Mission,money, and merit: Strategic decision making by nonprofit managers

Public and nonprofit organizations need to make strategic choices about where to invest their resources. They also need to expose hidden managerial assumptions and lack of adequate knowledge that prevent the attainment of consensus in strategic decision making. The approach we developed and tested in the field used a dynamic, three‐dimensional model that tracks individual programs in an organization's portfolio on their contribution to mission, money, and merit. The first dimension measures whether the organization is doing the right things; the second, whether it is doing things right financially; and the third, whether it doing things right in terms of quality. Senior managers provide their own evaluations of the organization's programs. Both the consensus view and the variation in individual assessments contribute to an improved managerial understanding of the organization's current situation and to richer discussions in strategic decision making. In field tests, this visual model proved to be a useful and powerful tool for illuminating underlying assumptions and variations in knowledge among managers facing the complex, multidimensional tradeoffs needed in strategic decision making.     

Stochastic Modelling of Times to Temperature for Furnaces Supplying Titanium Blooms to a Rolling Mill at TIMET

In conjunction with TIMET at Waunarlwydd (Swansea, UK) a model has been developed that will optimise the scheduling of various blooms to their eight furnaces so as to minimise the time taken to roll these blooms into the finished mill products. This production scheduling model requires reliable data on times taken for the various furnaces that heat the slabs and blooms to reach the temperatures required for rolling. These times to temperature are stochastic in nature and this paper identifies the distributional form for these times using the generalised F distribution as a modelling framework. The times to temperature were found to be similarly distributed over all furnaces. The identified distributional forms were incorporated into the scheduling model to optimise a particular campaign that was run at TIMET Swansea. Amongst other conclusion it was found that, compared to the actual campaign, the model produced a schedule that reduced the makespan by some 35%.     

Imagined Memories

Most discussions of popular music on the Internet focus on the utopian potential of new file-sharing technologies, yet applications that reproduce existing inequalities also deserve attention. Webcasting is the streaming (transmission) of digital video to multiple recipients in cyberspace. Paul McCartney's Webcast from the Cavern was a landmark case. It was a digital package staged for reproduction, and yet it felt live; in this article I explore why and offer a two-part explanation. The first part is that "live'-ness is based on an increasingly false opposition to recording, but, because that opposition still remains, Little Big Gig could seem live by adopting some trappings associated with it. The second part is that Internet use is mediated by daily life and computer users brought their own desire to the Webcast, in particular desires to see a Beatle play in the Cavern. Webcasting is an unanticipated use of the Internet that is being used to support corporate interests. With its widespread publicity, Little Big Gig helped naturalize that process.     

Bayesian inference for a discretely observed stochastic kinetic model

The ability to infer parameters of gene regulatory networks is emerging as a key problem in systems biology. The biochemical data are intrinsically stochastic and tend to be observed by means of discrete-time sampling systems, which are often limited in their completeness. In this paper we explore how to make Bayesian inference for the kinetic rate constants of regulatory networks, using the stochastic kinetic Lotka-Volterra system as a model. This simple model describes behaviour typical of many biochemical networks which exhibit auto-regulatory behaviour. Various MCMC algorithms are described and their performance evaluated in several data-poor scenarios. An algorithm based on an approximating process is shown to be particularly efficient.     

Asthma Patients' Activities and Air Pollution: A Semiparametric Censored Regression Analysis

Asthma patients' health status may be especially sensitive to some types of air pollution, but the evidence on this is mixed. We explore the effects of ground-level ozone on asthma patient's activities, breaking apart the usual aggregated category of leisure into indoor and outdoor activities, and differentiating those by whether the activities were active or inactive. Applying the semiparametric censored estimation method we demonstrate that even though the period over which activities were observed was relatively low in ozone levels, there is a significant impact of ozone on a few activities. The (non-ozone) economic and demographic variables in the model play significant roles in explaining the allocation of time among seven activities, suggesting the suitability of the approach for other household decision-making contexts.     

Bayesian locally optimal design of knockout tournaments

The elimination or knockout format is one of the most common designs for pairing competitors in tournaments and leagues. In each round of a knockout tournament, the losers are eliminated while the winners advance to the next round. Typically, the goal of such a design is to identify the overall best player. Using a common probability model for expressing relative player strengths, we develop an adaptive approach to pairing players each round in which the probability that the best player advances to the next round is maximized. We evaluate our method using simulated game outcomes under several data-generating mechanisms, and compare it to random pairings, to the standard knockout format, and to two variants of the standard format.     

Optimal scaling of Metropolis algorithms: Heading toward general target distributions

The authors provide an overview of optimal scaling results for the Metropolis algorithm with Gaussian proposal distribution. They address in more depth the case of high‐dimensional target distributions formed of independent, but not identically distributed components. They attempt to give an intuitive explanation as to why the well‐known optimal acceptance rate of 0.234 is not always suitable. They show how to find the asymptotically optimal acceptance rate when needed, and they explain why it is sometimes necessary to turn to inhomogeneous proposal distributions. Their results are illustrated with a simple example.