In many industrial companies, although data is ubiquitous, its quality is often problematic. Engineers find that the difficulty lies not in a lack of data, but in incomplete, inconsistent, or poorly structured data, which complicates analysis. Before embarking on analytical models, it is crucial to thoroughly understand the data sources and their transformations. To improve data quality, it is recommended to document the sources, standardize formats, automate data cleaning, and regularly verify data quality. This presents an opportunity for engineers to transform this data into powerful decision-making tools.