Skip to contents

Candidate locations must be supplied explicitly. Allowed and exclusion areas, spacing, cost, target areas, and any user scores remain visible. ps_candidate_network() supports coverage, support-gap, model-conditional kriging-variance reduction, or preserved user scores. Sequential greedy selection updates after each choice; it is not globally optimal and does not establish access, ownership, constructability, or drillability.

design <- ps_candidate_network(existing, candidates, "support_gap",
                               n_select=3, target=weak_support,
                               exclusion_area=exclusions)