Continuous prediction of manufacturing performance throughout the production lifecycle

We describe methods for continual prediction of manufactured product quality prior to final testing. In our most expansive modeling approach, an estimated final characteristic of a product is updated after each manufacturing operation. Our initial application is for the manufacture of microprocessors, and we predict final microprocessor speed. Using these predictions, early corrective manufacturing actions may be taken to increase the speed of expected slow wafers (a collection of microprocessors) or reduce the speed of fast wafers. Such predictions may also be used to initiate corrective supply chain management actions. Developing statistical learning models for this task has many complicating factors: (a) a temporally unstable population (b) missing data that is a result of sparsely sampled measurements and (c) relatively few available measurements prior to corrective action opportunities. In a real manufacturing pilot application, our automated models selected 125 fast wafers in real-time. As predicted, those wafers were significantly faster than average. During manufacture, downstream corrective processing restored 25 nominally unacceptable wafers to normal operation.     

Metallurgical Studies of Austenitic Stainless Steel 304 under Warm Deep Drawing

Austenitic stainless steel 304 was deep drawn with different blank diameters under warm conditions using 20 t hydraulic press. A number of deep drawing experiments both at room temperature and at 150 ℃ were conducted to study the metallography. Also, tensile test experiments were conducted on a universal testing machine up to 700 ℃ and the broken specimens were used to study the fractography of the material using scanning electron microscopy in various regions. The microstructure changes were observed at limiting draw ratio （LDR） when the cup is drawn at different temperatures. In austenitic stainless steel, martensite formation takes place that is not only affected by temperature, hut also influenced by the rate at which the material is deformed. In austenitic stainless steel 304, dynamic strain regime appears above 300 ℃ and it decreases the formability of material due to brittle fracture as studied in its fractography. From the metallographic studies, the maximum LDR of the material is observed at 150 ℃ before dynamic strain regime. It is also observed that at 150 ℃, grains are coarse in the drawn cups at LDR.     

Effect of preparation methods on the performance of Ni/Al2O3 catalysts for aqueous-phase reforming of ethanol: Part I-catalytic activity

The effect of preparation method on the performance of Ni/Al₂O₃ catalysts for aqueous-phase reforming of ethanol (EtOH) has been investigated. The first catalyst was prepared by a sol–gel (SG) method and for the second one the Al₂O₃ support was made by a solution combustion synthesis (SCS) route and then the metal was loaded by standard wet impregnation. The catalytic activity of these catalysts of different Ni loading was compared with a commercial Al₂O₃ supported Ni catalyst [CM (10%)] at different temperatures, pressures, feed flow rates, and feed concentrations. Based on the product distribution, the proposed reaction pathway is a mixture of dehydrogenation of EtOH to CH₃CHO followed by C–C bond breaking to produce CO + CH₄ and oxidation of CH₃CHO to CH₃COOH followed by decarbonylation to CO₂ + CH₄. CH₄(C₂H₆ and C₃H₈) also can form via Fischer–Tropsch reactions of CO/CO₂ with H₂. The CH₄ (C₂H₆ and C₃H₈) reacts to form hydrogen and carbon monoxide through steam reforming, while CO converts to CO₂ mostly through the water–gas shift reaction (WGSR). SG catalysts showed poorer WGSR activity than the SCS catalysts. The activation energies for H₂ and CO₂ production were 153, 155 and 167 kJ/mol and 158, 160 and 169 kJ/mol for SCS (10%), SG (10%), and CM (10%) samples, respectively.     

Aperture Field Estimation in E Plane and H-Plane T-Junctions Using Field Theoretic Analysis

Wireless Personal Communications - This paper presents method of moment based analysis of E Plane and H-Plane T-junctions to find the field distributions on different transverse and longitudinal...     

Inhalation risk of indoor aerosol spray products commercially available in Bangladesh

Today, urbanised people spend most of their time in indoor environments, and the risk of exposure to toxic materials in an indoor environment is of great concern. This paper considers sources for the inhalation risk of aerosol sprays by measuring the particle size distributions of some of the most popular products that are commercially available in Bangladesh, and examines the degree of toxicity and the health risk via the respirability of the products. Results show that thoracic and respirable fractions of the suspended particulate matter are present in the tested samples and hence the spray products may pose a significant health risk.     

Design of viral marketing strategies for product cross-sell through social networks

In this paper, we introduce a novel and generalized version of the influence maximization problem in social networks, which we call as Budgeted Influence Maximization with Cross-sell of Products (B-IMCP), and it considers simultaneously the following three practical aspects: (i) Often cross-sell among products is possible, (ii) Product-specific costs (and benefits) for promoting the products have to be considered, and (iii) Since a company often has budget constraints, the initial seeds have to be chosen within a given budget. In particular, we consider that the cross-sell relationships among the products of a single company are given by an arbitrary bipartite graph. We explore two variants of cross-sell, one weak and one strong, and also assume product-specific costs and benefits. This leads to two different versions of the B-IMCP problem. Given a fixed budget, one of the key issues associated with each version of the B-IMCP problem is to choose the initial seeds within this budget not only for the individual products, but also for promoting cross-sell phenomenon among these products. The following are the specific contributions of this paper: (i) We propose a novel influence propagation model to capture both the cross-sell phenomenon and the costs–benefits for the products; (ii) For each version of the B-IMCP problem, we note that the problem turns out to be NP-hard, and then, we present a simple greedy approximation algorithm for the same. We derive the approximation ratio of this greedy algorithm by drawing upon certain key results from the theory of matroids; (iii) We then outline three heuristics based on well-known concepts from the sociology literature; and (iv) Finally, we experimentally compare and contrast the proposed algorithms and heuristics using certain well-known social network data sets such as WikiVote trust network, Epinions, and Telco call detail records data. Based on the experiments, we consistently found that the stronger the cross-sell relationship between the products, the larger the overlap between the seeds of these products and lesser the distances among the corresponding non-overlapping seeds.