作者: Guolin Shi , Bing Xu , Xin Wang , Zhong Xue , Xinyuan Shi
DOI: 10.1155/2019/4139762
关键词: Time release technology 、 Quality (business) 、 Reliability (statistics) 、 Near-infrared spectroscopy 、 Graph (abstract data type) 、 Computer science 、 Multivariate calibration 、 Reliability engineering 、 Tolerance interval 、 Process analytical technology
摘要: The concept of real-time release testing (RTRT) has recently been adopted by the production pharmaceuticals in order to provide high-level guarantee product quality. Process analytical technology (PAT) is an attractive and efficient way for realizing RTRT. In this paper, near-infrared (NIR) determination cryptotanshinone tanshinoneIIA content tanshinone extract powders was taken as research object. aim NIR analysis reliably declare compliant with its specification limits or not. First, quantification method developed parameters multivariate calibration model were optimized. reliable concentration ranges covering two APIs successfully verified accuracy profile (AP) methodology. Then, designed validation data from AP, unreliability graph decision tool built. Innovatively, β-content, γ-confidence tolerance intervals (β-CTIs) around estimated. During routine use, boundary β-CTIs could help decide whether prediction results are acceptable. proposed quantified risk near confirmed that unreliable region useful quality a way. Such strategy be extended other PAT applications improve reliability results.