Integrating skills profiling and purchasing portfolio management:an opportunity for building purchasing capability

Kralijc’s (1983) purchasing portfolio approach holds that different types of purchases need different sourcing strategies, underpinned by distinct sets of resources and practices. The approach is widely deployed in business and extensively researched, and yet little research has been conducted on how knowledge and skills vary across a portfolio of purchases. This study extends the body of knowledge on purchasing portfolio management, and its application in the strategic development of purchasing in an organization, and on human resource management in the purchasing function. A novel approach to profiling purchasing skills is proposed, which is well suited to dynamic environments which require flexibility. In a survey, experienced purchasing personnel described a specific purchase and profiled the skills required for effective performance in purchasing that item. Purchases were categorized according to their importance to the organization (internally-oriented evaluation of cost and production factors) and to the supply market (externally-oriented evaluation of commercial risk and uncertainty). Through cluster analysis three key types of purchase situations were identified. The skills required for effective purchasing vary significantly across the three clusters (for 22 skills, p<0.01). Prior research shows that global organizations use the purchasing portfolio approach to develop sourcing strategies, but also aggregate analyses to inform the design of purchasing arrangements (local vs global) and to develop their improvement plans. Such organizations would also benefit from profiling skills by purchase type. We demonstrate how the survey can be adapted to provide a management tool for global firms seeking to improve procurement capability, flexibility and performance.

Simulation and Evaluation of High-Voltage Power Systems for Civil Aircraft

A status report of the modelling and simulation work that is being undertaken as part of the TIMES (Totally Integrated More Electric Systems) project is presented. Dynamic power quality simulations have been used to asses the performance of the electrical system of a EMA based actuation system for an Airbus A330 size aircraft, for both low voltage 115 V, and high voltage 230 V three-phase AC systems. The high voltage system is shown to have benefits in terms of power quality and reduced size and weight of equipment.

Modular construction of dynamic nucleodendrimers

Isoguanosine-containing dendritic small molecules self-assemble into decameric nucleodendrimers as observed by 1D NMR spectroscopy, 2D DOSY, and mass spectrometry. In particular, apolar building blocks readily form pentameric structures in acetonitrile while the presence of alkali metals promotes the formation of stable decameric assemblies with a preference for cesium ions. Remarkably, co-incubation of guanosine and isoguanosine-containing nucleodendrons results in the formation of decameric structures in absence of added salts. Further analysis of the mixture indicated that guanosine derivatives facilitate the formation, but are not involved in decameric structures; a process reminiscent of molecular crowding. This molecular system provides a powerful canvas for the rapid and modular assembly of polyfunctional dendritic macromolecules. © 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Molecular dynamics simulation of brittle fracture in silicon

The fracture process involves converting potential energy from a strained body into surface energy, thermal energy, and the energy needed to create lattice defects. In dynamic fracture, energy is also initially converted into kinetic energy. This paper uses molecular dynamics (MD) to simulate brittle frcture in silicon and determine how energy is converted from potential energy (strain energy) into other forms.

Improving scheduling techniques in heterogeneous systems with dynamic, on-line optimisations

Computational performance increasingly depends on parallelism, and many systems rely on heterogeneous resources such as GPUs and FPGAs to accelerate computationally intensive applications. However, implementations for such heterogeneous systems are often hand-crafted and optimised to one computation scenario, and it can be challenging to maintain high performance when application parameters change. In this paper, we demonstrate that machine learning can help to dynamically choose parameters for task scheduling and load-balancing based on changing characteristics of the incoming workload. We use a financial option pricing application as a case study. We propose a simulation of processing financial tasks on a heterogeneous system with GPUs and FPGAs, and show how dynamic, on-line optimisations could improve such a system. We compare on-line and batch processing algorithms, and we also consider cases with no dynamic optimisations.