Choice behavior in interactive multiple-criteria decision making

Choice behavior in an interactive multiple-criteria decision making environment is examined experimentally. A free search discrete visual interactive reference direction approach was used on a microcomputer by management students to solve two realistic and relevant multiple-criteria decision problems. The results revealed persistent patterns of intransitive choice behavior, and an unexpectedly rapid degree of convergence of the reference direction approach on a preferred solution. The results can be explained using Tversky' [20] additive utility difference model and Kahneman-Tversky's [5] prospect theory. The implications of the results for the design of interactive multiple-criteria decision procedures are discussed.     

Modification of cycloolefin copolymer and poly(vinyl chloride) surfaces by superimposition of nano- and microstructures

Cycloolefin copolymer (COC) and poly(vinyl chloride) (PVC) surfaces were patterned with nanopillars or with microbumps on which nanopillars were superimposed. The area of patterned surfaces was several square centimeters. Patterning was achieved by applying nanoporous anodized aluminum oxide (AAO) membrane as a mask in injection molding or imprinting. Nanostructures superimposed on microstructures were achieved by patterning the AAO mask with microstructures before anodization. Micro- and nanometer-sized structures could then be transferred simultaneously to polymer surfaces. Structures were characterized by SEM, AFM, and contact profilometry. The effect of different-sized structures on properties of the polymer surface was studied by contact angle measurements. Relative to the smooth surface, the increase in water contact angle on a COC surface with nanostructures superimposed on microstructures was up to 50°.     

Predicting the drug concentration in starch acetate matrix tablets from ATR-FTIR spectra using multi-way methods

The amounts of drug and excipient were predicted from ATR-FTIR spectra using two multi-way modelling techniques, parallel factor analysis (PARAFAC) and multi-linear partial least squares (N-PLS). Data matrices consisted of dissolved and undissolved parallel samples having different drug content and spectra, which were collected at axially cut surface of the flat-faced matrix tablets. Spectra were recorded comprehensively at different points on the axially cut surface of the tablet. The sample drug concentrations varied between 2 and 16% v/v. The multi-way methods together with ATR-FTIR spectra seemed to represent an applicable method for the determination of drug and excipient distribution in a tablet during the release process. The N-PLS calibration method was more robust for accurate quantification of the amount of components in the sample whereas the PARAFAC model provided approximate relative amounts of components.     

Co3O4-SiO2 nanocomposite: a very active catalyst for CO oxidation with unusual catalytic behavior

A high surface area Co(3)O(4)-SiO(2) nanocomposite catalyst has been prepared by use of activated carbon as template. The Co(3)O(4)-SiO(2) composite, the surface of which is rich in silica and Co(II) species compared with normal Co(3)O(4), exhibited very high activity for CO oxidation even at a temperature as low as -76 °C. A rather unusual temperature-dependent activity curve, with the lowest conversion at about 80 °C, was observed with a normal feed gas (H(2)O content ~3 ppm). The U-shape of the activity curve indicates a negative apparent activation energy over a certain temperature range, which has rarely been observed for the heterogeneously catalyzed oxidation of CO. Careful investigation of the catalytic behavior of Co(3)O(4)-SiO(2) catalyst led to the conclusion that adsorption of H(2)O molecules on the surface of the catalyst caused the unusual behavior. This conclusion was supported by in situ diffuse reflectance Fourier transform infrared (DRIFT) spectroscopic experiments under both normal and dry conditions.     

Photocrosslinkable polyesters and poly(ester anhydride)s for biomedical applications

Crosslinking is a feasible way to prepare biodegradable polymers with potential in biomedical applications such as controlled release of active agents and tissue engineering. A synthesis route in which functional telechelic aliphatic polyester oligomers are used as precursors for the preparation of crosslinked polyesters and poly(ester anhydride)s is described. Mechanical properties, degradation characteristics and rate, and bioactivity can be modified widely by controlling the chemical composition and architecture of the crosslinkable oligomers. In tissue engineering, photocrosslinking allows to use crosslinkable oligomers in advanced manufacturing techniques like micromolding in capillaries, stereolithography and two-photon polymerization.     

Performance of multicomponent self-organizing regression (MCSOR) in QSAR,QSPR, and multivariate calibration: comparison with partial least-squares (PLS) and validation with large external data sets

A novel method for underdetermined regression problems, multicomponent self-organizing regression (MCSOR), has been recently introduced. Here, its performance is compared with partial least-squares (PLS), which is perhaps the most widely adopted multivariate method in chemometrics. A potpourri of models is presented, and MCSOR appears to provide highly predictive models that are comparable with or better than the corresponding PLS models in large internal (leave-one-out, LOO) and pseudo-external (leave-many-out, LMO) validation tests. The “blind” external predictive ability of MCSOR and PLS is demonstrated employing large melting point, factor Xa, log?P and log?S data sets. In a nutshell, MCSOR is fast, conceptually simple (employing multiple linear regression, MLR, as a statistical tool), and applicable to all kinds of multivariate problems with single Y-variable.     

Automated quantum mechanical total line shape fitting model for quantitative NMR-based profiling of human serum metabolites

An automated quantum mechanical total line shape (QMTLS) fitting model was implemented for quantitative nuclear magnetic resonance (NMR)-based profiling of 42 metabolites in ultrafiltrated human serum samples. Each metabolite was described by a set of chemical shifts, J-couplings, and line widths. These parameters were optimized for each metabolite in each sample by iteratively minimizing the difference between the calculated and the experimental spectrum. In total, 92.0 to 98.1 % of the signal intensities in the experimental spectrum could be explained by the calculated spectrum. The model was validated by comparison to signal integration of metabolites with isolated signals and by means of standard additions. Metabolites present at average concentration higher than 50 μM were quantified with average absolute relative error less than 10 % when using different initial parameters for the fitting procedure. Furthermore, the biological applicability of the QMTLS model was demonstrated on 287 samples from an intervention study in 37 human volunteers undergoing an exercise challenge. Our automated QMTLS model was able to cope with the large dynamic range of metabolite concentrations in serum and proved to be suitable for high-throughput analysis.

Resource allocation for performance improvement

In this paper the problem of allocating resources among Decision Making Units is considered. This study covers the case in which several homogeneous units are operating under the supervision of a central unit. The resource allocation is carried out by the DM (central unit) in such a way that the overall performance of the system is improved. Performance is defined by means of a convex combination of the ratio of the efficiencies before and after the resource allocation. It is assumed that each unit is allowed to modify its resources within the current production possibility set. A novel model is proposed which aims at achieving the best performance of the system. The method is capable of dealing with some additional constraints, imposed by the DM. The model is illustrated by a simple numerical example and a real application.     

Fast Growing Solutions of Linear Differential Equations in the Unit Disc

For $n \in \mathbb{N}$ , the n-order of an analytic function f in the unit disc D is defined by $$\sigma _{{{M,n}}} (f) = {\mathop {\lim \sup }\limits_{r \to 1^{ - } } }\frac{{\log ^{ + }_{{n + 1}} M(r,f)}} {{ - \log (1 - r)}},$$ where log⁺ x  =  max{log x, 0}, log ⁺ ₁ x  =  log ⁺ x, log ⁺ _n+1 x  =  log ⁺ log ⁺ _n x, and M(r, f) is the maximum modulus of f on the circle of radius r centered at the origin. It is shown, for example, that the solutions f of the complex linear differential equation $$f^{{(k)}} + a_{{k - 1}} (z)f^{{(k - 1)}} + \cdots + a_{1} (z)f^{\prime} + a_{0} (z)f = 0,\quad \quad \quad (\dag)$$ where the coefficients are analytic in D, satisfy σ _M,n+1(f)  ≤  α if and only if σ _M,n (a _j )  ≤  α for all j  =  0, ..., k ? 1. Moreover, if q ∈{0, ..., k ? 1} is the largest index for which $\sigma _{M,n} ( a_{q}) = {\mathop {\max }\limits_{0 \leq j \leq k - 1} }{\left\{ {\sigma _{{M,n}} {\left( {a_{j} } \right)}} \right\}}$ , then there are at least k ? q linearly independent solutions f of ( $\dag$ ) such that σ _M,n+1(f) = σ _M,n (a _q ). Some refinements of these results in terms of the n-type of an analytic function in D are also given.     

Performance of multicomponent self-organizing regression (MCSOR) in QSAR, QSPR, and multivariate calibration: comparison with partial least-squares (PLS) and validation with large external data sets

A novel method for underdetermined regression problems, multicomponent self-organizing regression (MCSOR), has been recently introduced. Here, its performance is compared with partial least-squares (PLS), which is perhaps the most widely adopted multivariate method in chemometrics. A potpourri of models is presented, and MCSOR appears to provide highly predictive models that are comparable with or better than the corresponding PLS models in large internal (leave-one-out, LOO) and pseudo-external (leave-many-out, LMO) validation tests. The "blind" external predictive ability of MCSOR and PLS is demonstrated employing large melting point, factor Xa, log P and log S data sets. In a nutshell, MCSOR is fast, conceptually simple (employing multiple linear regression, MLR, as a statistical tool), and applicable to all kinds of multivariate problems with single Y-variable.