The capital, operating, and overall costs of a dedicated continuous manufacturing process to synthesize an active pharmaceutical ingredient (API) and formulate it into tablets are estimated for a production scale of 2000 t of tablets per year, with raw material cost, production yield, and API loading varied over broad ranges. Costs are compared to batch production in a dedicated facility. Synthesis begins with a key organic intermediate three synthetic steps before the final API; results are given for key intermediate (KI) costs of $100 to $3000/kg, with drug loadings in the tablet of 10 and 50 wt %. The novel continuous process described here is being developed by an interdisciplinary team of 20 researchers. Since yields are not yet well-known, and continuous processes typically have better yields than batch ones, the overall yields of the continuous processes with recycling were set equal to that of the batch process. Without recycling, yields are 10% lower, but less equipment is required. The continuous process has not been built at large scale, so Wroth factors and other assumptions were used to estimate costs. Capital expenditures for continuous production were estimated to be 20 to 76% lower, depending on the drug loading, KI cost, and process chosen; operating expenditures were estimated to be between 40% lower and 9% higher. The novel continuous process with recycling coupled to a novel direct tablet formation process yields the best overall cost savings in each drug loading/KI price scenario: estimated savings range from 9 to 40%. Overall cost savings are also given assuming the yield in the continuous case is 10% above and 10% below that of the batch process. Even when yields in the continuous case are lower than in the batch case, savings can still be achieved because the labor, materials handling, CapEx, and other savings compensate.

In continuous branch-and-bound algorithms, a very large number of boxes near global minima may be visited prior to termination. This so-called cluster problem (Du and Kearfott, 1994) is revisited and a new analysis is presented. Previous results are confirmed, which state that at least second-order convergence of the relaxations is required to overcome the exponential dependence on the termination tolerance. Additionally, it is found that there exists a threshold on the convergence order pre-factor which can eliminate the cluster problem completely for second-order relaxations. This result indicates that, even among relaxations with second-order convergence, behavior in branch-and-bound algorithms may be fundamentally different depending on the pre-factor. A conservative estimate of the prefactor is given for αBB relaxations.

Time-varying, or dynamic, experiments can produce richer data sets than sequences of steady-state experiments using less material and time. A case study demonstrating this concept for microreactor experiments is presented. Beginning with five kinetic model candidates for the reaction of phenylisocyanate with t-butanol, an initial dynamic experiment showed that two of the five models gave a similar quality of fit to the experimental data, whereas the remaining three gave significantly poorer fits. Next an optimal experiment was designed to discriminate between the remaining two models. This drove the two models to differ significantly in quality, leaving a single model and a set of kinetic parameter values that adequately described the data. This method can be applied to future kinetic studies to reduce material use and experimental time while validating a dynamic model of the physics and chemical kinetics.

Convergence-order analysis for differential-inequalitiesbased bounds and relaxations of the solutions of ODEsThe MIT Faculty has made this article openly available. Please share how this access benefits you. Your story matters.