Construction supervision
with real visibility of the works

An industrial plant on the river needed to extend its site. The digital terrain model that arrived as project background showed that at an elevation of 61.80 meters the water surface came dangerously close to the new structures. The client’s question was direct: how much higher do we need to fill?

Before answering, the technical team asked a different question: where does that model’s elevation come from? The background was an inherited terrain model with no documented traceability to the official vertical reference system. Auditing the base data meant a new aerial photogrammetric survey and a benchmark tied to the Villeta gauge and to Paraguay’s National Geodetic Network. That was extra work, extra budget, and another week of waiting on site.

When the new model was loaded onto the site, the critical elevation moved. The water surface sat seventeen centimeters below what the inherited model said. The fill the client was about to budget was not merely oversized: the problem it was meant to solve did not exist in that form. The structure was not compromised.

Outcome: the managers had a number they could defend, and the projected works came down in earthwork volume. Above all, the decision rested on base data verified against the official network, not on an inherited model of uncertain origin.

1. Traceability of the base data

Before supervising execution, you have to be able to defend the data execution is running on. Reference elevation, coordinate system, terrain model, project drawings: each has an origin and a chain of validations behind it. If that chain is broken, everything measured afterwards carries the error forward.

In practice this means insisting on benchmarks tied to the National Geodetic Network where the works are sensitive to elevation, asking for the documented origin of inherited terrain models, and running a cross-check against an independent survey when the cost of the error exceeds the cost of the audit. It is a step you go through once, at the start of the project, and it prevents decisions being taken on data that would not survive a review.

2. Physical and financial progress in the same view

Physical progress tells you what happened on site. Financial progress tells you what was certified. When they live in separate spreadsheets, the gap between what was executed and what was paid surfaces late — and when it surfaces late, corrective action has less room.

An S-curve updated with actual progress, next to the committed budget and the period’s certificate, answers a single question: is the project where it should be, on time and on cost? The schedule and cost performance indices from earned value analysis are the quantitative version of that question. Early detection of deviations depends on having one integrated view that updates at the cadence of the project.

3. Reporting differentiated by role

An investor needs an executive summary: where the project stands today, what changed this week that moves the horizon, and where a decision is required in the next two weeks. A project director needs the full dashboard: progress by trade, items showing deviation, pending certifications, active alerts. A site professional needs operational sheets: daily control, materials records, reports with photographs.

Those three levels describe the same project from three angles. When all three are generated from a single source of data, the information is consistent across roles, meetings get shorter, and each decision is taken with the level of detail appropriate to whoever signs it.