A new model for the viscosity of highly concentrated multi-sized particulate Bingham slurries

Bench scale tests were carried out to investigate the rheological properties of multi-sized particulate Bingham slurries at high solid concentrations ranging from 50% to 70% by weight. In addition, rheological data from Biswas et al. (2000 Biswas, A., B. K. Gandhi, S. N. Singh, and V. Seshari. 2000. Characteristics of coal ash and their role in hydraulic design of ash disposal pipelines. Indian Journal of Engineering and Materials Sciences 7:1–7.[Web of Science ®] , [Google Scholar]) and Chandel et al. (2009, 2010) have also been considered. Based on these extensive amount of rheological data, an empirical model is proposed for viscosity as a function of solid volume fraction (?), maximum solid volume fraction (?_m), median particle diameter (d₅₀), and coefficient of uniformity (C_u) using optimization and nonlinear least-square curve-fitting technique. The proposed model shows good agreement with the experimental data considered in the present study and is found to be much better than the previously developed models in predicting the viscosity of multi-sized particulate Bingham slurries at high solid concentrations.     

Patterns of Technological Innovation in Knowledge‐Intensive Business Services

Employing data from a sample of 1,161 small firms, the paper draws broad comparisons between patterns of innovation expenditure and output, innovation networking, knowledge intensity and competition within Knowledge‐Intensive Business Services (KIBS; N = 563) and manufacturing firms (N = 598). In so doing, KIBS are further disaggregated along lines proposed by Miles et al. (1995 Miles, I., Kastrinos, N., Flanagan, K., Bilderbeek, R., den Hertog, P., Huitink, W. and Bouman, M. 1995. Knowledge Intensive Business Services: Their Role as Users, Carriers and Sources of Innovation EIMS Publication No. 15, Innovation Programme, DGXIII, Luxembourg [Google Scholar]). That is, as technology‐based KIBS (t‐KIBS; N = 264) and professional KIBS (p‐KIBS; N = 299). However, detailing such broad patterns is preliminary. The principal interest of the paper is in identifying the factors associated with higher levels of innovativeness, within each sector, and the extent to which such “success” factors vary across sectors. The results of the analysis appear to offer support for some widely held beliefs about the relative roles of “softer” and “harder” sources of knowledge and technology within services and manufacturing (Tether, 2004 Tether, B. 2004. Do Services Innovate (Differently)?, Manchester: University of Manchester. CRIC Discussion Paper 66 [Google Scholar]). However, some important qualifications are also apparent.     

Exponential stability analysis of travelling waves solutions for nonlinear delayed cellular neural networks

In this paper, the exponential stability of travelling waves solutions for nonlinear cellular neural networks with distribute delays in the lattice is studied. The weighted energy method and comparison principle are employed to derive the sufficient conditions under which the networks proposed are exponentially stable. Following the study [13 X. Liu, P. Weng, and Z. Xu, Existence of traveling wave solutions in nonlinear delayed cellular neural networks, Nonlinear Anal. Real World Appl. 10(1) (2009), pp. 277–286.[Crossref], [Web of Science ®] , [Google Scholar]] on the existence of the travelling wave solutions in nonlinear delayed cellular neural networks, this paper is focused on the exponential stability of these travelling wave solutions.     

Non-Newtonian Fluids in Spout-Fluid Beds: Mass Transfer Coefficient at Low Reynolds Numbers

Following on from the work of Anabtawi et al. (2003) Anabtawi, M. Z., Hilal, N. and Al Muftah, A. E. 2003. Volumetric mass transfer coefficient in non-Newtonian fluids in spouted-fluid beds: Part I. Chem. Eng. Technol., 26: 759–764.  [Google Scholar], this study examined how the volumetric liquid-phase mass transfer coefficient, k _L a, of oxygen in air in three-phase spout-fluid beds was affected by varying the system parameters of bed height, bed diameter, gas velocity, and liquid velocity. The liquid used was 0.1% CMC solution, displaying a pseudo-plastic rheology, with 1.75 mm glass spheres as packing. The values of the Sherwood number were lower than in previous studies (Anabtawi et al., 2003 Anabtawi, M. Z., Hilal, N. and Al Muftah, A. E. 2003. Volumetric mass transfer coefficient in non-Newtonian fluids in spouted-fluid beds: Part I. Chem. Eng. Technol., 26: 759–764.  [Google Scholar]), in the range 9,000–186,000. Gas velocity had a similar effect on k _L a as in a bubble column, with results also giving good agreement with previous work on two-phase and three-phase spouted bed systems. The correlation obtained for the effect of liquid velocity on k _L a compared well with that of Schumpe et al. (1989) Schumpe, A., Deckwer, W. and Nigam, K. D. P. 1989. Gas-liquid mass transfer in three- phase fluidized beds with viscous pseudoplastic liquids. Can. J. Chem. Eng., 67: 873–877. [Crossref] , [Google Scholar]. An increase in the height of packing increased k _L a to the power of 0.319, with an increase in column diameter also causing an increase in k _L a, which is in agreement with the results of Akita and Yoshida (1973) Akita, K. and Yoshida, F. 1973. Gas hold-up and volumetric mass transfer coefficient in bubble columns. Ind. Eng. Chem. Process Des. Dev., 12: 76–80. [Crossref] , [Google Scholar].     

Magnetic-saturation zone model for two semipermeable cracks in magneto-electro-elastic medium

Extension of the PS model (Gao et al. [1 H. Gao, T. Y. Zhang, and P. Tong, “Local and global energy release rates for an electrically yielded crack in a piezoelectric ceramic,” J. Mech. Phy. Solids, vol. 45, pp. 491–510, 1997.[Crossref], [Web of Science ®] , [Google Scholar]]) in piezoelectric materials and the SEMPS model (Fan and Zhao [2 C. Y. Fan and M. H. Zhao, “Nonlinear fracture of 2-D magnetoelectroelastic media: analytical and numerical solutions,” Int. J. Solids Struct., vol. 48, pp. 2383–2392, 2011.[Crossref], [Web of Science ®] , [Google Scholar]]) in MEE materials, is proposed for two semi-permeable cracks in a MEE medium. It is assumed that the magnetic yielding occurs at the continuation of the cracks due to the prescribed loads. We have model these crack continuations as the zones with cohesive saturation limit magnetic induction. Stroh's formalism and complex variable techniques are used to formulate the problem. Closed form analytical expressions are derived for various fracture parameters. A numerical case study is presented for BaTiO₃ ? CoFe₂O₄ ceramic cracked plate.     

Designing an optimal water quality monitoring network for river systems using constrained discrete optimization via simulation

The problem of designing a water quality monitoring network for river systems is to find the optimal location of a finite number of monitoring devices that minimizes the expected detection time of a contaminant spill event while guaranteeing good detection reliability. When uncertainties in spill and rain events are considered, both the expected detection time and detection reliability need to be estimated by stochastic simulation. This problem is formulated as a stochastic discrete optimization via simulation (OvS) problem on the expected detection time with a stochastic constraint on detection reliability; and it is solved with an OvS algorithm combined with a recently proposed method called penalty function with memory (PFM). The performance of the algorithm is tested on the Altamaha River and compared with that of a genetic algorithm due to Telci, Nam, Guan and Aral (2009) Telci, I. T., K. Nam, J. Guan, and M.M. Aral, 2009. “Optimal Water Quality Monitoring Network Design for River Systems.” Journal of Environmental Management, 90 (3–4): 2987–2998. doi: 10.1016/j.jenvman.2009.04.011[Crossref], [PubMed], [Web of Science ®] , [Google Scholar].     

Supercritical Antisolvent Process Applied to the Pharmaceutical Industry

Pharmaceutical preparations are the final product of a technological process that gives the drugs the characteristics appropriate for easy administration, proper dosage, and enhancement of the therapeutic efficacy. The design of pharmaceutical preparations in nanoparticulate form has emerged as a new strategy for drug delivery (Pasquali, Bettini, and Giordano, 2006 Pasquali , I. , R. Bettini , and F. Giordano . 2006 . Solid-state chemistry and particle engineering with supercritical fluids in pharmaceutics . Eur. J. Pharm. Sci. 27 : 299 – 310 .[Crossref], [PubMed], [Web of Science ®] , [Google Scholar]). Particle size (PS) and particle size distribution (PSD) are critical parameters that determine the rate of dissolution of the drug in the biological fluids and, hence, have a significant effect on the bioavailability of those drugs that have poor solubility in water, for which the dissolution is the rate-limiting step in the absorption process (Perrut, Jung, and Leboeuf, 2005 Perrut , M. , J. Jung , and F. Leboeuf . 2005 . Enhancement of dissolution rate of poorly-soluble active ingredients by supercritical fluid processes: Part I: Micronization of neat particles . Int. J. Pharm. 288 : 3 – 10 .[Crossref], [PubMed], [Web of Science ®] , [Google Scholar]; Van Nijlen et al., 2003 Van Nijlen , T., G. Van Den Mooter , R. Kinget , P. Augustijns , N. Blaton , and K. Brennan . 2003 . Improvement of the dissolution rate of artemisinin by means of supercritical fluid technology and solid dispersions . Int. J. Pharm. 254 : 173 – 181 .[Crossref], [PubMed], [Web of Science ®] , [Google Scholar]). Supercritical antisolvent (SAS) processes have been widely used to precipitate active pharmaceutical ingredients (APIs) (Chattopadhyay and Gupta, 2001 Chattopadhyay , P. , and R. B. Gupta . 2001 . Production of antibiotic nanoparticles using supercritical CO₂ as antisolvent with enhanced mass transfer . Ind. Eng. Chem. Res. 40 : 3530 – 3539 .[Crossref], [Web of Science ®] , [Google Scholar]; Rehman et al., 2001 Rehman , M. , B. Y. Shekunov , P. York , and P. Colthorpe . 2001 . Solubility and precipitation of nicotinic acid in supercritical carbon dioxide . J. Pharm. Sci. 90 : 1570 – 1582 .[Crossref], [PubMed], [Web of Science ®] , [Google Scholar]) with a high level of purity, suitable dimensional characteristics, narrow PSD, and spherical morphologies. The SAS process is based on the particular properties of the supercritical fluids (SCFs). These fluids have diffusivities two orders of magnitude larger than those of liquids, resulting in a faster mass transfer rate SCF properties (solvent power and selectivity) can be also adjusted continuously by altering the experimental conditions (temperature and pressure). As a consequence, SCFs can be removed from the process by a simple change from the supercritical to room conditions, which avoids difficult post-treatments of waste liquid streams. Carbon dioxide (CO₂) at supercritical conditions, among all possible SCFs, is largely used because of its relatively low critical temperature (31.1°C) and pressure (73.8 bar), low toxicity, and low cost. In this article, we show some results about processed antibiotics (ampicillin and amoxicillin), two of the world's most widely prescribed antibiotics, when they are dissolved in 1-methyl-2-pyrrolidone (NMP) and carbon dioxide is used as antisolvent.     

Comparative Assessment of Two Constitutive Models for Superelastic Shape-Memory Wires Against Experimental Measurements

Two constitutive models representative of two well-known modeling techniques for superelastic shape-memory wires are reviewed. The first model has been proposed by Kim and Aberayatne in the framework of finite thermo-elasticity with non-convex energy [1 S.-J. Kim and R. Abeyaratne, On the effect of the heat generated during a stress-induced thermoelastic phase transformation, Continuum Mech. Thermodyn., vol. 7, pp. 311–332, 1995.[Crossref], [Web of Science ®] , [Google Scholar]]. In the present article this model has been modified in order to take into account the difference between elastic moduli of austenite and martensite and to introduce the isothermal approximation proposed in [1 S.-J. Kim and R. Abeyaratne, On the effect of the heat generated during a stress-induced thermoelastic phase transformation, Continuum Mech. Thermodyn., vol. 7, pp. 311–332, 1995.[Crossref], [Web of Science ®] , [Google Scholar]]. The second model has been developed by Auricchio et al. within the theory of irreversible thermodynamics with internal variables [2 F. Auricchio, D. Fugazza, and R. DesRoches, Rate-dependent thermo-mechanical modelling of superelastic shape-memory alloys for seismic applications, J. Intell. Mater. Syst. Struct., vol. 19, pp. 47–61, 2008.[Crossref], [Web of Science ®] , [Google Scholar]]. Both models are temperature and strain rate dependent and they take into account thermal effects. The focus in this article is on investigating how the two models compare with experimental data obtained from testing superelastic NiTi wires used in the design of a prototypal anti-seismic device [3 M. Indirli and M.G. Castellano, Shape memory alloy devices for the structural improvement of masonry heritage structures, Int. J. Archit. Heritage, vol. 2, no. 2, pp. 93–119, 2008.[Taylor &; Francis Online], [Web of Science ®] , [Google Scholar], 4 A. Chiozzi, M. Merlin, R. Rizzoni, and A. Tralli, Experimental comparison for two one-dimensional constitutive models for shape memory alloy wires used in anti-seismic applications. In: European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS 2012), e-Book Full Papers, pp. 4672–4682, 2012. [Google Scholar]]. After model calibration and numerical implementation, numerical simulations based on the two models are compared with data obtained from uniaxial tensile tests performed at two different temperatures and various strain rates.     

New relation between input and output fields based on Collins diffraction formula and the Wigner function in entangled state representation

Based on the correspondence between the Collins diffraction formula (optical Fresnel transform) and the transform matrix element of a three-parameter two-mode squeezing operator in the entangled state representation 1 Fan, H-Y and Lu, H-L. 2006. Opt. Lett., 31: 2622–2624.  [Google Scholar] we further explore the relationship between output field intensity determined by the Collins formula and the input field's probability distribution along an infinitely thin phase space strip, both in the spatial domain and the frequency domain. The entangled Wigner function is introduced for recapitulating the result.     

Dense-Phase Pneumatic Conveying of Slush Ice Through Plastic Pipes

Ice is being used in certain deep mines to transport refrigeration to underground areas. Research has been carried out previously into the pipeline conveying characteristics of ice with air, but there remains a lack of knowledge about some aspects of this complex flow. Previous articles (Sheer, 1995 Sheer , T. J. 1995 . Pneumatic conveying of ice particles through mine-shaft pipelines . Powder Technol. 85 : 203 – 219 . [Google Scholar]; Sheer et al., 2001 Sheer , T. J. , R. Ramsden , & M. Butterworth . 2001. The design of pipelines for transporting ice into deep mines. In Handbook of Conveying and Handling of Particulate Solids , eds. A. Levy and H. Kalman . New York : Elsevier Science. pp. 425–433. [Google Scholar]) have described experimental results on the pneumatic conveying of ice in particulate form (“hard” ice) through successive long horizontal and vertical sections of pipelines into mines. More recent research has been carried out to determine the conveying characteristics of “slush” ice that resembles wet snow, with an ice mass fraction range of 65–75%. Laboratory pneumatic conveying tests with slush ice were conducted through three horizontal plastic pipelines with inner diameters of 43, 54, and 69 mm, each pipeline being approximately 50 m long and including various bends. The tests yielded numerical and photographic data that were used to investigate the conveying characteristics of slush ice (including flow regime transition to plug flow and pressure gradients) and to compare them with the previous results for particulate ice. It was found that the conveying characteristics of the slush depend strongly on the water content. Correlations are proposed for multiphase friction factors.     

Prediction of Concentration Distribution in Pipeline Flow of Highly Concentrated Slurry

The Kaushal and Tomita's (2002a Kaushal , D. R. , and Y. Tomita . 2002a . An improved method for predicting pressure drop along slurry pipeline . Particulate Science and Technology 20 : 1 – 20 .[Taylor & Francis Online] , [Google Scholar]) model, which has already been found satisfactory for broadly graded multisized particulate zinc tailings slurry with moderate concentration up to 26%, flow velocity up to 3.5 m/s in 105 mm diameter pipe, mean diameter 140 μm, and geometric standard deviation of particles of 4.0, is tested for concentration distribution data collected by Kaushal et al. (2005 Kaushal , D. R. , K. Sato , T. Toyota , K. Funatsu , and Y. Tomita . 2005 . Effect of particle size distribution on pressure drop and concentration profile in pipeline flow of highly concentrated slurry . International Journal of Multiphase Flow. 31 : 809 – 823 .[Crossref], [Web of Science ®] , [Google Scholar]) on two sizes of glass beads, of which mean diameter and geometric standard deviation are 440 μm and 1.2, and 125 μm and 1.15, respectively, with concentration up to 50% and flow velocity up to 5 m/s in 54.9 mm diameter pipe. Kaushal and Tomita's (2002a) model gives more asymmetric concentration distributions. A modified model is proposed by alleviating some of the restrictive assumptions used in the existing model. Comparison of experimental data by Kaushal et al. (2005 Kaushal , D. R. , K. Sato , T. Toyota , K. Funatsu , and Y. Tomita . 2005 . Effect of particle size distribution on pressure drop and concentration profile in pipeline flow of highly concentrated slurry . International Journal of Multiphase Flow. 31 : 809 – 823 .[Crossref], [Web of Science ®] , [Google Scholar]), Gillies and Shook (1994 Gillies , R. G. , and C. A. Shook . 1994 . Concentration distributions of sand slurries in horizontal pipe flow . Particulate Science Technology 12 : 45 – 69 .[Taylor & Francis Online], [Web of Science ®] , [Google Scholar]), and Matousek (2009 Matousek , V. 2009 . Concentration profiles and solids transport above stationary deposit in enclosed conduit . Journal of Hydraulic Engineering 135 : 1101 – 1106 .[Crossref], [Web of Science ®] , [Google Scholar]) with the proposed model is satisfactory.     

Interactions and Collisions of Bubbles in Thermocapillary Motion: A 3D Study

Two- and three-dimensional simulations (created using volume of fluid-fluent) concerning the rise and interactions of two/multiple thermocapillary bubbles arranged horizontally and perpendicular to a hot surface are investigated and presented in this paper. The results indicate that thermocapillary bubble agglomeration can occur in zero gravity. Furthermore, the temperature gradient and bubble diameter were found to have a major impact on the collision between bubbles. The results of Nas and Tryggvason (1993 Nas, S., and G. Tryggvason. 1993. Computational investigation of the thermal migration of bubbles and drops. 175 Fluid Mechanics Phenomena In Microgravity 71–83. [Google Scholar]) in their three-dimensional numerical study reported that no such collisions could occur in zero gravity and that bubbles repel each other due to the cold liquid carried between particles during migration. Their results contrast with both the present results and those recorded onboard the Chinese 22nd recoverable satellite experiment by Kang et al. (2008 Kang, Q., H. L. Cui, L. Hu, and L. Duan. 2008. On-board experimental study of bubble thermocapillary migration in a recoverable satellite. Microgravity Science And Technology 20:67–71.[Crossref], [Web of Science ®] , [Google Scholar]), who observed a total of 19 coalescences between the air bubbles injected in the direction of the temperature gradient of the stagnant heated liquid.     

Hyperelastic analysis based on a polygonal finite element method

In this contribution, we present a novel polygonal finite element method applied to hyperelastic analysis. For generating polygonal meshes in a bounded period of time, we use the adaptive Delaunay tessellation (ADT) proposed by Constantinu et al. [1 A. Constantinu, P. Steinmann, T. Bobach, G. Farin, and G. Umlauf, The adaptive Delaunay tessellation: A neighborhood covering meshing technique., Comput. Mech., vol. 42, no. 5, pp. 655–669, 2008.[Crossref], [Web of Science ®] , [Google Scholar]]. ADT is an unstructured hybrid tessellation of a scattered point set that minimally covers the proximal space around each point. In this work, we have extended the ADT to nonconvex domains using concepts from constrained Delaunay triangulation (CDT). The proposed method is thus based on a constrained adaptive Delaunay tessellation (CADT) for the discretization of domains into polygonal regions. We involve the metric coordinate (Malsch) method for obtaining the interpolation over convex and nonconvex domains. For the numerical integration of the Galerkin weak form, we resort to classical Gaussian quadrature based on triangles. Numerical examples of two-dimensional hyperelasticity are considered to demonstrate the advantages of the polygonal finite element method.     

A cuckoo search algorithm by Lévy flights for solving reliability redundancy allocation problems

A new metaheuristic optimization algorithm, called cuckoo search (CS), was recently developed by Yang and Deb (2009 Yang, X.S. and Deb, S., 2009. Cuckoo search via Lévy flights. In: Proceedings of world congress on nature &; biologically inspired computing (NaBIC 2009), 9–11 December 2009, Coimbatore, India. Piscataway, NJ: IEEE Press, 210–214.[Crossref] , [Google Scholar], 2010 Yang, X.S. and Deb, S., 2010. Engineering optimisation by cuckoo search. International Journal of Mathematical Modelling and Numerical Optimisation, 1, 330–343. doi: 10.1504/IJMMNO.2010.035430[Crossref] , [Google Scholar]). This article uses CS and Lévy flights to solve the reliability redundancy allocation problem. The redundancy allocation problem involves setting reliability objectives for components or subsystems in order to meet the resource consumption constraint, e.g. the total cost. The difficulties facing the redundancy allocation problem are to maintain feasibility with respect to three nonlinear constraints, namely, cost, weight and volume-related constraints. The redundancy allocation problems have been studied in the literature for decades, usually using mathematical programming or metaheuristic optimization algorithms. The performance of the algorithm is tested on five well-known reliability redundancy allocation problems and is compared with several well-known methods. Simulation results demonstrate that the optimal solutions obtained by CS are better than the best solutions obtained by other methods.     

Rank 2 proximal Cantor systems are residually scrambled

Downarowicz and Maass [7 T. Downarowicz and A. Maass, Finite-rank Bratteli–Vershik diagrams are expansive, Ergod. Th. Dynam. Sys. 28 (2008), pp. 739–747.[Web of Science ®] , [Google Scholar]] proposed topological ranks for all homeomorphic Cantor minimal dynamical systems using properly ordered Bratteli diagrams. In this study, we adopt this definition to the case of the essentially minimal zero-dimensional systems. We consider the cases in which topological ranks are 2 and unique minimal sets are fixed points. Akin and Kolyada [2 E. Akin and S. Kolyada, Li–Yorke sensitivity, Nonlinearity 16(4) (2003), pp. 1421–1433.[Crossref], [Web of Science ®] , [Google Scholar]], had shown that if the unique minimal set of an essentially minimal system is a fixed point, then the system must be proximal. The finite topological rank implies expansiveness; furthermore, in the case of proximal Cantor systems with topological rank 2, the expansiveness is always from the lowest degree. Rank 2 proximal Cantor systems are residually scrambled. We present a necessary and sufficient condition for the unique ergodicity of these systems. In addition, we show that the number of ergodic measures of the systems that are topologically mixing can be 1 and 2. Moreover, we present examples that are topologically weakly mixing, not topologically mixing, and uniquely ergodic. Finally, we show that the number of ergodic measures of the systems that are not weakly mixing can be 1 and 2.     

Minimising makespan heuristics for scheduling a single batch machine processing machine with non-identical job sizes

In this paper, the problem of minimising maximum completion time on a single batch processing machine is studied. A batch processing is performed on a machine which can simultaneously process several jobs as a batch. The processing time of a batch is determined by the longest processing time of jobs in the batch. The batch processing machine problem is encountered in many manufacturing systems such as burn-in operations in the semiconductor industry and heat treatment operations in the metalworking industries. Heuristics are developed by iterative decomposition of a mixed integer programming model, modified from the successive knapsack problem by Ghazvini and Dupont (1998 Ghazvini, F.J. and Dupont, L. 1998. Minimising mean flow times criteria on a single batch processing machine with non-identical jobs sizes. International Journal of Production Economics, 55: 273–280. [Crossref], [Web of Science ®] , [Google Scholar], Minimising mean flow times criteria on a single batch processing machine with non-identical jobs sizes. International Journal of Production Economics 55: 273–280) and the waste of batch clustering algorithm by Chen, Du, and Huang (2011 Chen, H., Du, B. and Huang, G.Q. 2011. Scheduling a batch processing machine with non-identical job sizes: a clustering perspective. International Journal of Production Research, 49(19): 5755–5778. [Taylor &; Francis Online], [Web of Science ®] , [Google Scholar], Scheduling a batch processing machine with non-identical job sizes: a clustering perspective. International Journal of Production Research 49 (19): 5755–5778). Experimental results show that the suggested heuristics produce high-quality solutions comparable to those of previous heuristics in a reasonable computation time.     

Flexible Expectile Regression in Reproducing Kernel Hilbert Spaces

Expectile, first introduced by Newey and Powell in 1987 Newey, W. K., and Powell, J. L. (1987), “Asymmetric Least Squares Estimation and Testing,” Econometrica, 55, 819–847.[Crossref], [Web of Science ®] , [Google Scholar] in the econometrics literature, has recently become increasingly popular in risk management and capital allocation for financial institutions due to its desirable properties such as coherence and elicitability. The current standard tool for expectile regression analysis is the multiple linear expectile regression proposed by Newey and Powell in 1987 Newey, W. K., and Powell, J. L. (1987), “Asymmetric Least Squares Estimation and Testing,” Econometrica, 55, 819–847.[Crossref], [Web of Science ®] , [Google Scholar]. The growing applications of expectile regression motivate us to develop a much more flexible nonparametric multiple expectile regression in a reproducing kernel Hilbert space. The resulting estimator is called KERE, which has multiple advantages over the classical multiple linear expectile regression by incorporating nonlinearity, nonadditivity, and complex interactions in the final estimator. The kernel learning theory of KERE is established. We develop an efficient algorithm inspired by majorization-minimization principle for solving the entire solution path of KERE. It is shown that the algorithm converges at least at a linear rate. Extensive simulations are conducted to show the very competitive finite sample performance of KERE. We further demonstrate the application of KERE by using personal computer price data. Supplementary materials for this article are available online.