The application of a reinforcement learning agent to a multi-product manufacturing facility

An intelligent agent-based scheduling system, consisting of a reinforcement learning agent and a simulation model has been developed and tested on a classic scheduling problem. The production facility studied is a multiproduct serial line subject to stochastic failure. The agent goal is to minimise total production costs, through selection of job sequence and batch size. To explore state space the agent used reinforcement learning. By applying an independent inventory control policy for each product, the agent successfully identified optimal operating policies for a real production facility.

Content-based video indexing for sports applications using integrated multi-modal approach

To sustain an ongoing rapid growth of video information, there is an emerging demand for a sophisticated content-based video indexing system. However, current video indexing solutions are still immature and lack of any standard. This doctoral consists of a research work based on an integrated multi-modal approach for sports video indexing and retrieval. By combining specific features extractable from multiple audio-visual modalities, generic structure and specific events can be detected and classified. During browsing and retrieval, users will benefit from the integration of high-level semantic and some descriptive mid-level features such as whistle and close-up view of player(s).

Optimising discrete event simulation models using a reinforcement learning agent

A reinforcement learning agent has been developed to determine optimal operating policies in a multi-part serial line. The agent interacts with a discrete event simulation model of a stochastic production facility. This study identifies issues important to the simulation developer who wishes to optimise a complex simulation or develop a robust operating policy. Critical parameters pertinent to 'tuning' an agent quickly and enabling it to rapidly learn the system were investigated.

Realistic grasping of rigid virtual objects through event-based haptics

This research presents a novel haptic grasping interface and demonstrates its ability within multi-point event-based feedback. Through experimental methodology, the dynamics involved in grasp contact interactions are modelled based on first principles. The proposed approach demonstrates a method of realistically representing grasp contact with rigid virtual objects through multi-point interaction.

Design tools for event-driven software systems

Proposes a design method for the development of software in Visual BASIC whose programming environment is multi-paradigm. Extensions to structure charts and pseudocode to incorporate the three paradigms have been developed based on the premise that procedural models, event-driven forms and objects are mutually exclusive in operation.

Event effectiveness : analysing the perceptions of key stakeholders of the Lexmark Indy 300/champ car event, Gold Coast.

The present study assessed the perceptions of the Lexmark Indy 300/Champ Car Event’s key stakeholders namely, spectators, volunteers and paid workers, as part of an evaluation tool for its managers. We developed a model, which included visible, interaction and management elements that we proposed influenced the perceptions and experiences of stakeholders relating to the event effectiveness, strengths, weaknesses and improvements. Specifically, we used a modified version of Preble’s (2005) six-step multi-stakeholder assessment approach to examining the perceived overall effectiveness of the event. Fifty stakeholder interviews were conducted on-site and their opinions were content analysed. The key findings of the study were that the majority of the stakeholders perceived the event to be effective. Spectators perceived the strengths to be interaction components, and weaknesses and improvements to be visible elements such as track layout. Volunteers perceived strengths to be management related, but also noted that these were weaknesses and needed improvement. Lastly, Paid workers perceived interaction elements to be both strengths and weaknesses, and suggested that management elements could be improved. Important implications for similar motoring events were suggested.

A multi-criteria decision approach to decommissioning of offshore oil and gas infrastructure

Thousands of the world's offshore oil and gas structures are approaching obsolescence and will require decommissioning within the next decade. Many nations have blanket regulations requiring obsolete structures to be removed, yet this option is unlikely to yield optimal environmental, societal and economic outcomes in all situations. We propose that nations adopt a flexible approach that allows decommissioning options to be selected from the full range of alternatives (including 'rigs-to-reefs' options) on a case-by-case basis. We outline a method of multi-criteria decision analysis (Multi-criteria Approval, MA) for evaluating and comparing alternative decommissioning options across key selection criteria, including environmental, financial, socioeconomic, and health and safety considerations. The MA approach structures the decision problem, forces explicit consideration of trade-offs and directly involves stakeholder groups in the decision process. We identify major decommissioning options and provide a generic list of selection criteria for inclusion in the MA decision process. To deal with knowledge gaps concerning environmental impacts of decommissioning, we suggest that expert opinion feed into the MA approach until sufficient data become available. We conducted a limited trial of the MA decision approach to demonstrate its application to a complex and controversial decommissioning scenario; Platform Grace in southern California. The approach indicated, for this example, that the option 'leave in place intact' would likely provide best environmental outcomes in the event of future decommissioning. In summary, the MA approach will allow the environmental, social, and economic impacts of decommissioning decisions to be assessed simultaneously in a transparent manner. © 2013 Elsevier Ltd.

A multi-level approach to planning and scheduling resources for aviation training

This paper describes a multi-level system dynamics (SD) / discrete event simulation (DES) approach for assessing planning and scheduling problems within an aviation training continuum. The aviation training continuum is a complex system, consisting of multiple aviation schools interacting through interschool student and instructor flows that are affected by external triggers such as resource availability and the weather.
SD was used to model the overall training continuum at a macro level to ascertain relationships between system entities. SD also assisted in developing a shared understanding of the training continuum, which involves constructing the definitions of the training requirements, resources and policy objectives. An end-to-end model of the continuum is easy to relate to, while dynamic visualisation of system behaviour provides a method for exploration of the model.
DES was used for micro level exploration of an individual school within the training continuum to capture the physical aspects of the system including resource capacity requirements, bottlenecks and student waiting times. It was also used to model stochastic events such as weather and student availability. DES has the advantage of being able to represent system variability and accurately reflect the limitations imposed on a system by resource constraints.
Through sharing results between the models, we demonstrate a multi-level approach to the analysis of the overall continuum. The SD model provides the school’s targeted demand to the DES model. The detailed DES model is able to assess schedules in the presence of resource constraints and variability and provide the expected capacity of a school to the high level SD model, subjected to constraints such as instructor availability or budgeted number of training systems. The SD model allows stakeholders to assess how policy and planning affect the continuum, both in the short and the long term.
The development of this approach permits moving the analysis of the continuum between SD and DES models as appropriate for given system entities, scales and tasks. The resultant model outcomes are propagated between the continuum and the detailed DES model, iteratively generating an assessment of the entire set of plans and schedule across the continuum. Combining data and information between SD and DES models and techniques assures relevance to the stakeholder needs and effective problem scoping and scaling that can also evolve with dynamic architecture and policy requirements.
An example case study shows the combined use of the two models and how they are used to evaluate a typical scenario where increased demand is placed on the training continuum. The multi-level approach provides a high level indication of training requirements to the model of the new training school, where the detailed model indicates the resources required to achieve those particular student levels.

Vertical distribution of benthic invertebrate larvae during an upwelling event along a transect off the tropical Brazilian continental margin

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Jet and underlying event properties as a function of charged-particle multiplicity in proton-proton collisions at root s=7 TeV

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Vertical distribution of benthic invertebrate larvae during an upwelling event along a transect off the tropical Brazilian continental margin

Abundance and composition of marine benthic communities have been relatively well studied in the SE Brazilian coast, but little is known on patterns controlling the distribution of their planktonic larval stages. A survey of larval abundance in the continental margin, using a Multi-Plankton Sampler, was conducted in a cross-shelf transect off Cabo Frio (23 degrees S and 42 degrees W) during a costal upwelling event. Hydrographic conditions were monitored through discrete CDT casts. Chlorophyll-a in the top 100 m of the water column was determined and changes in surface chlorophyll-a was estimated using SeaWiFS images. Based on the larval abundances and the meso-scale hydrodynamics scenario, our results suggest two different processes affecting larval distributions. High larval densities were found nearshore due to the upwelling event associated with high chlorophyll a and strong along shore current. On the continental slope, high larval abundance was associated with a clockwise rotating meander, which may have entrapped larvae from a region located further north (Cabo de Sao Tome, 22 degrees S and 41 degrees W). In mid-shelf areas, our data suggests that vertical migration may likely occur as a response to avoid offshore transport by upwelling plumes and/or cyclonic meanders. The hydrodynamic scenario observed in the study area has two distinct yet extremely important consequences: larval retention on food-rich upwelling areas and the broadening of the tropical domain to southernmost subtropical areas. (C) 2009 Elsevier B.V. All rights reserved.

Geochronological constraints on the age of a Permo-Triassic impact event: U-Pb and Ar-40/Ar-39 results for the 40 km Araguainha structure of central Brazil

Impact cratering has been a fundamental geological process in Earth history with major ramifications for the biosphere. The complexity of shocked and melted rocks within impact structures presents difficulties for accurate and precise radiogenic isotope age determination, hampering the assessment of the effects of an individual event in the geological record. We demonstrate the utility of a multi-chronometer approach in our study of samples from the 40 km diameter Araguainha impact structure of central Brazil. Samples of uplifted basement granite display abundant evidence of shock deformation, but U/Pb ages of shocked zircons and the Ar-40/Ar-39 ages of feldspar from the granite largely preserve the igneous crystallization and cooling history. Mixed results are obtained from in situ Ar-40/Ar-39 spot analyses of shocked igneous biotites in the granite, with deformation along kink-bands resulting in highly localized, partial resetting in these grains. Likewise, spot analyses of perlitic glass from pseudotachylitic breccia samples reflect a combination of argon inheritance from wall rock material, the age of the glass itself, and post-impact devitrification. The timing of crater formation is better assessed using samples of impact-generated melt rock where isotopic resetting is associated with textural evidence of melting and in situ crystallization. Granular aggregates of neocrystallized zircon form a cluster of ten U-Pb ages that yield a "Concordia" age of 247.8 +/- 3.8 Ma. The possibility of Pb loss from this population suggests that this is a minimum age for the impact event. The best evidence for the age of the impact comes from the U-Th-Pb dating of neocrystallized monazite and Ar-40/Ar-39 step heating of three separate populations of post-impact, inclusion-rich quartz grains that are derived from the infill of miarolitic cavities. The Pb-206/U-238 age of 254.5 +/- 3.2 Ma (2 sigma error) and Pb-208/Th-232 age of 255.2 +/- 4.8 Ma (2 sigma error) of monazite, together with the inverse, 18 point isochron age of 254 +/- 10 Ma (MSWD = 0.52) for the inclusion-rich quartz grains yield a weighted mean age of 254.7 +/- 2.5 Ma (0.99%, 2 sigma error) for the impact event. The age of the Araguainha crater overlaps with the timing of the Permo-Triassic boundary, within error, but the calculated energy released by the Araguainha impact is insufficient to be a direct cause of the global mass extinction. However, the regional effects of the Araguainha impact event in the Parana-Karoo Basin may have been substantial. (C) 2012 Elsevier Ltd. All rights reserved.

Multi-strange baryon production in pp collisions at root s=7 TeV with ALICE

A measurement of the multi-strange Xi(-) and Omega(-) baryons and their antiparticles by the ALICE experiment at the CERN Large Hadron Collider (LHC) is presented for inelastic proton-proton collisions at a centre-of-mass energy of 7 TeV. The transverse momentum (p(T)) distributions were studied at mid-rapidity (vertical bar y vertical bar < 0.5) in the range of 0.6 < p(T) < 8.5 GeV/c Xi(-) for and Xi(+) baryons, and in the range of 0.8 < P-T < 5 GeV/c for Omega(-) and<(Omega)over bar>(+). Baryons and antibaryons were measured as separate particles and we find that the baryon to antibaryon ratio of both particle species is consistent with unity over the entire range of the measurement. The statistical precision of the current data has allowed us to measure a difference between the mean p(T) of Xi(-) ((Xi) over bar)(+) and Omega(-) ((Omega) over bar (+)). Particle yields, mean pi, and the spectra in the intermediate pi range are not well described by the PYTHIA Perugia 2011 tune Monte Carlo event generator, which has been tuned to reproduce the early LHC data. The discrepancy is largest for Omega(-)((Omega) over bar (+)). This PYTHIA tune approaches the pi spectra of Xi(-) and Xi(+) baryons below p(T) <0.85 GeV/c and describes the Xi(-) and Xi(+) spectra above p(T) > 6.0 GeV/c. We also illustrate the difference between the experimental data and model by comparing the corresponding ratios of (Omega(-) +(Omega) over bar (+))/(Xi(-) + Xi(+)) as a function of transverse mass. (C) 2012 CERN. Published by Elsevier B.V. All rights reserved.