No description available.

Data volume is growing for each day as more and more is digitalized which puts the data management on test. The smart tools developed by Atlas Copco saves and transmits data to the cloud as a service to find errors in tightening's for their customers to review. A problem is the amount of data that is lost in this process. A tightening cycle usually contains thousands of data points and the storage space for it is too great for the tool's hardware. Today many of the data points are deleted and a small portion of scattered data of the cycle is saved and transmitted. To avoid overfilling the storage space the data need to be minimized. This study is focus on comparing data compression algorithms that could solve this problem.   In a literature study in the beginning, numerous data compression algorithms were found with their advantages and disadvantages. Two different types of compression algorithms are also defined as lossy compression, where data is compressed by losing data points or precision, and lossless compression, where no data is lost throughout the compression. Two lossy and two lossless algorithms are selected to be avaluated with respect to their compression ratio, speed and error tolerance. Poor Man's Compression - Midrange (PMC-MR) and SWING-filter are the lossy algorithms while Gorilla and Fixed-Point Compression (FPC) are the lossless ones.   The reached compression ratios, in percentage, could range from 39\% to 99\%. As combinations of a lossy and a lossless algorithm yields best compression ratios with lower error tolerance, PMC-MR with Gorilla is suggested to be the best suited for Atlas Copco's needs.