Statistical processing, quantitative

Quantitative index of risk levels produced by AHP (Analytic Hierarchy program) technique has been compared in order to analyze safety cost rate by project size, by project kind, by project risk. This can be referred to making out determination equation to select the number of safety engineers, assigned to construction sites by project amount, by project risk level. Also, questioimaire items and statistic process can be done, in reference to this advanced smdy result. 

Data of planimetric and stereometric analyses are subjected to comprehensive statistical analysis aimed at obtaining complex quantitative information on soil microstructure. It is done both by functionality of the software STIMAN-3D and by program packages for statistical processing from different manufactures. Owing to full automation... 

In this work, we describe an approach to integrate multivariate statistical process monitoring and online HAZOP analysis for abnormal event management of batch processes. The framework consists of three main parts process monitoring and fault detection, automated online HAZOP analysis module and a coordinator. Multiway PCA is used for batch process monitoring and fault detection. When abnormal event is detected, signal-to-symbol transformation technique based on variable contribution is used to transfer quantitative sensor readings to qualitative states. Online HAZOP analysis is based on PHASuite, an automated HAZOP analysis tool, to identify the potential causes, adverse consequences and potential operator options for the identified abnormal event. 

De Thomas etal. [Ill] studied the production of polyurethanes and showed that NIRS can be used successfully to monitor the course of the reaction in real time. Spectral data were obtained with a dispersive instrument, using standard transflectance probes. An MLR model was derived for the quantitative determination of isocyanate concentrations during the urethane polymerization reaction. Model predictions were used to build statistical process control charts and to detect trends along the polymerization reaction. The authors suggested that the integration of NIRS with process control routines could lead to improvements of product quality and consistency, while minimizing reaction time. However, model predictions were not used as feedback information for any sort of correction of the process trajectory. Similar studies were performed by Dallin [112] for prediction of the acid number during the production of polyesters. 

A quantitative theory of rate processes has been developed on the assumption that the activated state has a characteristic enthalpy, entropy and free energy the concentration of activated molecules may thus be calculated using statistical mechanical methods. Whilst the theory gives a very plausible treatment of very many rate processes, it suffers from the difficulty of calculating the thermodynamic properties of the transition state. 

Our own earlier work on the chlorination of toluene had been subject to similar constraints. In this case, chlorination with ferf-butyl hypochlorite had proved to be advantageous. In the presence of silica gel as catalyst the yield of chlorotoluenes was quantitative but the regioselectivity was more or less statistical (ref. 8). However, the use of proton-exchanged zeolite X allowed the production of chlorotoluenes with a para-selectivity of more than 90 % (Fig. 4) (ref. 9). No HCl is generated in this process since the by-product is tert-butanol, and there is no inhibition of the catalyst. Indeed, the catalyst can be reused if necessary. 

The models which we have developed can be classified as follows. Some are intended to represent physicochemical processes and properties by mimicking quantitatively concepts which have become accepted by chemists in general. A simple example would be the transfer of electronic charge between two atoms of differing electronegativities. Other models are statistical in nature. We have applied parameters quantified by the physicochemical models to series of chemical data. The relationships thus derived by various statistical techniques, and their form, is such that they are readily applicable to the task of quantifying the evaluation process in EROS. Further discussion of these points is a major feature of this article. 

---