Respirable dust in coal mining environments has long been a concern for occupational health. Over the past several decades, much effort has been devoted to reducing dust exposures in these environments, and rates of coal workers’ pneumoconiosis (CWP) have dropped significantly. However, in some regions, including parts of Central Appalachia it appears that incidence of CWP has recently been on the rise. This trend is yet unexplained, but a possible factor might be changes in specific dust characteristics, such as particle composition, size or shape. Prior work in our research group has developed a standardized methodology for analyzing coal mine dust particles on polycarbonate filter media using scanning electron microscopy with energy dispersive x-ray (SEM-EDX). While the method allows individual particles to be characterized, it is very time-intensive because the instrument user must interrogate each particle manually; this limits the number of particles that can practically be characterized per sample. Moreover, results may be somewhat user-dependent since classification of particle composition involves some interpretation of EDX spectra. To overcome these problems, we aim to automate the current SEM-EDX method. The ability to analyze more particles without user bias should increase reproducibility of results as well as statistical confidence (i.e., in applying characteristics of the analyzed particles to the entire dust sample.) Some challenges do exist in creating an automated routine, which are primarily related to ensuring that the available software is programmed to differentiate individual particles from anomalies on the sample filter media, select and measure an appropriate number of particles across a sufficient surface area of the filter, and classify particle compositions similarly to a trained SEM-EDX user following a manual method. This paper discusses the benefits and challenges of an automated routine for coal mine dust characterization, and progress to date toward this effort.