One of the primary challenges in explosive detection using fluorescence quenching is the identification and quantification of detected targets. In this work, we explore the reliability of aerosol jet printed sensor arrays for the discrimination of nitroaromatic traces using linear discriminant analysis (LDA). We varied the amount of the deposited material by controlling the printer's shutter to investigate the impact on the detection reliability. For a twofold variation of the amount of the deposited material, we report excellent classification rates between 81 and 96% for the discrimination of nitrobenzene, 1,3-dinitrobenzene, and 2,4-dinitrotoluene at 1, 3, and 10 parts per billion in air, respectively. Our results close to the detection limits indicate a remarkable identification and quantification of explosive trace vapors because of high control of the printing process. This work demonstrates the high potential of digitally printed fluorescence quenching sensor arrays and the excellent capabilities of LDA as a simple supervised statistical learning technique.