The {WeCo2} platform for assisted multi-well correlation: {Current} {Status}.

@inproceedings{Caumon2026RMa,
 abstract = {WeCo2 is a set of Python functions and executables to address challenges in stratigraphic correlation in terms of automation, reproducibility and uncertainty quantification. To implement assisted stratigraphic correlation, WeCo2 rests on discrete dynamic time warping (DTW) principles to match two finite data series based on a local cost function. DTW results are best translated into correlation panels when applied on point data, but both point and interval data can be used to define local cost functions in WeCo2. As stratigraphic correlation is fundamentally an ill-posed problem, WeCo2 implements n-best DTW and a stochastic DTW version to generate several acceptable correlation scenarios, represented as a directed acyclic graph. In a multi-well settings, this allows to generalize pairwise multi-scenario DTW by correlating two already aligned groups of wells in an incremental way. A graph connecting the wells may be used to define the incremental correlation order. WeCo2 also implements multiscale correlation, whereby a vertically detailed correlation is constrained by a coarser-scale solution using depth-based mapping. The goal of these improvements is not only to accelerate computations, but also to explore more effectively the correlation search space. WeCo2 implements simple data-driven cost functions such as data variance and previously published rules translating sedimentological knowledge such as distality or depositional profile. It also defined an extensible framework for new cost functions. From a technical standpoint, WeCo2 combines executables and a rich Python API. Developers can use the Python API to define new correlation cost functions depending on the type of data and the geological context. For performance, the core data structures and algorithms are implemented in C++, and multi-well incremental correlation is multithreaded.},
 author = {Caumon, Guillaume and Antoine, Christophe},
 booktitle = {2026 {RING} meeting},
 language = {en},
 pages = {188--202},
 publisher = {ASGA},
 title = {The {WeCo2} platform for assisted multi-well correlation: {Current} {Status}.},
 year = {2026}
}