Cost model & CAPEX

Precise vs fast mode

The accuracy/speed trade-off when estimating CAPEX, and resampling.

CO2GIS offers two ways to read the cost factors along the route. You pick one in the Calculation Mode of the Price Estimation tab.

Precise (default)

Cell-based, with resampling:

• resamples all cost rasters to a common grid (nearest-neighbour, to preserve categorical values);
• for each route segment, finds the crossed cells with the Bresenham line algorithm;
• computes the exact L_cell by line–cell geometric intersection;
• reads each factor per cell; N is the count of features intersecting both the segment and the cell (capped at 10).

Use it for final estimates.

Fast (point sampling)

Segment-based, no resampling and no geometric intersection:

Note — “segment” here means a vertex-to-vertex edge of the pipeline polyline, not a segment between booster stations (see Segments & booster stations).

• processes vertex-to-vertex segments;
• samples 5 points at t ∈ {0, 0.25, 0.5, 0.75, 1.0} and reads the native-resolution rasters;
• takes the maximum of the five (a conservative, worst-case choice);
• uses the segment length as L_cell.

Use it for scenario exploration and sensitivity analysis.

Fast mode introduces a small positive bias (the max-of-five over-estimates the average). Prefer precise for the number you report.

Performance & resampling

Resampling the cost surface to a coarser cell size is the biggest speed lever — going to 50 m has ~25× fewer cells and cuts least-cost-path time by about 97%. On the most demanding scenario (CS4 — see Validation):

So a 50 m + precise run is usually the sweet spot: minutes instead of over an hour, with sub-1% deviation.