Optimization of Dew Point, Humidity, and Temperature in Data Centers using Genetic Algorithm with Multiple Linear Regression

This study provides help in the Department of Public Works & Highways - Information Management Service Technology Support Division Systems Administration Section of the Philippines in bringing optimization analysis and metric standardization in their current data centers. The researchers proposes that the relationship between dew point, humidity and temperature be known in order to produce standards in maintaining the server reliability of their data centers. The researchers use a MATLAB program to create a genetic algorithm code in producing the best preferred optimization values with the help of the multiple linear regression as its fitness function. The results in the multiple linear regression analysis proved that there is a strong significant relationship between dew point (negative relationship) and humidity (positive relationship) to temperature based on its multiple r (correlation), r-squared and ANOVA- significance F. The generated optimized results from the MATLAB genetic algorithm code are the following; dew point be 12 degrees Celsius, humidity be in the range of 58 to 61 percent, and temperature be in the range of 20.027 to 20.253 degrees Celsius.