No abstract

No abstract

No abstract

Both the increasing number and diversity of illicit-drug seizures complicate forensic drug identification. Traditionally, colorimetric tests are performed on-site, followed by transport to a laboratory for confirmatory analysis. Higher caseloads increase laboratory workload and associated transport and chain-of-evidence assurance performed by police officers. Colorimetric tests are specific only for a small set of drugs.The rise of new psychoactive substances therefore introduces risks for erroneous results. Near-infrared (NIR)-based analyzers may overcome these encumbrances by their compound-specific spectral selectivity and broad applicability. This work introduces a portable NIR analyzer that combines a broad wavelength range (1300-2600 nm) with a chemometric model developed specifically for forensic samples. The application requires only a limited set of reference spectra for time-efficient model training. This calibration-light approach thus eliminates the need of extensive training sets including mixtures. Performance was demonstrated with 520 casework samples resulting in a 99.6% true negative and 97.6% true positive rate for cocaine. Similar results were obtained for MDMA, methamphetamine, ketamine, and heroin. Additionally, 236 samples were analyzed by scanning directly through their plastic packaging. Also here, a >97% true positive rate was obtained. This allows for non-invasive, operator-safe chemical identification of potentially potent drugs of abuse. Our results demonstrate the applicability for multiple drug-related substances. Ideally, the combination of this NIR approach with other portable techniques, such as Raman and IR spectroscopy and electrochemical tests, may eventually eliminate the need for subsequent laboratory analysis; therefore, saving tremendous resources in the overall forensic process of confirmatory illicit drug identification.

No abstract

No abstract

No abstract

This paper describes the concepts of statistical batch process monitoring and the associated problems. It starts with an introduction to process monitoring in general which is then extended to batch process monitoring. The performance of control charts for batch process monitoring is discussed by means of two performance indices: the overall type I error and the action signal time. Problems associated with the existing approach are discussed and highlighted. Improvements are suggested and checked with the performance indices. To evaluate the effect of the proposed improvements as well as to assess the performance of the existing approach, an industrial batch production process is used.