Validation, comparative modelling process

In this section, we present two examples with different scenarios. The first example illustrates the performance of the model on a single site total refinery planning problem where we compare the results of the model to an industrial scale study from Favennec et al. (2001). This example serves to validate our model and to make any necessary adjustments. The second example extends the scale of the model application to cover three complex refineries in which we demonstrate the different aspects of the model. The refineries considered are of large industrial-scale refineries and actually mimic a general set-up of many areas around the world. The decisions in this example include the selection of crude blend combination, design of process integration network between the three refineries, and decisions on production units expansion options and operating levels. 

Validation of models is desired but can be difficult to achieve. Models are empirically validated by examining how output data (predictions) compare with observed data (such comparisons, of course, must be conducted on data sets that have not been used to create or specify the model). However, model validations conducted in this manner are difficult given limitations on data sources. As an alternative approach, model credibility can be assessed by a careful examination of the subcomponents of the model and inputs. One should ask the question Does the selection of input variables and the way they are processed make sense Also, confidence in the model may be augmented by peer reviews and the opinion of the scientific community. Common faults and shortcomings are... 

Compared to efficacy, safety is typically more multifactorial, as it is dependent on homeostasis of virtually all cellular processes. A wider number and diversity of potential molecular and cellular effects of compound interactions may affect safety than may affect efficacy or bioavailability. Accordingly, cytotoxicity assessment is less specific, more multi parametric and extrapolatable with less certainty, unless there are specific safety signals indicated by the chemical structure or by its precedents. Extrapolation of safety biomarker data needs a greater foundation of mechanistic understanding of both in vitro and in vivo pathogenesis of toxicities, as well as rigorous, empirical validation of models. 

Once the above-discussed components of the model have been determined, they are added to the final model of a monolith (or even filter) reactor. The monolith reactor model has already been described in Section III. The next stage is to validate the model by comparing the predictions of the model based on laboratory data, with the real-world data measured on an engine bench or chassis dynamometer. At this stage the reason(s) for any discrepancies between the prediction and experiment need to be determined and, if required, further work on the kinetics done to improve the prediction. This process can take a number of iterations. Model validation is described in more detail in Section IV. D. Once all this has been done the model can be used predictively with confidence. 

Once the parameters are calculated, the final stage is the construction of a model, i.e., the validation. This stage consists of checking the results of the model for several different experimental cases. Its goal is to confirm that simulations carried out with the model reproduce the behavior of the processes in a proper way. Some statistical procedures are typically applied in this stage. The best way to validate a model is to compare experimental and modeled data by means of two different tests ... 

Validation is one of the most difficult aspects of environmental QSAR development due to the comparatively small size of the database. Cross-validation has been useful in validating the effectiveness of the model. In this method, one compound is removed from the database, the equation is recalculated, and the toxicity of the omitted compound is estimated. The process is repeated for all compounds in the dataset and the results are tabulated. In this manner, a calculation of the accuracy of prediction of continuous data and the rate of misclassification for categorial data can be compiled. A more useful estimate of the validity of the QSAR model is its ability to predict the toxicity of new compounds. Generally, this is difficult to accomplish in a statistically significant way due to the slow accumulation of new data that meet the criteria used in the modeling process and the associated expense. 

Ever since the development and application of mathematical models for the design of SMB processes, beginning in the 1980s, efforts have been made to validate these models by comparing measured and simulated data. SMB and TMB models of different complexity have been used for this task, for example the ideal and equilibrium... 

The NIPALS algorithm can tolerate missing data. It is therefore possible to compute a principal components model if data are left out from the data matrix during the modelling process. This can be used to determine whether or not a new component is significant by examining how well the expanded model with the new component can predict left-out data, as compared to the model without the new component. If the new component does not improve the predictions, it is considered not to be significant. The cross validation procedure can be summarized as follows ... 

A qualitative PCR assay was developed for the detection of peanut in foods and validated further with the collaboration of six participant laboratories (Watanabe et al., 2007). Autoclaved, roasted, boiled, and nonprocessed doughs made out of Japanese yam spiked with different levels of defatted peanut flour (0, 0.001, 0.01, 0.1%) were used as food models. Results were compared with ELISA, which showed decreased protein levels in the processed food models, especially in the autoclaved dough. No protein was detected in aU of the nonspiked dough. PCR results from the six labs correctly identified dough samples (processed and nonprocessed), which correlated with results obtained from ELISA analysis. The assay was shown to be specific, reproducible, reliable, and applicable for the detection of peanut in the model processed food. 

These values show that from the two possible alternatives of ion formation that one is preferred, which leads to the formation of an anion with the largest number of F-, after which, of Cl-ligands. It is remarkable that ionization in Lewis acid mixtures is favoured versus that in pure Lewis acids in all cases. This could be the reason why the polymerization conversion increases when using Lewis acid mixtures as initiators. However, it should be pointed out here that the quantum chemical reaction energies employed are only then comparable with each other, when they are valid for the same process used for modelling the reactions. 

---