Planning a recurring stockpile measurement workflow

Recurring stockpile measurement starts with a decision rather than a flight plan: which inventory question has to be answered, how often, and with what evidence standing behind it? The workflow described below is representative, and it does not report a completed engagement, a measured result, or a delivered timeline. Rather, it sets out the decisions a recurring program has to settle in advance, so that a future engagement can be scoped to the site, the operating conditions, the intended use, and the required deliverables.

The recurring decision

On a representative aggregate site, two groups want different things from the same measurement. Finance needs month-end quantities that use the same units, the same pile definitions, and the same review path from one cycle to the next, because a quantity that cannot be compared to last month's quantity is not usable for reconciliation. Operations needs capture windows that account for active equipment, site access, weather, airspace, and safe separation from working crews. A recurring program succeeds or fails on whether those two requirement sets are reconciled before the first flight rather than after the third.

Aerial measurement fits when the required output is a repeatable geometric volume and the relevant surface can be reconstructed from visible imagery, which covers a large share of routine inventory work. However, it does not fit every question asked of a stockpile: hidden geometry, material identification, a density determination, a legal survey product, and any evidence beyond the scoped accuracy mode all sit outside what a camera can establish. Naming that boundary early is considerably cheaper than discovering it in a reconciliation meeting.

Scope the workflow

Each cycle in a representative program begins with an approved area of interest, a pile inventory, exclusion rules, a coordinate reference system, units, a flight design, and a control plan, all recorded before anyone launches. Ground control and independent checkpoints are included only as the intended use and the contracted accuracy mode require, since checkpoints nobody will cite are cost without evidence. While reusing a documented boundary and reference-surface convention makes cycle-to-cycle comparison substantially more interpretable, every cycle still requires review for site changes, because a boundary that was correct in March may enclose a haul road by August.

A first processing pass runs in the field, in the mobile lab, and serves as a completeness check rather than an accuracy check. The crew reviews planned extent, gaps, overlap, alignment, control behavior, and obstructions while a safe recovery flight may still be possible, which is the narrow window in which a capture problem is still inexpensive. That check reduces the risk of discovering the problem after demobilization; it does not, by itself, establish the final dataset's positional accuracy.

The accuracy statement matches the evidence, and under the American Society for Photogrammetry and Remote Sensing (ASPRS) standard the two available statements are not interchangeable. Produced to ASPRS describes acquisition and processing under the agreed methodology, with resolution, coordinate system, control approach, and limitations stated; it does not imply independently measured accuracy, and it should not be read as though it did. Tested to ASPRS adds independent surveyed checkpoints withheld from the adjustment, documented residuals and root mean square error (RMSE), the checkpoint count and distribution, the supported class, and any limitations that apply.

For scheduling context, eligible standard-scope deliverables are typically available within 24 hours after flight. That figure is a scope-qualified service expectation and not a claim that this representative workflow has been flown, processed, or delivered on that timeline. Larger areas, complex surfaces, multi-day captures, and tested-to scopes each receive a stated schedule at quote rather than an assumed one.

Review and finance handoff

Before a cycle enters the inventory record, a named reviewer confirms the capture date, the cycle and dataset identifiers, the pile list, boundaries, units, base or comparison surfaces, exceptions, accuracy mode, and any density source used for conversion. The word carrying the weight in that sentence is named: a review that belongs to everyone belongs to no one, and an inventory figure with no accountable reviewer behind it is difficult to defend later. The handoff package itself includes the volume table, the map or orthomosaic, the scoped surfaces, the methodology and limitations record, and the review status.

Finance then decides whether that package is fit for the organization's reconciliation and control process, which is a judgment belonging to the organization rather than to the measurement provider. The workflow does not replace accounting policy, material-density controls, independent audit requirements, or any professional survey responsibility. Rather, it supplies a documented, repeatable geometric input that those processes can accept, reject, or reconcile against their own records.

In this representative workflow, each accepted delivery carries a lifecycle record listing the capture date, dataset and version identifiers, coordinate reference system, units, boundary-set version, reference-surface version, density source if one was used, accuracy statement, processing notes, exceptions, revisions, and review status. Additionally, a revision creates a new version rather than silently overwriting the earlier record, so a comparison run two years from now can identify exactly which inputs and which files produced each figure. In other words, the record exists so that a disagreement about a number becomes a lookup rather than an argument.

The linked Lifetime Data Support policy covers in-scope reprocessing, revisions, and re-delivery of delivered datasets for the life of the project, and it covers those things only. A re-flight is a new engagement. Raw imagery retention is a separate matter with its own clock: 60 days at no charge unless an extension is selected at scoping. Any later request that depends on raw source imagery therefore depends on the applicable retention window, which is a reason to settle the retention tier while the project is being scoped rather than when the request arrives.

The scope boundary

Overall, a recurring aerial program is appropriate when a team needs the same stockpile decision on a schedule and can define a measurement method stable enough to repeat, and it is inappropriate when either of those conditions is missing. Before committing, the scope should name the frequency, operating window, capture boundary, control and accuracy mode, reference surfaces, density owner, finance review path, deliverable package, delivery schedule, version record, support boundary, and raw-retention tier. This representative workflow does not, by itself, establish cost, safety, or turnaround outcomes for any particular site. Perhaps the more useful test of a measurement program is not how accurate a single cycle proves to be, but rather whether the twelfth cycle can still be compared honestly to the first.