Robust optimization of uncertain multistage inventory systems with inexact data in decision rules

In production-inventory problems customer demand is often subject to uncertainty. Therefore, it is challenging to design production plans that satisfy both demand and a set of constraints on e.g. production capacity and required inventory levels. Adjustable robust optimization (ARO) is a technique to solve these dynamic (multistage) production-inventory problems. In ARO, the decision in each stage is a function of the data on the realizations of the uncertain demand gathered from the previous periods. These data, however, are often inaccurate; there is much evidence in the information management literature that data quality in inventory systems is often poor. Reliance on data “as is” may then lead to poor performance of “data-driven” methods such as ARO. In this paper, we remedy this weakness of ARO by introducing a model that treats past data itself as an uncertain model parameter. We show that computational tractability of the robust counterparts associated with this extension of ARO is still maintained. The benefits of the new model are demonstrated by a numerical test case of a well-studied production-inventory problem. Our approach is also applicable to other ARO models outside the realm of production-inventory planning.     

Quadrupole moment of the 11(-) intruder isomer in 196Pb and its implications for the 16(-) shears band head

The quadrupole moment of the 11(-) isomer in 196Pb has been measured by the level mixing spectroscopy method. This state has a pi(3s(-2)(1/2)1h(9/2)1i(13/2))11(-) configuration which is involved in most of the shears band heads in the Pb region. The first directly measured value of Q(s)(11(-)) = (-)3.41(66) b, coupled to the previously known quadrupole moment of the nu(1i(-2)(13/2))12(+) isomer allows us to estimate the quadrupole moment of the 16(-) shears band head as Q(s)(16(-)) = -0.32(10) b. The experimental values are compared to tilted axis cranking calculations, giving insight into the validity of the additivity approach to couple quadrupole moments and on the amount of deformation in the shears bands.     

Evaluation of process parameters of ultrasonic treatment of bacterial suspensions in a pilot scale water disinfection system

In this study, several process parameters that may contribute to the efficiency of ultrasound disinfection are examined on a pilot scale water disinfection system that mimics realistic circumstances as encountered in an industrial environment. The main parameters of sonication are: (i) power; (ii) duration of treatment; (iii) volume of the treated sample. The specific energy (E_s) is an indicator of the intensity of the ultrasound treatment because it incorporates the transferred power, the duration of sonication and the treated volume. In this study, the importance of this parameter for the disinfection efficiency was assessed through changes in volume of treated water, water flow rate and electrical power of the ultrasonic reactor. In addition, the influences of the initial bacterial concentration on the disinfection efficiency were examined. The disinfection efficiency of the ultrasonic technique was scored on a homogenous and on a mixed bacterial culture suspended in water with two different types of ultrasonic reactors (Telsonic and Bandelin). This study demonstrates that specific energy, treatment time of water with ultrasound and number of passages through the ultrasonic reactor are crucial influential parameters of ultrasonic disinfection of contaminated water in a pilot scale water disinfection system. The promising results obtained in this study on a pilot scale water disinfection system indicate the possible application of ultrasound technology to reduce bacterial contamination in recirculating process water to an acceptable low level. However, the energy demand of the ultrasound equipment is rather high and therefore it may be advantageous to apply ultrasound in combination with another treatment.     

Study of the carbonyl products of terpene/OH radical reactions: detection of the 2,4-DNPH derivatives by HPLC-MS

The oxidation of the terpenes - and -pinene, limonene and ³-carene by hydroxyl radicals has been investigated in a fast-flow reactor coupled to a liquid nitrogen trap for collecting the carbonyl compounds. Identification of the products was performed via 2,4-dinitrophenylhydrazone (DNPH) derivatization of the carbonyls to form the mono- and di-DNPH derivatives, which were analysed by high-performance liquid chromatographic (HPLC)-DAD (diode array detector) and HPLC-mass spectrometry (HPLC-MS). Both electrospray ionization [ESI(–)] and atmospheric pressure chemical ionization [APCI(–)] were suitable for the detection of the DNPH derivatives of formaldehyde, acetaldehyde, myrtanal, campholene aldehyde, perillaldehyde, acetone, nopinone, trans-4-hydroxynopinone and 4-acetyl-1-methylcyclohexene. Also the mono-DNPH derivatives of the dicarbonyl compounds pinonaldehyde, endolim and caronaldehyde could be identified. The MS² spectra generated in the ion trap of the mass spectrometer allowed us to distinguish between aldehydes and ketones on the basis of the characteristic fragment ion m/z 163 for the aldehydes. For the quantitative analysis of the mono-DNPH derivatives, ESI(–) in combination with single ion monitoring (SIM) detection showed the lowest detection limits. For the quantification of the dicarbonyl compounds, the acid-sensitive di-DNPH derivatives had to be formed by keeping the acidity in the acid-catalysed derivatization reaction at about 1.7 mM H₂SO₄. Detection of these dicarbonyl compounds can only be performed by APCI(–) with somewhat lesser sensitivity than by HPLC-DAD.     

Validation of HPLC Analysis of 2-Phenoxyethanol, 1-Phenoxypropan-2-ol, Methyl, Ethyl, Propyl, Butyl and Benzyl 4-Hydroxybenzoate (Parabens) in Cosmetic Products, with Emphasis on Decision Limit and Detection Capability

The Commission Decision of August 12, 2002 on the performance of analytical methods and the interpretation of results was applied to the HPLC method for the analysis of parabens, 2-phenoxyethanol and 1-phenoxypropan-2-ol in cosmetic products. This method is published in the seventh Directive 96/45/EC of the European Commission. Non-compliant concentrations, taking into account the data distribution (CC) and the probability of false negative values (CC) were determined. The repeatability and reproducibility amount to <4% and <7%, respectively. These values were obtained with blanc samples that were fortified in the laboratory. Calibration linearity was confirmed by absence of lack of fit for all seven preservatives. Matrix effects on the determinations of the preservatives in body milk or shampoo are negligible.     

Molecular clusters in two-dimensional surface-confined nanoporous molecular networks: structure, rigidity, and dynamics

The self-assembly of a series of hexadehydrotribenzo[12]annulene (DBA) derivatives has been investigated by scanning tunneling microscopy (STM) at the liquid/solid interface in the absence and presence of nanographene guests. In the absence of appropriate guest molecules, DBA derivatives with short alkoxy chains form two-dimensional (2D) porous honeycomb type patterns, whereas those with long alkoxy chains form predominantly dense-packed linear type patterns. Added nanographene molecules adsorb in the pores of the existing 2D porous honeycomb type patterns or, more interestingly, they even convert the guest-free dense-packed linear-type patterns into guest-containing 2D porous honeycomb type patterns. For the DBA derivative with the longest alkoxy chains (OC20H41), the pore size, which depends on the length of the alkoxy chains, reaches 5.4 nm. Up to a maximum of six nanographene molecules can be hosted in the same cavity for the DBA derivative with the OC20H41 chains. The host matrix changes its structure in order to accommodate the adsorption of the guest clusters. This flexibility arises from the weak intermolecular interactions between interdigitating alkoxy chains holding the honeycomb structure together. Diverse dynamic processes have been observed at the level of the host matrix and the coadsorbed guest molecules.     

Self-consistent field modeling of adsorption from polymer/surfactant mixtures

We report on the development of a self-consistent field model that describes the competitive adsorption of nonionic alkyl-(ethylene oxide) surfactants and nonionic polymer poly(ethylene oxide) (PEO) from aqueous solutions onto silica. The model explicitly describes the response to the pH and the ionic strength. On an inorganic oxide surface such as silica, the dissociation of the surface depends on the pH. However, salt ions can screen charges on the surface, and hence, the number of dissociated groups also depends on the ionic strength. Furthermore, the solvent quality for the EO groups is a function of the ionic strength. Using our model, we can compute bulk parameters such as the average size of the polymer coil and the surfactant CMC. We can make predictions on the adsorption behavior of either polymers or surfactants, and we have made adsorption isotherms, i.e., calculated the relationship between the surface excess and its corresponding bulk concentration. When we add both polymer and surfactant to our mixture, we can find a surfactant concentration (or, more precisely, a surfactant chemical potential) below which only the polymer will adsorb and above which only the surfactant will adsorb. The corresponding surfactant concentration is called the CSAC. In a first-order approximation, the surfactant chemical potential has the CMC as its upper bound. We can find conditions for which CMC < CSAC . This implies that the chemical potential that the surfactant needs to adsorb is higher than its maximum chemical potential, and hence, the surfactant will not adsorb. One of the main goals of our model is to understand the experimental data from one of our previous articles. We managed to explain most, but unfortunately not all, of the experimental trends. At the end of the article we discuss the possibilities for improving the model.     

4-metalated condensed pyrimidines: their preparation and reaction with aldehydes under Barbier-type conditions

[reaction: see text] The organometallic intermediate obtained from halogen-metal exchanges of 4-iodo-6-phenylthieno[2,3-d]pyrimidine under Barbier-type conditions was reacted with aldehydes to form the corresponding alcohols in moderate yields. The reaction involving an organolithium intermediate proceeded only at low temperature, whereas the reaction involving a magnesium ate intermediate also proceeded at room temperature. A crystal structure confirms that the expected constitutional alcohol isomer is formed, where no migration has taken place. The conditions were also suitable for 9-benzyl-6-iodopurine.     

Continuum formulation of the Scheutjens-Fleer lattice statistical theory for homopolymer adsorption from solution

Homopolymer adsorption from a dilute solution on an interacting (attractive) surface under static equilibrium conditions is studied in the framework of a Hamiltonian model. The model makes use of the density of chain ends n(1,e) and utilizes the concept of the propagator G describing conformational probabilities to locally define the polymer segment density or volume fraction phi; both n(1,e) and phi enter into the expression for the system free energy. The propagator G obeys the Edwards diffusion equation for walks in a self-consistent potential field. The equilibrium distribution of chain ends and, consequently, of chain conformational probabilities is found by minimizing the system free energy. This results in a set of model equations that constitute the exact continuum-space analog of the Scheutjens-Fleer (SF) lattice statistical theory for the adsorption of interacting chains. Since for distances too close to the surface the continuum formulation breaks down, the continuum model is here employed to describe the probability of chain configurations only for distances z greater than 2l, where l denotes the segment length, from the surface; instead, for distances z < or = 2l, the SF lattice model is utilized. Through this novel formulation, the lattice solution at z = 2l provides the boundary condition for the continuum model. The resulting hybrid (lattice for distances z < or = 2l, continuum for distances z > 2l) model is solved numerically through an efficient implementation of the pseudospectral collocation method. Representative results obtained with the new model and a direct application of the SF lattice model are extensively compared with each other and, in all cases studied, are found to be practically identical.