Although $\beta $ -ray measurements remain one of the main methods for determining the concentration of dust in mines, the accuracy of direct measurements is impacted by radiation-intensity fluctuations across the probe beam. We collected coal-dust samples on membrane filters and compared the dust-concentration levels from a basic algorithm for determining the concentration by $\beta $ -ray absorption through the membranes with the levels calculated using a newly developed algorithm. The two algorithms both used standardized weights of coal deposited on the membranes but differed in the particular statistical variables used to describe the $\beta $ -ray characteristics. Our analysis revealed how $\beta $ -ray intensity fluctuations affected measurement accuracy and precision, thereby highlighting the limitations of the basic algorithm. We evaluated the standard deviation and envelope mean functions that characterized the fluctuations in the probe-beam flux, and we then used those statistics to derive new algorithms for better concentration accuracy. We employed a data fusion technique to establish a new algorithm. The new algorithm was tested experimentally and was shown to reduce the influence of the intensity fluctuations on the measurement of coal-dust concentrations, enhancing both the accuracy and reliability of the readings.