Worst-case experiment

The roles of the research scientist and the regulation officer differ. Laboratory scientists are obligated to get the facts and not prejudge the situation worst-case experiments may be their stock-in-trade. Regulatory decisions are difficult to make, and demand the best and most comprehensive data base available. 

Drilling of wells with less than the previously required pressure to prevent blowout. That required pressure had been based on worst case experience. 

A potential outlier should always be examined carefully. First we check if there was a simple mistake during the recording or the early manipulation of the data. If this is not the case, we proceed to a careful examination of the experimental circumstances during the particular experiment. We should be very careful not to discard ("reject") an outlier unless we have strong non-statistical reasons for doing so. In the worst case, we should report both results, one with the outlier and the other without it. 

If there is no possibility to cover the complete range with one detector, there may be the possibility to use two detectors which are placed in different distances from the sample. In the worst case the experiment has to be performed several times with different setups. 

For non-Newtonian liquids and suspensions, an apparent viscosity is determined using correlations which include power input and the Reynolds number. Scale-up comparisons based on heat generation data only were determined by comparison of results from RC1 experiments and from a 675-liter reactor [208]. In the experiments, a Bingham plastic fluid was used to determine the film heat transfer coefficient. This presents a worst case because of the low thermal conductivity of the Bingham plastic. Calculated inside film heat transfer coefficients determined in the RC1 tests were about 60% lower than the values determined in the pilot plant reactor, even though substantial effort was made to obtain both geometric and kinematic similarity in the pilot reactor. 

In an experiment, we compare the optimization strategies introduced for different spot price scenarios. Two alternative demand spot price scenarios best case and worst case with equal probability of 0.25 are defined in addition to the standard scenario with probability 0.5. The best case assumes a continuous price increase while the worst case assumes a continu-... 

Mr. Don Clay, Director of the Office of Toxic Substances, discussed the premanufacture review procedures and experience with PMNs to date at a meeting of the Organization for Economic Cooperation on Development (OECD) Chemicals Forum in December, (10). He noted that EPA s chemistry, toxicology, and exposure assessment teams normally complete their preliminary evaluation within a week of receipt of a PMN, and, that preliminary assessment eliminates about 50 percent of the substances as chemicals of low concern. They then proceed to structure activity analysis and reasonable worst case assumptions to assess unreasonable risk or the need for more data. 

A statistical significant effect is not always relevant from the practical point of view. Therefore, a worst-case level combination experiment with regard to the studied response (e.g., resolution) is determined and performed with replicates. In this experiment, only the method parameters with major effects (both statistically significant and almost significant) are considered. As can be seen in Figure 15, the major effect, temperature and pH, are easily detected. The worst-case combinations... 

This effect is shown in Figure 7.2, where the upper left element of the (Y Y)" matrix (the element used to estimate the variance of b is plotted as a function of the center of various two-experiment designs. The worst case illustrated involves two experiments separated by two units in the factor jc, as the design is moved away from jCj = 0, the uncertainty in b begins to rise steeply. The best case illustrated is for two experiments separated by ten units in the factor x the uncertainty in b increases very slowly as the center of the design is moved away from jc, = 0. Note that in all cases, the element of (X X) associated with the variance of bf, is equal to its minimum value of 0.5 when the design is centered at jc, = 0. Note also that the... 

Further information on stability can be derived from the provider s experience or from literature. It might also be necessaiy worst-case conditions. 

This probably reflects a common range of methods in practice, although from our own experience I would add that it seems better not to allow too much (or any) prior discussion of forthcoming reductions in dosage, as many reasons tend to be given to avoid these Again as examined in Chapter 2 there is virtually no support for benzodiazepine maintenance as there is for that with methadone and buprenorphine, and almost certainly benzodiazepines are far more often diverted to others than the opioid substitutes whose consumption can be supervised, with in the worst cases the tablets representing little more than currency. 

Two trials were carried out using the same levels as experiment 12 =8 kN, 7>2=400-800 pm, 7 3=12 kN and T 4=35% ethylcellulose. Table 10 gives the results of the three trials in this validation. The predicted responses from the linear models and the relative errors are presented. If 3, Y4 and 8 present the worst case, the results on the main responses are suitable. 

Pilot Plant Design, Experiments, and Results. Design of the pilot plant scale-up was based upon the earlier study involving quinaldic acid breakthrough. Because this compound may represent a worst case, other factors must be considered. 

Fletcher and Ley ffer (1994) studied the worst-case performance of GOA in an attempt to present the potential limitations that the outer approximation algorithm of Duran and Grossmann (1986a) may exhibit despite the encouraging experience obtained from application to engineering problems. 

In most of the cases where experimental values are available B3LYP overestimates the dipole moments. The exception is ScO, which is underestimated by 0.55D. The largest deviation is for TiO (+0.7D). The LSD calculations of Piechota and Suffczynski [19] give dipole moments in better agreement with experiment, but for the two worst cases (ScO and TiO) the values are only marginally better than the B3LYP results. 

The heat release rate of the reaction at process temperature can be estimated from an experiment in a reaction calorimeter. If it is unknown, as in a worst-case assumption, the cooling capacity of the reactor can be used instead, since for an isothermal process the neat heat release rate of the reaction is certainly inferior to the cooling capacity. Some examples of the acceleration factor are represented graphically in Figure 10.7. 

A series of isothermal experiments are performed at different temperatures in a calorimeter, for example, in a DSC. On each curve, the maximum heat release rate, which represents the worst case, is measured (Figure 11.4). The experimental procedure must reach the desired temperature as fast as possible. For this purpose, in a DSC, the oven is preheated to the desired temperature with the reference cmcible in place. The sample cmcible is then placed into the oven and the measurement begins once thermal equilibrium is achieved, which usually takes approximately 2 minutes. During this time no measurement is possible, but 2 minutes is a very short time relative to the total experimental time of several hours. Thus, the achieved conversion before the measure really starts is negligible. It is left as an exercise for the reader to verify this point Moreover, the difference may be corrected graphically. 

---