Difference Image Analysis (DIA) is the cornerstone of the transient detection discovery pipelines in digital sky surveys. This technique for finding new objects and also subtle brightness variability of known astrophysical sources has been in use and developement for more than two decades, extending the reach of time domain astronomy. The forthcoming of The Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) is bringing new load on the shoulders of transient detection scientific tools, making them the rendezvous point of solar system science, galactic astronomy, and Cosmology. The Dark Energy Science Collaboration (DESC) is preparing scientific pipelines for transient detection for Cosmology with DIA during the LSST survey 10 year operation. Subtracting images can be done using different techniques that face an inverse problem: each image is being convoluted by a varying optical transfer function (or Point Spread Function PSF), creating the necessity of calibrating and optimizing these tools. In this presentation we will discuss the tools and analysis that DESC is doing in order to quantify and predict the performance of the DIA pipelines during LSST operations and its impact on Cosmology science results.