Web based expert elicitation of uncertainties in environmental model inputs

When constructing and using environmental models, it is typical that many of the inputs to the models will not be known perfectly. In some cases, it will be possible to make observations, or occasionally physics-based uncertainty propagation, to ascertain the uncertainty on these inputs. However, such observations are often either not available or even possible, and another approach to characterising the uncertainty on the inputs must be sought. Even when observations are available, if the analysis is being carried out within a Bayesian framework then prior distributions will have to be specified. One option for gathering or at least estimating this information is to employ expert elicitation. Expert elicitation is well studied within statistics and psychology and involves the assessment of the beliefs of a group of experts about an uncertain quantity, (for example an input / parameter within a model), typically in terms of obtaining a probability distribution. One of the challenges in expert elicitation is to minimise the biases that might enter into the judgements made by the individual experts, and then to come to a consensus decision within the group of experts. Effort is made in the elicitation exercise to prevent biases clouding the judgements through well-devised questioning schemes. It is also important that, when reaching a consensus, the experts are exposed to the knowledge of the others in the group. Within the FP7 UncertWeb project (http://www.uncertweb.org/), there is a requirement to build a Webbased tool for expert elicitation. In this paper, we discuss some of the issues of building a Web-based elicitation system - both the technological aspects and the statistical and scientific issues. In particular, we demonstrate two tools: a Web-based system for the elicitation of continuous random variables and a system designed to elicit uncertainty about categorical random variables in the setting of landcover classification uncertainty. The first of these examples is a generic tool developed to elicit uncertainty about univariate continuous random variables. It is designed to be used within an application context and extends the existing SHELF method, adding a web interface and access to metadata. The tool is developed so that it can be readily integrated with environmental models exposed as web services. The second example was developed for the TREES-3 initiative which monitors tropical landcover change through ground-truthing at confluence points. It allows experts to validate the accuracy of automated landcover classifications using site-specific imagery and local knowledge. Experts may provide uncertainty information at various levels: from a general rating of their confidence in a site validation to a numerical ranking of the possible landcover types within a segment. A key challenge in the web based setting is the design of the user interface and the method of interacting between the problem owner and the problem experts. We show the workflow of the elicitation tool, and show how we can represent the final elicited distributions and confusion matrices using UncertML, ready for integration into uncertainty enabled workflows.We also show how the metadata associated with the elicitation exercise is captured and can be referenced from the elicited result, providing crucial lineage information and thus traceability in the decision making process.

Regression with input-dependent noise: A Gaussian process treatment

Gaussian processes provide natural non-parametric prior distributions over regression functions. In this paper we consider regression problems where there is noise on the output, and the variance of the noise depends on the inputs. If we assume that the noise is a smooth function of the inputs, then it is natural to model the noise variance using a second Gaussian process, in addition to the Gaussian process governing the noise-free output value. We show that prior uncertainty about the parameters controlling both processes can be handled and that the posterior distribution of the noise rate can be sampled from using Markov chain Monte Carlo methods. Our results on a synthetic data set give a posterior noise variance that well-approximates the true variance.

Inventory parameter management and focused continuous improvement for repetitive batch manufacturers

What this thesis proposes is a methodology to assist repetitive batch manufacturers in the adoption of certain aspects of the Lean Production principles. The methodology concentrates on the reduction of inventory through the setting of appropriate batch sizes, taking account of the effect of sequence dependent set-ups and the identification and elimination of bottlenecks. It uses a simple Pareto and modified EBQ based analysis technique to allocate items to period order day classes based on a combination of each item's annual usage value and set-up cost. The period order day classes the items are allocated to are determined by the constraints limits in the three measured dimensions, capacity, administration and finance. The methodology overcomes the limitations associated with MRP in the area of sequence dependent set-ups, and provides a simple way of setting planning parameters taking this effect into account by concentrating on the reduction of inventory through the systematic identification and elimination of bottlenecks through set-up reduction processes, so allowing batch sizes to reduce. It aims to help traditional repetitive batch manufacturers in a route to continual improvement by: Highlighting those areas where change would bring the greatest benefits. Modelling the effect of proposed changes. Quantifying the benefits that could be gained through implementing the proposed changes. Simplifying the effort required to perform the modelling process. It concentrates on increasing flexibility through managed inventory reduction through rationally decreasing batch sizes, taking account of sequence dependent set-ups and the identification and elimination of bottlenecks. This was achieved through the development of a software modelling tool, and validated through a case study approach.

Practicing social e-atmospherics:French and Turkish web designers' perspectives

E-atmospherics have motivated an emerging body of research which reports that both virtual layouts and atmospherics encourage consumers to modify their shopping habits. While the literature has analyzed mainly the functional aspect of e-atmospherics, little has been done in terms of linking its characteristics’ to social (co-) creation. This paper focuses on the anatomy of social dimension in relation to e-atmospherics, which includes factors such as the aesthetic design of space, the influence of visual cues, interpretation of shopping as a social activity and meaning of appropriate interactivity. We argue that web designers are social agents who interact within intangible social reference sets, restricted by social standards, value, beliefs, status and duties embedded within their local geographies. We aim to review the current understanding of the importance and voluntary integration of social cues displayed by web designers from a mature market and an emerging market, and provides an analysis based recommendation towards the development of an integrated e-social atmospheric framework. Results report the findings from telephone interviews with an exploratory set of 10 web designers in each country. This allows us to re-interpret the web designers’ reality regarding social E-atmospherics. We contend that by comprehending (before any consumer input) social capital, daily micro practices, habits and routine, deeper understanding of social e-atmospherics preparatory, initial stages and expected functions will be acquired.

The human factors of computer input using a mouse device

This thesis investigates how people select items from a computer display using the mouse input device. The term computer mouse refers to a class of input devices which share certain features, but these may have different characteristics which influence the ways in which people use the device. Although task completion time is one of the most commonly used performance measures for input device evaluation, there is no consensus as to its definition. Furthermore most mouse studies fail to provide adequate assurances regarding its correct measurement.Therefore precise and accurate timing software were developed which permitted the recording of movement data which by means of automated analysis yielded the device movements made. Input system gain, an important task parameter, has been poorly defined and misconceptualized in most previous studies. The issue of gain has been clarified and investigated within this thesis. Movement characteristics varied between users and within users, even for the same task conditions. The variables of target size, movement amplitude, and experience exerted significant effects on performance. Subjects consistently undershot the target area. This may be a consequence of the particular task demands. Although task completion times indicated that mouse performance had stabilized after 132 trials the movement traces, even of very experienced users, indicated that there was still considerable room for improvement in performance, as indicated by the proportion of poorly made movements. The mouse input device was suitable for older novice device users, but they took longer to complete the experimental trials. Given the diversity and inconsistency of device movements, even for the same task conditions, caution is urged when interpreting averaged grouped data. Performance was found to be sensitive to; task conditions, device implementations, and experience in ways which are problematic for the theoretical descriptions of device movement, and limit the generalizability of such findings within this thesis.

Generation of triangular pulses in normally dispersive fibre

In this scheme, nonlinearity and dispersion in the NDF lead to various reshaping processes of an initial, conventional pulse according to the chirping value and power level at the input of the fibre. In particular, we have observed that triangular-shaped pulses can be generated for sufficiently high energies and a positive initial chirp parameter. In our experiments, 2.8 ps-FWHM, transform-limited pulses generated from a mode-locked fibre laser source at a repetition rate of 1.25 GHz were pre-chirped by propagating the pulses through different lengths of standard mono-mode fibre. The chirped pulses were then amplified to different power levels before being launched into a 2.3 km section of True Wave fibre (TWF). The corresponding numerically calculated pulse temporal intensity profile and numerical and experimental second-harmonic generation frequency-resolved optical gating (SHG FROG) spectrograms were also derived. In conclusion, we have presented numerical modelling results which show the system design parameters required for the generation of triangular-shaped pulses in a nonlinear NDF, and experimentally demonstrated triangular pulse shaping in conventional NDF.

Simulating the value of collaboration in multi-actor conservation planning

The loss of habitat and biodiversity worldwide has led to considerable resources being spent on conservation interventions. Prioritising these actions is challenging due to the complexity of the problem and because there can be multiple actors undertaking conservation actions, often with divergent or partially overlapping objectives. We explore this issue with a simulation study involving two agents sequentially purchasing land for the conservation of multiple species using three scenarios comprising either divergent or partially overlapping objectives between the agents. The first scenario investigates the situation where both agents are targeting different sets of threatened species. The second and third scenarios represent a case where a government agency attempts to implement a complementary conservation network representing 200 species, while a non-government organisation is focused on achieving additional protection for the ten rarest species. Simulated input data was generated using distributions taken from real data to model the cost of parcels, and the rarity and co-occurrence of species. We investigated three types of collaborative interactions between agents: acting in isolation, sharing information and pooling resources with the third option resulting in the agents combining their resources and effectively acting as a single entity. In each scenario we determine the cost savings when an agent moves from acting in isolation to either sharing information or pooling resources with the other agent. The model demonstrates how the value of collaboration can vary significantly in different situations. In most cases, collaborating would have associated costs and these costs need to be weighed against the potential benefits from collaboration. Our model demonstrates a method for determining the range of costs that would result in collaboration providing an efficient use of scarce conservation resources.

The innovation value chain and export performance:a study of Taiwanese manufacturing industries

This thesis involves the secondary data of 1806 innovative manufacturing firms derived from the database of 2nd Taiwanese Innovation Survey. Three topics are researched. The first topic investigates the innovation value chain (IVC) in Taiwanese manufacturing firms. Previous IVC studies are all done in developed countries such as UK, Ireland, Northern Ireland and Switzerland, and it leaves the gap of those non-developed countries. The result shows the overall knowledge sourcing pattern of Taiwanese manufacturing firms presenting a complementary relationship which is consistent to the previous IVC studies. The main innovation input is still derived from internal R&D which suggests more utilisation of external knowledge may boost innovation outcome. Product innovation does enhance firm growth while process innovation reduces a firm’s productivity. The second topic uses the lens of IVC to investigate the difference of the innovation process from knowledge linkages to value added between high-tech and low- tech sectors. The findings indicate (1) there are significant differences in the IVC between high- and low-tech sectors, however these are defined; (2) how you define ‘sector’ matters i.e. the nature of the high-tech and low-tech differences varies depending on whether the technology definition is carried out at the industry or firm level; and (3) the high uncertainty of innovation cause the difficulty to predict firm performance especially for those firms with high intensity of innovation. The third topic investigates the innovation-exporting relationship and explores the determinants of export performance. Product innovation enhances export performance once a firm enters international markets while process innovation affects negatively on a firm’s likelihood of being an exporter. Furthermore, IP protection is found to affect directly export performance positively.

A framework for assessing business-incubation input, process and output performance:proceedings of the RENT XXVII Conference, 20-22 November 2013, Vilnius, Lithuania

With business incubators deemed as a potent infrastructural element for entrepreneurship development, business incubation management practice and performance have received widespread attention. However, despite this surge of interest, scholars have questioned the extent to which business incubation delivers added value. Thus, there is a growing awareness among researchers, practitioners and policy makers of the need for more rigorous evaluation of the business incubation output performance. Aligned to this is an increasing demand for benchmarking business incubation input/process performance and highlighting best practice. This paper offers a business incubation assessment framework, which considers input/process and output performance domains with relevant indicators. This tool adds value on different levels. It has been developed in collaboration with practitioners and industry experts and therefore it would be relevant and useful to business incubation managers. Once a large enough database of completed questionnaires has been populated on an online platform managed by a coordinating mechanism, such as a business incubation membership association, business incubator managers can reflect on their practices by using this assessment framework to learn their relative position vis-à-vis their peers against each domain. This will enable them to align with best practice in this field. Beyond implications for business incubation management practice, this performance assessment framework would also be useful to researchers and policy makers concerned with business incubation management practice and impact. Future large-scale research could test for construct validity and reliability. Also, discriminant analysis could help link input and process indicators with output measures.

Optimal design for correlated processes with input-dependent noise

Optimal design for parameter estimation in Gaussian process regression models with input-dependent noise is examined. The motivation stems from the area of computer experiments, where computationally demanding simulators are approximated using Gaussian process emulators to act as statistical surrogates. In the case of stochastic simulators, which produce a random output for a given set of model inputs, repeated evaluations are useful, supporting the use of replicate observations in the experimental design. The findings are also applicable to the wider context of experimental design for Gaussian process regression and kriging. Designs are proposed with the aim of minimising the variance of the Gaussian process parameter estimates. A heteroscedastic Gaussian process model is presented which allows for an experimental design technique based on an extension of Fisher information to heteroscedastic models. It is empirically shown that the error of the approximation of the parameter variance by the inverse of the Fisher information is reduced as the number of replicated points is increased. Through a series of simulation experiments on both synthetic data and a systems biology stochastic simulator, optimal designs with replicate observations are shown to outperform space-filling designs both with and without replicate observations. Guidance is provided on best practice for optimal experimental design for stochastic response models. © 2013 Elsevier Inc. All rights reserved.

Essential BIM input data study for housing refurbishment:homeowners’ preferences in the UK

Construction customers are persistently seeking to achieve sustainability and maximize value as sustainability has become a major consideration in the construction industry. In particular, it is essential to refurbish a whole house to achieve the sustainability agenda of 80% CO2 reduction by 2050 as the housing sector accounts for 28% of the total UK CO2 emission. However, whole house refurbishment seems to be challenging due to the highly fragmented nature of construction practice, which makes the integration of diverse information throughout the project lifecycle difficult. Consequently, Building Information Modeling (BIM) is becoming increasingly difficult to ignore in order to manage construction projects in a collaborative manner, although the current uptake of the housing sector is low at 25%. This research aims to investigate homeowners’ decision making factors for housing refurbishment projects and to provide a valuable dataset as an essential input to BIM for such projects. One-hundred and twelve homeowners and 39 construction professionals involved in UK housing refurbishment were surveyed. It was revealed that homeowners value initial cost more while construction professionals value thermal performance. The results supported that homeowners and professionals both considered the first priority to be roof refurbishment. This research revealed that BIM requires a proper BIM dataset and objects for housing refurbishment.

Exploring demographic information in social media for product recommendation

In many e-commerce Web sites, product recommendation is essential to improve user experience and boost sales. Most existing product recommender systems rely on historical transaction records or Web-site-browsing history of consumers in order to accurately predict online users’ preferences for product recommendation. As such, they are constrained by limited information available on specific e-commerce Web sites. With the prolific use of social media platforms, it now becomes possible to extract product demographics from online product reviews and social networks built from microblogs. Moreover, users’ public profiles available on social media often reveal their demographic attributes such as age, gender, and education. In this paper, we propose to leverage the demographic information of both products and users extracted from social media for product recommendation. In specific, we frame recommendation as a learning to rank problem which takes as input the features derived from both product and user demographics. An ensemble method based on the gradient-boosting regression trees is extended to make it suitable for our recommendation task. We have conducted extensive experiments to obtain both quantitative and qualitative evaluation results. Moreover, we have also conducted a user study to gauge the performance of our proposed recommender system in a real-world deployment. All the results show that our system is more effective in generating recommendation results better matching users’ preferences than the competitive baselines.

Novel input-output prediction approach for biomass pyrolysis

Biomass pyrolysis to bio-oil is one of the promising sustainable fuels. In this work, relation between biomass feedstock element characteristic and pyrolysis process outputs was explored. The element characteristics considered in this study include moisture, ash, fix carbon, volatile matter, carbon, hydrogen, nitrogen, oxygen, and sulphur. A semi-batch fixed bed reactor was used for biomass pyrolysis with heating rate of 30 °C/min from room temperature to 600 °C and the reactor was held at 600 °C for 1 h before cooling down. Constant nitrogen flow rate of 5 L/min was provided for anaerobic condition. Rice husk, Sago biomass and Napier grass were used in the study to form different element characteristic of feedstock by altering mixing ratio. Comparison between each element characteristic to total produced bio-oil yield, aqueous phase bio-oil yield, organic phase bio-oil yield, higher heating value of organic phase bio-oil, and organic bio-oil compounds was conducted. The results demonstrate that process performance is associated with feedstock properties, which can be used as a platform to access the process feedstock element acceptance range to estimate the process outputs. Ultimately, this work evaluated the element acceptance range for proposed biomass pyrolysis technology to integrate alternative biomass species feedstock based on element characteristic to enhance the flexibility of feedstock selection.