MSC-clustering and forward stepwise regression for virtual metrology in highly correlated input spaces

Increasingly semiconductor manufacturers are exploring opportunities for virtual metrology (VM) enabled process monitoring and control as a means of reducing non-value added metrology and achieving ever more demanding wafer fabrication tolerances. However, developing robust, reliable and interpretable VM models can be very challenging due to the highly correlated input space often associated with the underpinning data sets. A particularly pertinent example is etch rate prediction of plasma etch processes from multichannel optical emission spectroscopy data. This paper proposes a novel input-clustering based forward stepwise regression methodology for VM model building in such highly correlated input spaces. Max Separation Clustering (MSC) is employed as a pre-processing step to identify a reduced srt of well-conditioned, representative variables that can then be used as inputs to state-of-the-art model building techniques such as Forward Selection Regression (FSR), Ridge regression, LASSO and Forward Selection Ridge Regression (FCRR). The methodology is validated on a benchmark semiconductor plasma etch dataset and the results obtained are compared with those achieved when the state-of-art approaches are applied directly to the data without the MSC pre-processing step. Significant performance improvements are observed when MSC is combined with FSR (13%) and FSRR (8.5%), but not with Ridge Regression (-1%) or LASSO (-32%). The optimal VM results are obtained using the MSC-FSR and MSC-FSRR generated models. © 2012 IEEE.

Engineering Simulations

Over recent years, ionic liquids have emerged as a class of novel fluids that have inspired the development of a number of new products and processes. The ability to design these materials with specific functionalities and properties means that they are highly relevant to the growing philosophy of chemical-product design. This is particularly appropriate in the context of a chemical industry that is becoming increasingly focussed on small-volume, high-value added products with relatively short times to market. To support such product and process development, a number of tools can be utilised. A key requirement is that the tool can predict the physical properties and activity coefficients of multi-component mixtures and, if required, model the process in which the materials will be used. Multi-scale simulations that span density functional theory (DFT) to process-engineering computations can address the relevant time and length scales and have increased in usage with the availability of cheap and powerful computers. Herein we will discuss the area of engineering calculations relating to the design of ionic liquid processes, that is, the computational tools that bridge this gap and allow for process simulation tools to utilise and assist in the design of ionic liquids. It will be shown that, at present, it is possible to use available tools to estimate many important properties of ionic liquids and mixtures containing them with a sufficient level of accuracy for preliminary design and selection.

Waste Not, Want Not: Managing Perishables in Small and Medium Retail Enterprises

Purpose of this paper:
Recent literature indicates that around one third of perishable products finish as waste (Mena et al., 2014): 60% of this waste can be classified as avoidable (EC, 2010) suggesting logistics and operational inefficiencies along the supply chain. In developed countries perishable products are predominantly wasted in wholesale and retail (Gustavsson et al., 2011) due to customer demand uncertainty the errors and delays in the supply chain (Fernie and Sparks, 2014). While research on logistics of large retail supply chains is well documented, research on retail small and medium enterprises’ (SMEs) capabilities to prevent and manage waste of perishable products is in its infancy (c.f. Ellegaard, 2008) and needs further exploration. In our study, we investigate the retail logistics practice of small food retailers, the factors that contribute to perishable products waste and the barriers and opportunities of SMEs in retail logistics to preserve product quality and participate in reverse logistics flows.

Design/methodology/approach:
As research on waste of perishable products for SMEs is scattered, we focus on identifying key variables that contribute to the creation of avoidable waste. Secondly we identify patterns of waste creation at the retail level and its possibilities for value added recovery. We use explorative case studies (Eisenhardt, 1989) and compare four SMEs and one large retailer that operate in a developed market. To get insights into specificities of SMEs that affect retail logistics practice, we select two types of food retailers: specialised (e.g. greengrocers and bakers) and general (e.g. convenience store that sells perishable products as a part of the assortment)

Findings:
Our preliminary findings indicate that there is a difference between large retailers and SME retailers in factors that contribute to the waste creation, as well as opportunities for value added recovery of products. While more factors appear to affect waste creation and management at large retailers, a small number of specific factors appears to affect SMEs. Similarly, large retailers utilise a range of practices to reduce risks of product perishability and short shelf life, manage demand, and manage reverse logistics practices. Retail SMEs on the other hand have limited options to address waste creation and value added recovery. However, our findings show that specialist SMEs could successfully minimize waste and even create possibilities for value added recovery of perishable products. Data indicates that business orientation of the SME, the buyersupplier relationship, and an extent of adoption of lean principles in retail coupled with SME resources, product specific regulations and support from local authorities for waste management or partnerships with other organizations determine extent of successful preservation of a product quality and value added recovery.

Value:
Our contribution to the SCM academic literature is threefold: first, we identify major factors that contribute to the generation waste of perishable products in retail environment; second, we identify possibilities for value added recovery for perishable products and third, we present opportunities and challenges for SME retailers to manage or participate in activities of value added recovery. Our findings contribute to theory by filling a gap in the literature that considers product quality preservation and value added recovery in the context of retail logistics and SMEs.

Research limitations/implications:
Our findings are limited to insights from five case studies of retail companies that operate within a developed market. To improve on generalisability, we intend to increase the number of cases and include data obtained from the suppliers and organizations involved in reverse logistics flows (e.g. local authorities, charities, etc.).

Practical implications:
With this paper, we contribute to the improvement of retail logistics and operations in SMEs which constitute over 99% of business activities in UK (Rhodes, 2015). Our findings will help retail managers and owners to better understand the possibilities for value added recovery, investigate a range of logistics and retail strategies suitable for the specificities of SME environment and, ultimately, improve their profitability and sustainability.

A high-value, low-cost bubble continuous positive airway pressure system for low-resource settings:technical assessment and initial case reports

Acute respiratory infections are the leading cause of global child mortality. In the developing world, nasal oxygen therapy is often the only treatment option for babies who are suffering from respiratory distress. Without the added pressure of bubble Continuous Positive Airway Pressure (bCPAP) which helps maintain alveoli open, babies struggle to breathe and can suffer serious complications, and frequently death. A stand-alone bCPAP device can cost $6,000, too expensive for most developing world hospitals. Here, we describe the design and technical evaluation of a new, rugged bCPAP system that can be made in small volume for a cost-of-goods of approximately $350. Moreover, because of its simple design--consumer-grade pumps, medical tubing, and regulators--it requires only the simple replacement of a <$1 diaphragm approximately every 2 years for maintenance. The low-cost bCPAP device delivers pressure and flow equivalent to those of a reference bCPAP system used in the developed world. We describe the initial clinical cases of a child with bronchiolitis and a neonate with respiratory distress who were treated successfully with the new bCPAP device.