<p>Vehicle trajectory data is becoming increasingly essential for data-driven applications that support smarter and greener cities. From optimizing transportation networks to supporting the shift toward Electromobility, these applications rely on vast amounts of detailed vehicle trajectory data. However, the growing volume of this data introduces two major challenges: efficient storage of large datasets and privacy preservation, as trajectory data itself is classified as personal information. To address these challenges, both data compression and privacy preservation mechanisms are critical. Despite their different objectives, these two applications share a common need: preserving data usability. Traditional data compression techniques, such as line simplification, can reduce data size but often result in the loss of important and irrecoverable information. Similarly, anonymization methods face a trade-off between protecting privacy and maintaining the utility of the data, where stronger anonymization often leads to less usable data. Building on this, this paper presents a unified framework that addresses both challenges simultaneously by leveraging Tensor-based dimensional reduction. By creating a lower-dimensional representation of the trajectory data, the framework achieves two goals at once: (1) it compresses the data, reducing its size for storage purposes, and (2) it anonymizes the data by representing it in a space that lacks direct physical meaning, thereby protecting privacy. The results demonstrate that this approach achieves high compression rates with minimal loss of data accuracy, maintaining usability for downstream applications. Additionally, the structure of the compressed data supports varying levels of anonymization by controlling access to different portions of the lower-dimensional representation.</p>