim 12 , wherein the optimality determining circuit modifies the convex hull analysis to determine designs clustered by the convex hull designs.
17 . The apparatus of claim 12 , wherein the optimality determining circuit modifies the convex hull analysis to determine twins of the convex hull designs.
18 . The apparatus of claim 12 , wherein the optimality determining circuit modifies the convex hull analysis to determine siblings of the convex hull designs.
19 . The apparatus of claim 12 , wherein the optimality determining circuit modifies the convex hull analysis to determine second level convex hull designs.
20 . A system comprising:
an electronic device including an electronic display; and
a server in electronic communication with the electronic device and including at least one processor structured to:
obtain trial design simulation results for a set of trial designs;
determine a score for each trial design based on a performance criteria;
evaluate the designs in the set of trial designs to determine a convex hull for the set of trial designs;
filter designs based on the convex hull; and
transmit the filtered designs to the electronic device;
wherein the electronic device is structured to display the filtered designs on the electronic display.