No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

This thesis is a report of an investigation into the possible advantages to be gained by running computer simulations interactively. Models relevant to administrative, teaching and research work were constructed for the study, and the merits of the interactive use of deterministic models, stochastic equations and Monte-Carlo simulations were examined. In order to be able to draw worthwhile conclusions from the investigation it was necessary to study sustantial systems which had proved to require computer simulation for their elucidation. The first model described is a deterministic portrayal of the use of university lecture rooms which was written as an aid towards evaluating the need for additional rooms in an expanding university. The part played by interaction is to make easier the incorpration of human experience into the planning mechanism. Population genetics provided the next system, and it is shown how a model using a set of stochastic equations in conjunction with fast graphical output may be of value in teaching. A Monte-Carlo approach is demonstrated to be unsuitable for interactive use. The final simulation also employs stochastic equations, designed to represent a system from the field of cell cycle kinetics whose action is not fully understood. Using the model interactively allows the researcher to form an appreciation of the consequences of altering its parameters and to fit experimental data with more perception than is possible using purely algorithmic methods. Introductions are given to the two biological systems dealt with, so that the results of using the models can be discussed in relation to the genetics and cytokinetics Lnvo lved as well as purely in the context of the interactive use of simulmtion.

It is now commonplace for complex physical systems such as the climate system to be studied indirectly via computer simulations. Often, the equations that govern the underlying physical system are known but detailed or highresolution computer models of these equations (“governing models”) are not practical because of limited computational resources; so the models are simplified or “parameterised”. However, if the output of a simplified model is to lead to conclusions about a physical system, we must prove that these outputs reflect reality and are not merely artifacts of the simplifications. At present, simplifications are usually based on informal, ad-hoc methods making it difficult or impossible to provide such a proof rigorously. Here we introduce a set of formal methods for generating computer models. We present a newly developed computer program, “iGen”, which syntactically analyses the computer code of a high-resolution, governing model and, without executing it, automatically produces a much faster, simplified model with provable bounds on error compared to the governing model. These bounds allow scientists to rigorously distinguish real world phenomena from artifact in subsequent numerical experiments using the simplified model. Using simple physical systems as examples, we illustrate that iGen produces simplified models that execute typically orders of magnitude faster than their governing models. Finally, iGen is used to generate a model of entrainment in marine stratocumulus. The resulting simplified model is appropriate for use as part of a parameterisation of marine stratocumulus in a Global Climate Model.

This thesis addresses certain problems encountered by teams of engineers when modelling complex structures and processes subject to cost and other resource constraints. The cost of a structure or process may be ‘read off’ its specifying model, but the language in which the model is expressed (e.g. CAD) and the language in which resources may be modelled (e.g. spreadsheets) are not naturally compatible. This thesis demonstrates that a number of intermediate steps may be introduced which enable both meaningful translation from one conceptual view to another as well as meaningful collaboration between team members. The work adopts a diagrammatic modelling approach as a natural one in an engineering context when seeking to establish a shared understanding of problems. Thus, the research question to be answered in this thesis is: ‘To what extent is it possible to improve user-driven software development through interaction with diagrams and without requiring users to learn particular computer languages?’ The goal of the research is to improve collaborative software development through interaction with diagrams, thereby minimising the need for end-users to code directly. To achieve this aim a combination of the paradigms of End-User Programming, Process and Product Modelling and Decision Support, and Semantic Web are exploited and a methodology of User Driven Modelling and Programming (UDM/P) is developed, implemented, and tested as a means of demonstrating the efficacy of diagrammatic modelling. In greater detail, the research seeks to show that diagrammatic modelling eases problems of maintenance, extensibility, ease of use, and sharing of information. The methodology presented here to achieve this involves a three step translation from a visualised ontology, through a modelling tool, to output to interactive visualisations. An analysis of users groups them into categories of system creator, model builder, and model user. This categorisation corresponds well with the three-step translation process where users develop the ontology, modelling tool, and visualisations for their problem. This research establishes and exemplifies a novel paradigm of collaborative end-user programming by domain experts. The end-user programmers can use a visual interface where the visualisation of the software exactly matches the structure of the software itself, making translation between user and computer, and vice versa, much more direct and practical. The visualisation is based on an ontology that provides a representation of the software as a tree. The solution is based on translation from a source tree to a result tree, and visualisation of both. The result tree shows a structured representation of the model with a full visualisation of all parts that leads to the computed result. In conclusion, it is claimed that this direct representation of the structure enables an understanding of the program as an ontology and model that is then visualised, resulting in a more transparent shared understanding by all users. It is further argued that our diagrammatic modelling paradigm consequently eases problems of maintenance, extensibility, ease of use, and sharing of information. This method is applicable to any problem that lends itself to representation as a tree. This is considered a limitation of the method to be addressed in a future project.

No description available.